OPTIMISM AND PESSIMISM FROM

A HISTORICAL PERSPECTIVE
BRIAN DOMINO AND DANIEL W. CONWAY

The psychological accounts of optimism and pessimism delineated in

this volume represent the most recent attempts by humankind to determine
its place in the world. The roots of these contemporary accounts are found,
we believe, in the various treatments of optimism and pessimism that have
been advanced by leading philosophers of the modern period. In this chapter,
we trace the historical development of philosophical understanding of optimism and pessimism from their emergence in the writings of Descartes;
through their articulation by protopsychologists such as Schopenhauer and
Nietzsche; and finally to their evaluation by two eminent psychologistphilosophers, Sigmund Freud and William James.
In philosophy, optimism and pessimism designate competing, antithetical
positions. Philosophers are regarded as optimistic if they treat the cosmos
as generally hospitable to the aims and aspirations of human beings; they
are regarded as pessimistic if they treat the cosmos as generally indifferent or even hostile to the flourishing of human beings and civilizations.
As philosophical positions, both optimism and pessimism presuppose the

13

capacity of (some) human beings to arrive at an accurate forecast of the

future. This capacity to forecast the future usually depends, in turn, on a
reliable interpretation of the salient events of the past and present, such
that the forecasted future might be presented as following naturally (or
logically) from the general trends that characterize the past and present.
Philosophical statements of either optimism or pessimism typically assume
one of two forms: (a) Through the use of a priori reasoning (i.e., speculative),
philosophers attempt to defend a particular forecast of the future by appealing
to various general principles that are held as unchallenged truths, for example
the existence of God, or (b) through the use of a posteriori reasoning (i.e.,
empirical), philosophers attempt to defend a particular forecast of the future
by appealing to verifiable patterns found in nature. Although statements of
the first form are perhaps more prevalent in the history of Western philosophy, statements of the second form are generally held to be more persuasive.

17TH AND 18TH CENTURIES

Descartes
The historical emergence of optimism and pessimism as identifiable
philosophical positions is usually associated with the inception of the modern
period of philosophy in the 17th century. In this period, it became increasingly common for philosophers to maintain that the successful application
of reason to the cosmos warranted a philosophical outlook characterized by
either optimism or pessimism. The formulation of an optimistic philosophical
position can be traced to the writings of the French philosopher Ren6
Descartes (1596-1650). Widely recognized as the father of modern philosophy, Descartes contributed significantly to the transition from the Catholic
Church-influenced philosophy of the Middle Ages to the more secularly
oriented philosophy of the modern period. An accomplished mathematician
and scientist, Descartes keenly understood the need for philosophers to
escape the shadows of dogma and superstition. Where the Church had
largely failed to improve the material conditions of human life, he believed,
science would succeed in delivering a secularized world devoid of fear,
scarcity, and disease. The source of Descartess optimism can therefore be
traced to his conviction that a methodical application of human reason can
unlock the mysteries of the natural world (Descartes, 1628/1985). This
conviction informs all of his writings, from the early Rules for the Direction
of the Mind (Descartes, 1701/1985) to his final treatise, The Passions of the
Soul (Descartes, 1649/1985), wherein he asserted, There is no soul so weak
that it cannot, if well-directed, acquire an absolute power over its passions

14

DOMINO AND CONWAY

(050, p. 348). From an enhanced understanding of the natural world arises

the power to blunt the ill effects of nature on the welfare and progress of
humankind. In general, Descartes concluded, the continually improving
state of the world warrants the optimism that pervades his philosophy.
Descartes also contributed significantly to the emergence of a distinctly
moral sense of optimism and pessimism by his insistence that human beings
are fully capable of improving the state of the world by their own efforts.
In his Discourse on the Method, Descartes ( 1628/1985)famously asserted
that science will enable human beings to become the masters and possessors
of nature; they will thereby enjoy the fruits of the earth and maintain their
health. Although this claim may strike us as innocuous, it was considered
daring, and perhaps even blasphemous, by many 17th-century philosophers
and scientists. According to Descartes, human beings should be seen as
participating creatively in the ongoing improvement of the material conditions of human life; they need not resign themselves to destinies determined
solely by divine will and inscrutable fate. In particular, Descartes expressed
hope that medicine would eventually prevent illness and the enfeeblement
of old age. Because ill health caused a variety of social problems, he hoped
that the scientifically fortified practice of medicine might deliver us to an
Eden of our own making.
Pope

A similar optimism is expressed by Alexander Pope (1688-1744) in

his highly influential Essay on Man ( 1733-1 734). Unlike Descartes, however,
Pope derived his optimism from an a priori insight into the goodness and
interconnectedness of the cosmos. He thus challenged the meaningfulness
of the claim that humankind could be better off than it is. Any such claim,
he maintained, expresses the misguided wish that human beings become
embodied gods. Indeed, for Pope, to insist that the human intellect could
(and should) be more formidable is to forget that we are not divine. Pope
appeals persuasively to the integrated imagery of the Great Chain of Being,
wherein each creation is appointed (and must maintain) its own unique
place. To complain, for example, that our senses could be more acute is to
deny our enduring nature. In many cases, he maintained, an increase in
sensitivity would only bring about our own destruction. For example, if our
sense of smell were to become substantially more acute, then we might die
of a rose in aromatic pain (Pope, 1733-1734, VI). In fact, he concluded,
our senses are as perfectly acute as is consistent with our obligation to
maintain the unique station to which we have been appointed. Humankind
is located precisely at its optimal position in the Great Chain of Being; to
change our status even in the slightest would trigger a disruption throughout

HISTORICAL PERSPECTIVE

15

the entire plenum of nature. Or, as Pope famously concluded, One truth
is clear, Whatever is, is right (1733-1734, I, p. 10).

Leibniz
As a technical philosophical term, optimism has its roots in the writings
of Gottfried Leibniz (1646-1 7 16). In particular, his Theodicy (Leibniz, 17lo/
1996) offers a more rigorous proof of the optimistic conclusions reached by
Pope. Leibniz used the term optimum to name the unique maximum or
minimum instance of an infinite class of possibilities. In particular, he argued
that God followed the principle of the optimum in the creation of the
world. Although he admitted that God alone can perform an infinite analysis,
Leibniz held that the human mind is capable of comprehending proofs
involving the infinite (1710/1996, $69). It is perhaps not surprising, then,
that Leibnizs demonstration that this is the best of all possible worlds bears
a distinct resemblance to a mathematical proof. According to this influential
demonstration, Gods infinite knowledge includes the ideas of all possible
universes. Gods moral perfection caused Him to choose to create the best
of these possible universes-namely, the one that exists, which Leibniz
famously described as the best of all possible worlds. That this is the best
of all possible worlds, however, does not entail that it is also perfect. As
Leibniz argued, all things in the cosmos are interconnected; to change one
part in the whole is, therefore, to change the whole. Although another
possible world (one like ours but lacking some particular evil) might appear
to be better than our world, this appearance arises from the mistaken belief
that the elimination of a particular evil is a discrete, self-contained event.
Our belief in a possible world better than ours thus proves not that such a
world exists, but that we are unable to complete an infinite analysis. Leibniz
argued, for example, that the elimination of one evil would entail the
absence of the good produced by that evil, which would then yield a world
worse than the present one. Once we are able to comprehend this analysis,
he believed, we will agree that this is, in fact, the best of all possible worlds.

Voltaire
Leibnizs Theodicy is ridiculed by Voltaire (1694-1778) in Candide, or
Optimism (1759/1959). The characters inCandide suffer all manner of torture,
tragedy, affliction, and reversal of fortune. Throughout these misadventures,
the unflinchingly optimistic Pangloss maintains that, indeed, all is for the
best. Under the Leibnizian tutelage of Pangloss, Candide dutifully parrots
his teachers optimism. As a result, he habitually fails to learn from his
impressive travels and diverse experiences. Toward the end of the book,

16

DOMINO AND CONWAY

Candide announces that he has finally learned his lesson: He and the others
should renounce the speculative philosophy practiced by Pangloss, as well
as the false optimism that he preaches. They should busy themselves instead
with the practical labor involved in cultivating a garden, which will enable
them to remain productive as they resist the blandishments of philosophical
speculation (Voltaire, 1758/1959).
On a superficial level, Voltaires specific charges against philosophical
optimism could easily be answered by Pope and Leibniz. Neither thinker
maintains that anyones (much less everyones) life is perfect. Voltaires
point, though, is not only to present a narrowly philosophic refutation
of optimism, but also to demonstrate that optimism sanctions a numbing
indifference to human suffering. In fact, he wishes to criticize both optimism
and pessimism insofar as they rest on general, speculative principles that
do not admit of empirical verification. For example, the purported interconnectedness of the cosmos, to which many philosophical optimists appeal
when constructing their theodicies, is not confirmed by an empirical investigation of the cosmos. According to Voltaire, if reason is properly (i.e.,
empirically) applied to the world, it warrants neither optimism nor pessimism
as a general philosophical outlook. He consequently portrayed the character
Martin, an inveterate pessimist, as no less ridiculous than the optimist Pangloss.
In his Philosophical Dictionary, Voltaire ( 1764/1962) again marshaled
his rhetorical powers against the view that all is good [tout est bien].
Dropping the satirical excesses of Cadi&, Voltaire observed that optimism
seems plausible only to young aristocrats, whose lives are indeed pleasurable.
Such optimism is easily refuted as a general philosophical principle. A critic
need merely stick his head out the window, he will see unhappy people
enough; let him catch a fever, he will be unhappy himself (Voltaire, 1764/
1962, p. 420f). Against Leibniz, Voltaire charged that the claim all is good
means nothing more than all is governed by immutable laws (p. 425).
That is, Leibniz failed to consider a sufficiently wide range of examples in
presenting his defense of philosophical optimism. Voltaire suggested that
he add the following illustration of the immutable laws of nature:
When a stone is formed in my bladder, it is by an admirable piece of
machinery: gravelly juices pass little by little into my blood, filter into
the kidneys, pass through the urethras, deposit themselves in my bladder,
assemble there by an excellent Newtonian attraction; the stone is
formed, grows larger; I suffer evils a thousand times worse than death,
by the most beautiful arrangement in the world; a surgeon. . .comes to
thrust a sharp and pointed iron into the perineum and seizes my stone
with his pincers. It breaks under his efforts by a necessary mechanism;
and, by this same mechanism, I die in horrible torments. All this is

HISTORICAL PERSPECTZVE

17

good, all this is the evident consequence of unchangeable physical

principles; I agree, and I know it as well as you do. (1764/1962, pp.
lZOf, 4250

Hume
Like Voltaire, Scottish philosopher David Hume (1711-1776) challenged the reasonableness of philosophical optimism. In his witty Dialogues
Concerning Natural Religion (1775/1980), Hume debunked the popular claim
that a robust appreciation of God can be derived from an experience of the
order and beauty of the natural world. Showing that any appeal to natural
religion must trade on some version of the argument from design, also known
as the argument from analogy), he proceeded to demonstrate just how little
we can actually infer about the designer of the cosmos. Speaking through
the skeptical character Philo, Hume insisted that no general truths about
the cosmos can be derived from the argument from design.
Restricting oneself to an empirical investigation of the beauty, order,
and complexity arrayed throughout the natural world, one can infer at best
that the universe was created by something bearing a remote resemblance
to human intelligence. Whether the designer of the universe possesses
any innate moral or aesthetic qualities, however, cannot be conclusively
determined, nor does an empirical survey of the natural world establish
whether this creator still exists. Rather than support a full-blown philosophical optimism, natural religion at best sanctions a modest commitment to
deism. If an optimistic position is to be maintained, Hume skeptically concluded, it cannot rest solely on the evidence provided by natural religion.
Relying solely on an analogy with human design, philosophical optimists
cannot claim to have evidence of the complexity and necessity that they
attribute (and Hume denied) to the cosmos as a whole. According to Hume,
in fact, an application of a posteriori reasoning to the cosmos leads us neither
to optimism or pessimism, but to skepticism. From this skeptical position,
we may choose to invest our faith in optimism or pessimism, but in no
event will we do so with sufficient reason (Hume, 1775/1980).
Kant
Like Voltaire, Immanuel Kant ( 1724-1804) dismissed the optimistic
claim that the existence of evil mysteriously contributes to Gods larger
plan. In his aptly entitled essay On the Miscarriage of all Philosophical
Attempts in Theodicy, Kant rejected this claim as need[ing] no refutation;
surely it can be freely given over to the detestation of every human being
who has the least feeling for morality (1791/1996, p. 38). Beginning with
his Critique of Pure Reason (Kant, 1781/1965), he thus maintained that both

18

DOMINO AND CONWAY

optimism and pessimism are metaphysically indefensible positions. Just as

human beings can never know that God exists, so is it impossible to prove
that God is impossible (1817/1966, pp. 335-452).
Kant nevertheless wished to claim that optimism holds great value as
a practical position, warranted by the postulates of pure practical reason
(Kant 1788/1956,p. 12). The following summarizes Kants reasoning: Kant
called the summum bonum the convergence of virtue and happiness. Philosophers have long sought to establish a basis for anticipating the summum
bonum. This anticipation has been notoriously difficult to foster, however,
especially in light of the widespread belief that immoral people are often
happier than moral people. It is not surprising, then, that many religiously
inclined philosophers have attempted to locate the convergence of virtue
and happiness in an afterlife or afterworld. Kant himself acknowledged
that if the summum bonum were unattainable, then the moral law which
commands that it be furthered must be fantastic, directed to empty imaginary
ends, and consequently inherently false (1788/1956, p. 118).
To avoid the contradiction involved in believing the moral law to be
false, Kant recommended that we postulate the existence of God and the
immortality of the soul as necessary conditions of the summum bonum, even
though we cannot objectively determine that its realization in fact depends
on them. A rational faith in these postulates of pure practical reason,
although objectively insufficient, thus facilitates the pursuit of happiness.
In fact, only by virtue of ones belief in these postulates can one avoid what
Kant called the absurdum practicum-namely, the life of the scoundrel who
chooses to disobey the moral law (Wood, 1970, p. 33). If we assume that
God exists and that this world supports the possible convergence of virtue
and happiness, then we will have sufficient motivation to act morally and
to maintain our desire to do so. Such assumptions are, Kant admitted,
matters of faith rather than of knowledge. But the faith invested in them
is nevertheless rational, because both the existence of God and the optimality
of this world are logically possible; that is, they do not contradict reason.
As a practical matter, then, it is better to believe that God exists and that
perfect virtue can coincide with earthly happiness than to believe otherwise.

19TH AND 20TH CENTURIES

Hegel
The most influential figure of 19th-century philosophy was Georg
Wilhelm Friedrich Hegel (1770-1831), who is equally famous for his optimistic account of history and his pessimistic understanding of the development of human consciousness through conflict and domination. T o describe
HISTORZCAL PERSPECTIVE

19

the teleological mechanism by which human consciousness has developed,

Hegel referred to the operation of the cunning of Reason ( 1837/1953,
p. 89). Working behind the scenes of human history, Reason has orchestrated the various collisions between forms of consciousness that have collectively sparked the spiritual progress of humankind. In Hegels view, then, the
teleological progress of human consciousness, as facilitated by the cunning of
Reason, is the cause for great optimism. Even the seeming disasters of world
history, wherein entire nations have been bent to the will of emperors and
dictators, are recuperated within this teleological frame. All that has befallen
humankind along the way has contributed to the freedom that characterizes
the modern epoch. Although Hegels optimism is more sophisticated than
Leibnizs, it is similarly vulnerable to Voltaires charge of insensitivity toward
human suffering. According to Hegels developmental account, historical
atrocities are merely the necessary, if unpleasant, consequences of the articulation of increasingly more adequate conceptions of freedom. Had countless
peoples not been sacrificed on the slaughter-bench of history, the modern
world would never have developed the mature notion of freedom that now
informs our laws and institutions (Hegel, 1830/1975,p. 69). One could, as
Voltaire warned, easily dismiss the myriad horrors of world history as the
inevitable growing pains of the human spirit.
Hegels world-historical optimism is balanced by his account of the
development of consciousness, which many critics see as inherently pessimistic. Hegel understood humans as social beings who acquire their sense of
identity through the recognition of others. He thus traced the origins of
civilization to the inescapable battle for power and recognition between
individuals. Stated briefly, each individual wants both to master his or her
environment and to command recognition of this mastery from others. By
risking ones life in the absence of any biological threat or justification, one
asserts ones independence from nature, but one also requires a witness of
this assertion, for ones identity is dependent on the recognition of others.
This desire for recognition by the Other, which the Other shares reciprocally,
leads to a struggle to the death, from which only one party can emerge as
master. For any individual, the optimum outcome of this primal struggle
involves the voluntary submission of the other, from which one receives
recognition and confirmation of ones mastery over nature. Thus emerge
the polar positions of lordship and bondage, which are mutually defined
in terms of the masters demand that the slave embody the masters independence from nature-that is, by working on the world and bestowing on the
master the fruits of this forced labor (Hegel, 1807/1977).
According to Hegel, this original arrangement of lordship and bondage
eventually reverses itself. The existence of the master is now dependent on
the labor of the slave. The erstwhile master has become merely a consumer,
whereas the former slave has become a creator. The master is now enslaved

20

DOMINO A N D CONWAY

to a slave, while the slave now masters the master (Hegel, 1807/1977).
Such tensions and reversals, developed more fully by Freud and other depth
psychologists, lie at the very origins of human civilization. Although Hegel
envisions an optimistic resolution to this drama whereby the slaves eventually inherit the earth, the pessimism inherent in his account of the development of human consciousness is difficult to ignore. If identity is dependent
on recognition, and if recognition is achieved only under conditions of
struggle and domination, then it is not immediately clear how human
civilization might someday outgrow, as Hegel optimistically promised, its
well-documented thirst for blood, sweat, and tears.

Schopenhauer
As gloomy as it may be, the pessimism suggested by Hegels account
of social development pales in comparison to that of the pessimist par
excellence, Arthur Schopenhauer (1788-1860). In his major work, The World
as Will and Representation (1818/1958), Schopenhauer argued not merely
that this is not the best of all possible worlds, but that it is demonstrably
the worst of all possible worlds. Schopenhauer began this famous argument
by defining a possible world as one that could endure (as opposed to one
that could exist only for a brief moment). This definition thus precludes
any appeal to ostensibly worse worlds that simply could not endure (e.g., a
world like ours, but in which all means of producing oxygen were suddenly
to vanish; Schopenhauer, 1818/1958). If our world is not the worst of all
possible worlds, then it must be possible to describe a worse world. Citing
example after example, Schopenhauer argued that our world teeters on the
brink of destruction. Any changes we might imagine that would demonstrate
the possibility of a worse world would either cause our world to collapse,
thus making it an impossible world, or create a better possible world. Schopenhauer thus concluded, Consequently, the world is as bad as it can
possibly be, if it is to exist at all (1818/1958, Vol. 2, p. 584).
One might attempt to counter Schopenhauers pessimism by noting
that some people in this world are happy, even if they are exceptional in
this respect. A worse world would therefore be one in which fewer people
were happy. According to Schopenhauer, however, it would be a mistake
to assume that even some people are happy. By his account, human beings
cannot be happy, and aspiring optimists are merely ignorant that this supposedly worse world is indeed our own. According to Schopenhauer, human
beings are motivated primarily by a blind, unconscious force that he called
the will, which is nothing but an incessant striving for a goal that cannot
be adequately identified. Indeed, the wills object is never chosen, because
the will is even more basic than reason itself. The absolute freedom of the
will, or so Schopenhauer claimed, is therefore illusory. Instead, he explained,
HISTORICAL PERSPECTIVE

21

the will blindly desires but is never satisfied. If happiness is understood in

terms of the satiation of desire, as Schopenhauer maintained, then in this
world no one is happy. What some people mistakenly take to be happiness
is simply the transient illusion of satisfaction, which Schopenhauer attributed
to a temporary reduction of pain. The only way to mitigate the wills constant
strivings is by means of an ascetic abrogation of the will within oneself.

Nietzsche
The influence of Schopenhauers pessimism is most evident in the
writings of Friedrich Nietzsche ( 1844-1900). Both philosophers share the
view that human beings blindly follow their unconscious impulses as they
attempt to derive immanent meaning from lives devoid of transcendent
meaning. Both also agree that typical human beings cannot resolutely embrace this truth and must instead invest their faith in the saving fictions of
religion and philosophy. From this common misanthropic ground, however,
the two philosophers part company. Whereas Schopenhauer recommended
his brand of pessimism as constituting the only honest response to the blind
strivings of the will, Nietzsche apparently promoted a joyful, affirmative
response to the absurdity of human existence. This difference of philosophical opinion stems from their widely divergent accounts of the psychological
disposition of those individuals who can bear the truth of the meaninglessness
of life. In the symptomatological terms that Nietzsche preferred in his later
writings, in fact, Schopenhauers vaunted pessimism signifies the decline of
life, whereas the spontaneous affirmation expressed by Nietzsches Uberc
mensch bespeaks the ascent and plenitude of life.
In his first book, The Birth of Tragedy (1872/1967), Nietzsche presented
his most sustained discussion of the psychophysiological states represented,
respectively, by optimism and pessimism. Influenced by Schopenhauers
claim that the world presents itself to us as both will and representation,
Nietzsche attempted to account for the birth of Attic tragedy in the
synthesis of the Apollinian and the Dionysian impulses (Nietzsche, 1872/
1967). The Apollinian impulse signifies those accomplishments for which
the Golden Age of Greece is justifiably famous: measure and harmony in
aesthetics, the clarifying light of reason, and a celebration of the dream
world of myth and appearance. Yet, Nietzsche conjectured, the Apollinian
impulse arises only in response to a more powerful, countervailing impulse
that Attic tragedy also expresses; namely, the Dionysian impulse toward
dissolution and disintegration. The Dionysian impulse engenders the ecstatic
bliss that occurs when human beings surrender their illusory individuality
and are once again subsumed within the eternal stream of life (Nietzsche,
1872/1967). The Dionysian impulse thus represents both the terror and joy

22

DOMINO AND CONWAY

that one feels at the destruction of ones self; it is the painful truth that
lies beneath the Apollinian illusions.
By virtue of its dual allegiance to Apollo and Dionysus, Greek tragedy
thus reflects a pessimism that Nietzsche initially interprets as indicative of
great health and strength of will. Tragedy is pessimistic, he explained, insofar
as it squarely confronts the inevitable demise of even the greatest human
exemplars and accomplishments. Rather than promote a nihilistic resignation, however, tragedy encourages human beings to strive for greatness and
excellence in their transient individual existence, enticing them with the
grand illusions and fine appearances crafted under the aegis of Apollo.
The pessimism of strength expressed by Attic tragedy thus manifests an
affirmation of life itself, including the dissolution and demise associated
with the Dionysian impulse.
Nietzsches articulation of this diagnosis evidences the extent of his
departure from the so-called existential claims that support Schopenhauers
pessimism. In the process of conducting his genealogy of morality,
Nietzsche (1887/1989b) observed that for much of Western history, the
twin forces of Socratic hyperrationality and Christian asceticism have all
but destroyed the type of humans who could give life meaning. He also
believed, however, that these destructive forces are gradually withering and
that new, creative superhumans may yet emerge to redeem the meaninglessness of human existence. The question that dominates his later writings is
whether humankind has become simply too enervated to produce any creative geniuses of the magnitude necessary to warrant its future. In Thus
Spoke Zarathustra (1883-1885/1983), he called this fatal condition the reign
of the last man, and he entrusted his character Zarathustra with the task
of rousing humankind from its moribund condition. It is not clear whether
Nietzsche intended Thus Spoke Zarathustra to serve notice of his own
(guarded) optimism concerning the future humankind, or if he intended to
engender in others an optimism that he himself could not muster. In any
event, Zarathustru is widely regarded as a stirring attempt to find meaning
even in the aftermath of the death of God.
In the writings from his final year of sanity, 1888, Nietzsche largely
abandoned the bombast of his Zarathustra and defended an optimism predicated on the teaching of umor fati (love of fate). The highest human beings
are those who are able to affirm the cosmos as a whole, understood as a
grand, interconnected network of fatalities. Against this fatalistic backdrop,
any wish to change even the smallest element of ones life is understood
in Nietzsches scheme to harbor the nihilistic wish that the whole of ones
life-and the cosmic totality in which it inheres-be different. Having a
genuinely optimistic affirmation of life would mean that one wants nothing
to be different, not forward, not backward, not in all eternity. Not merely

HISTORICAL PERSPECTIVE

23

bear what is necessary, still less conceal it-all idealism is mendaciousness

in the face of what is necessary-but love it (Nietzsche, 1908/1989a, p. 258).

Freud
Like Nietzsche, Sigmund Freud (1856-1939) is neither an optimist h
la Leibniz, nor a pessimist h la Schopenhauer. He too believed that civilization could be improved if the proper sort of individuals were to emerge, but
his hopes for this emergence were even more tepid than Nietzsches (Freud,
1927/1961a, p. 8). Although best known for his pioneering work in depth
psychology and psychoanalysis, Freud endeavored late in his career to develop the sociological and anthropological implications of his basic principles
of analysis. It is with respect to this latter enterprise that we may best
appreciate his contributions to the philosophical treatment of optimism
and pessimism.
Freuds considered opinion on the future of humankind can be gleaned
from his response to Albert Einsteins letter concerning the necessity of war
(1933/1961b). On the one hand, Freud emphatically dismissed any attempt
to annul humankinds aggressive inclinations. O n the other, he noted that
whatever furthers the growth of civilization simultaneously frustrates the
outbreak of war. A t bottom, then, civilization is riddled with tensions for
which there is little hope of resolution. This tension can also be located
in Freuds depth-psychological model of the soul, which figures the (civilized)
soul as the site of an unavoidable, internecine battle that ends only in
death. Indeed, if happiness is defined in the ideal terms that Freud proposed,
comprising the spontaneous outward expression of ones unconscious drives
and impulses, then no civilized human being can be deemed happy. It is in
this sense that Freud presented himself as a pessimist. If, however, happiness is
defined in the more restricted terms favored by civilization itself, involving
a delayed (but not pathological) gratification of ones unconscious drives and
impulses, then it is possible for some human beings to secure for themselves a
modicum of happiness (albeit only with the help of psychoanalysis). It is
in this sense that Freud presented himself as a guarded optimist.
Continuing a long philosophical tradition, Freud asserted that human
beings naturally strive for happiness. This pursuit takes two forms: Humans
wish both to avoid suffering and to experience extreme feelings of pleasure
(Freud, 1930/1961c). Two of the three sources of suffering are the superior
power of nature and the frailty of our own bodies. Although we cannot
eradicate either of these sources of suffering, we can minimize their impact
by banding together for mutual support and protection-by doing so the
first families and communities were formed. This banding together granted
the additional benefit of providing an ostensibly secure means of regularly

24

DOMINO AND CONWAY

obtaining sexual satisfaction (at least for the male), which is the highest
form of pleasure. By making one person the source of his happiness, however,
the individual actually places his happiness at risk, for his lover may withdraw, abdicate, or die. Fearful of this inevitable loss, the individual then
instinctually sublimates erotic urges and engages in aim-inhibited love or
nonsexual relationships with others (Freud, 1930/1961c, pp. 102ff). T o
carry on its business, civilization encourages the individual to have such
relationships, thereby directly confronting Eros and deflecting its advance.
Civilization further encourages such relationships because they assist in its
project to control the individuals aggressiveness. Thus individuals are drawn
toward civilization for the security it offers in the face of a potentially hostile
natural world, but they simultaneously rail against its enforced restrictions
of their sexual and aggressive instincts. The best the individual can hope
for is the formation of relatively small groups of people who visit their
aggressions on people outside the group. In other words, there is no hope
of eliminating the pains of life, only of redirecting them.
Of course, civilization would quickly expend its resources if it were
obliged to enforce its prohibitions in every instance. Pursuing a line of
interpretation developed by Nietzsche, Freud described the psychological
mechanism by which civilization deputizes each individual as his or her
own enforcer: The individual internalizes the founding taboos of civilization
and consequently follows its dictates under the threat of self-punishment,
or guilt. Guilt arises as the super ego deflects onto the ego the aggressive
energies intended by the individual for outward expression (Freud, 1930/
196lc).
Yet no matter how successful individuals become at policing their
sexual and aggressive impulses, civilization ultimately fails in its mission to
protect them from nature and from each other. Thus the individual remains
a child in that he or she continues to require protection from superior
powers. It is not surprising, then, that humans lend these powers fatherly
attributes. Indeed, the comforting illusions created by religion have significantly furthered the advance of civilization, much as a devoted father contributes to his childrens progress toward adulthood. Just as a paternalistic
father can imprison his children in an arrested state of adolescence, however,
so does religion now harm both individuals and civilization. As Freud saw
it, religion has largely outlived its usefulness as a means of accommodating
individuals to the demands (and benefits) of civilization. Particularly in
The Future of an Illusion, Freud argued that the psychic energies of
humanity are best spent forwarding the project of science. The new god
Freud wished to install is Logos. Although Logos cannot make promises on
the order of those made by religious gods, it actually fulfills many of the
promises that it does make. Although science lacks the emotional

HISTORICAL PERSPECTIVE

25

components of religion, Freud believed that civilization will eventually shed

its infantile need for such comforts. Although our instincts often overpower
our reason, this victory is only temporary:
The voice of the intellect is a soft one, but it does not rest till it has
gained a hearing. Finally, after a countless succession of rebuffs, it
succeeds. This is one of the few points on which one may be optimistic
about the future of mankind, but it is in itself a point of no small
importance. And from it one can derive yet other hopes.. . . [The
intellect] will presumably set itself the same aims as those whose realization you expect from your God . . . namely the love of man and the
decrease of suffering. (Freud, 1927/1961a, p. 53)

Written only a few years after On the Future of an Illusion, Civilization

and Its Discontents closed Freuds distinguished career on a pessimistic note.
Although he officially refused in this book to comment on the future of
humankind, his gathering pessimism is nevertheless manifest. As the unprecedented slaughter of World War I made painfully clear, science has empowered humankind to annihilate itself. This grim realization apparently shatters
the hopeful picture of science that informs Freuds earlier work. Like religion,
science merely furnishes saving illusions, which only temporarily distract
humankind from the discontents legislated by civilization. When the scientific illusion fades, human beings find themselves in full possession not only
of their discontents, but also of weapons of mass destruction. Freud thus
hypothesized that we have nearly reached a point of world-historical crisis:
The psychic demands placed on us by civilization may soon exceed both
the benefits that we receive and the native pains of a pre- or postcivilized
existence. Whatever net value humankind has heretofore derived from
civilization is now virtually exhausted.
Alarmed by the success of Hitlers Nazi party in the 1930parliamentary
elections in Germany, Freud added a new, final sentence to Civilization and
Its Discontents. This sentence effectively deflates the optimism with which
he had originally intended to close the book. Eros may in fact assert himself
in the struggle with his equally immortal adversary, as Freud had originally
concluded. But, he now skeptically adds, who can foresee with what
success and with what result? (1930/1961c, p. 112). Throughout the 1930s,
Freuds skepticism gradually yielded to a full-blown pessimism. Forced to
emigrate from his home in Vienna by the recrudescence of nationalism and
anti-Semitism and slowly dying from a painful throat cancer, Freud now
foresaw the final victory of Thanatos over Eros. At the time of his death
in 1939, in fact, Freud was well aware that the animosities that would
precipitate World War I1 were inexorably gathering force. Having apparently
learned nothing from the atrocities of World War I, humankind was preparing itself for yet another bloodletting of global dimensions. Could the demise
26

DOMINO AND CONWAY

of civilization itself be far off?These deathbed sentiments thus edged Freud

closer to Schopenhauer than to Nietzsche, for unlike Nietzsche, Freud
reserved little hope for the future of humankind.

James

A similarly pessimistic conclusion is reached by Freuds American

contemporary, William James (1842-1910). A fellow pioneer in the new
science of psychology, James shared several of Freuds greater doubts about
the prospects for human happiness within the walls of civilization. James also
shared Freuds concern about humankinds increasingly uncritical reliance on
science to deliver the material conditions of happiness. James lent a far
more personal-and therefore more ominous-voice to the possibility of
the utter disruption of the self:
Whilst in this state of philosophic pessimism and general depression of
spirits about my prospects, I went one evening into a dressing-room in
the twilight to procure some article. . .; when suddenly there fell upon
me without any warning, just as if it came out of the darkness, a horrible
fear of my own existence. Simultaneously there arose in my mind the
image of an epileptic patient whom 1 had seen in the asylum, a blackhaired youth with greenish skin, entirely idiotic, who used to sit all day
on one of the benches, or rather shelves against the wall, with his knees
drawn up against his chin, and the coarse gray undershirt, which was
his only garment, drawn over them inclosing his entire figure. He sat
there . . . moving nothing but his black eyes and looking absolutely
non-human. This image and my fear entered into a species of combination with each other. That shape am I, I felt, potentially. Nothing that
I possess can defend me against that fate, if the hour for it should strike
for me as it struck for him. There was such a horror of him, and such
a perception of my own merely momentary discrepancy from him, that
it was as if something hitherto solid within my breast gave way entirely,
and I became a mass of quivering fear. (James, 1902, pp. 160ff)

As Jamessanecdote demonstrates, medical science exerts a democratizing influence on humankind. The difference between sane and insane, for
example, is no longer understood as a difference in kind, but rather as a
relatively small organic difference. Although science creates the possibility
of a cure, and thus gives us reason to be optimistic, it simultaneously elides
the perceived difference between oneself and the diseased other.
It is important to note that James radically shifted the topos within
which discussions of optimism and pessimism can occur. The combined forces
Although James presented this example as drawn from an anonymous source, his son claimed that
the narrator is James himself (see Allen, 1967, pp. 164ff.).

HISTORICAL PERSPECTIVE

27

of Kants critique and scientific materialism render metaphysical questions

otiose. The rise of empirical science might lead one to expect that the
debate between Pope and Voltaire could be definitively answered, at least
in theory. Yet as both James and Nietzsche argued, albeit in very different
ways, science can never capture (or banish) the inescapably subjective
element of human experience. Thus for James, the philosophical question
of optimism versus pessimism can be decided not at any metaphysical or
anthropological level, but only at the level of the individual.
If the choice between optimism and pessimism ultimately lies with
each individual, then it is difficult at first to see why anything like a profound
decision must be made. All things being equal, it seems far preferable to
go through life with a Leibnizian or Popean attitude than a Schopenhauerian
one. Indeed, one might argue that Jamess designations for these competing
attitudes, healthymindedness and the sick soul, merely reiterate our
customary preference for happiness over unhappiness (James, 1902, Lectures
IV). Nevertheless, James does not lapse into the bourgeois optimism of
which Voltaire accused Leibniz and Pope. Optimism resembles a veil that
shields us from the inequities of life. Like a veil, it may be blown away by
a sudden gust of wind, or pushed aside by our own curiosity. One might be
able to lead an optimistic life, James conceded, but this is unlikely given
the insecurity of natural goods and the virtual certainty that one will fail
at something meaningful to ones sense of flourishing.

REFERENCES
Allen, G. W. (1967). William James: A biography. New York: Viking.
Descartes, R. (1985a). Discourse on the method. In J. Cottingham, R. Stoothoff,
& D. Murdoch (Trans.), The philosophical writings of Descartes (Vol. 1, pp.
111-175). New York: Cambridge University Press. (Original work published
1628)
Descartes, R. (1985b). Rules for the direction of the mind. In J. Cottingham, R.
Stoothoff, & D. Murdoch (Trans.), 7% philosophical writings of Descartes (Vol.
1, pp. 9-78). New York: Cambridge University Press. (Original work written
1628, published 1701)
Descartes, R. (1985~).The passions of the soul. In J. Cottingham, R. Stoothoff, &
D. Murdoch (Trans.), The philosophical writings of Descartes (Vol. 1, pp. 326404). New York: Cambridge University Press. (Original work published 1649)
Freud, S. (1961a). Civilization and its discontents. (J. Strachey, Trans.). New York:
W. W. Norton. (Original work published 1930)
Freud, S. (1961b). Letter to Albert Einstein. In J. Strachey (Ed. and Trans.), The
standard edition of the complete psychological works of Sigmund F r e d (Vol. 22,
pp. 203-215). London: Hogarth Press. (Original work written 1933)

28

DOMINO A N D CONWAY

Freud, S. (1961~).The future of an illusion. In J. Strachey (Ed. and Trans.), The

standard edition of the compkte psychological works of Sigmund Freud (Vol. 21).
London: Hogarth Press. (Original work published 1927)
Hegel, G. W. F. (1953). Reason in history. (R. S. Hartman, Trans.). Indianapolis,
IN: Bobbs-Merrill. (Original work published 1837)
Hegel, G. W. F. (1975). Natural law. Philadelphia: University of Pennsylvania
Press. (Originally published 1830)
Hegel, G. W. F. (1977). Phenomenology of spirit. (A. V. Miller, Trans.). New York:
Oxford University Press. (Original work published 1807)
Hume, D. (1980). Dialogues concerningnaturalreligion, (R. H. Popkin, Ed.). Indianapolis, IN: Hackett Press. (Original work published 1775)
James, W. ( 1902). The varieties of religious experience. New York: Longmans, Green
& co.
Kant, I. (1956). Critique of practical reuson. (L. W. Beck, Trans.). Indianapolis, IN:
Bobbs-Merrill. (Original work published 1788)
Kant, I. (1965). Critique of pure reason. (N. K. Smith, Trans.). New York: St.
Martins Press. (Original work published 1781)
Kant, I. (1969). Lectures on the philosophical doctrine of religion. In A. Wood
(Trans.) & G. DiGiovanni (Ed.), Religion and rational theology (pp. 335-452).
New York: Cambridge University Press. (Original work published 1817)
Kant, I. (1996). On the miscarriage of all philosophical attempts in theodicy. In
A. Wood & G. DiGiovanni (Trans. and Eds.), Religion and rational theology
(pp. 19-38). New York: Cambridge University Press. (Original work published 1791)
Leibniz, G. (1996). Theodicy. (E. M. Huggard, Trans.). La Salle, IL: Open Court.
(Original work published 1710)
Nietzsche, F. (1967). The birth of tragedy. (W. Kaufmann, Trans.). New York: Vintage. (Original work published 1872)
Nietzsche, F. (1983). Thus spoke Zarathustra. In W. Kaufmann (Trans.), The portable
Nietzxche (pp. 115-439). New York: Penguin. (Original work published 18831885)
Nietzsche, F. (1989a). Ecce homo. In W. Kaufmann (Trans.), O n the genealogy of
morals and ecce homo (pp. 215-335). New York: Vintage. (Original work
published 1908)
Nietzsche, F. (1989b). On the genealogy of morals. In W. Kaufmann (Trans.), O n
the genealogy ofmorals and ecce homo (pp. 31-163). New York: Vintage. (Original work published 1887)
Pope, A. (1994). An essay on m n . New York: Dover. (Original work published 17331734)
Schopenhauer, A. (1958). The world as will and representation (Vols. 1-2; E. F. J.
Payne, Trans.). New York: Dover Publications. (Original work published 1818)
HISTORICAL PERSPECTIVE

29

Voltaire, F. (1959). Cad&, or optimism. (L. Bair, Trans.). New York: Bantam
Books. (Original work published 1759)
Voltaire, F. (1962). Philosophical dictionary. (Vols. 1-2, P. Gay, Trans.). New York:
Basic Books. (Original work published 1764)
Wood, A. (1970). Kants moral religion. Ithaca, NY: Cornell University Press.

30

DOMINO AND CONWAY

2
OPTIMISM, PESSIMISM,
AND SELFeREGULATION
CHARLES S. CARVER AND MICHAEL E SCHEIER

Optimists are people who expect good experiences in the future. Pessimists are people who expect bad experiences. These concepts have a long
history in folk wisdom, as well as in early attempts to categorize people
according to their qualities of personality (see chap. 1, this volume). It
has long been believed that this fundamental difference among people is
important in many, perhaps all, facets of life. In the past 15 years or so,
this belief has been supported by a good deal of systematic research on the
correlates and consequences of this individual difference variable. Later
chapters of this book describe some of that research.
In this chapter we provide some conceptual background to this topic.
More specifically,we describe the theoretical principles that underlie our own
use of the concepts of optimism and pessimism. The colloquial definitions of
these terms (and those found in dictionaries) turn on peoples expectations
Preparation of this chapter was facilitated by grants CA64710 and CA64711 from the National
Cancer Institute.

31

for the future-positive and negative, respectively. Scientific approaches to

these ideas (including ours) also turn on expectations for the future. In so
doing, these scientific approaches link the concepts of optimism and pessimism to a long tradition of expectancy-value models of motivation. This
link means that the optimism construct, although rooted in folk wisdom,
is also firmly grounded in decades of theory and research on motives and how
they are expressed in behavior. Our approach to understanding behavior, an
approach that is identified with the term self-regulation, is one variant within
this broad tradition.
We begin by briefly exploring the elements of the expectancy-value
approach to motivation, to better understand the dynamics that underlie
optimism and pessimism and how they influence human experience. We
then turn to a description of the specific self-regulatory model within which
we address these qualities of personality in our own thinking and writing.

EXPECTANCY-VALUE MODELS OF MOTIVATION

The expectancy-value viewpoint on motivation begins with the assumption that behavior is organized around the pursuit of goals. Goals
provide the vulue element of the expectancy-value approach. Goals are
qualities that people take as desirable or as undesirable (you might think
of the latter as anti-goals). People try to fit their behaviors (indeed fit
their very selves) to values they see as desirable, and people try to keep
away from values they see as undesirable. The more important these goals
and anti-goals are to the person, the greater is the element of value in the
persons motivation with respect to those goals. Without having a goal that
is valued to at least some degree, people have no reason to act.
The second conceptual element in this approach is expectancy-a sense
of confidence or doubt about the goals attainability (or the anti-goals
avoidability). If the person lacks confidence, again, there will be no action.
Thats why a lack of confidence is sometimes referred to with the phrase
crippling doubt. Doubt can impair effort both before the action begins
and also while its ongoing. Only if people have sufficient confidence will
they move into action, and only if they retain sufficient confidence will
they remain engaged in effort. When people have confidence about an
eventual outcome, their efforts will continue, even in the face of enormous adversity.

SELF-REGULATION OF BEHAVIOR
Our particular version of the expectancy-value approach to motivation
also uses another idea: that behavior embodies feedback control processes.

32

CARVER AND SCHEIER

The basic unit of analysis in this view is the discrepancy-reducing feedback

loop. This is a system of four elements in a particular organization (cf.
MacKay, 1966; Miller, Galanter, & Pribram, 1960). The elements are an
input function, a reference value, a comparator, and an output function
(see Figure 2.1). An input function brings information in. This function is
equivalent to perception. The reference value is a second source of information
(i.e., apart from whats in the input function). The reference values in the
loops were interested in are goals. The comparator is a device that compares
input and reference value. The comparison yields one of two outcomes:
either the values are discriminably different from one another or theyre
not. Comparators can vary in sensitivity, however, so some can detect very
small discrepancies, whereas others can detect only much larger ones.
Following this comparison is an output function. We treat this as equivalent to behavior, although sometimes the behavior is internal. If the comparison yields no difference, the output function remains whatever it now is.
This may mean no output, or it may mean that the ongoing output continues.
If the comparison yields discrepancy, however, the output function
changes.
There are two kinds of feedback loops, corresponding to regulation of
present states with respect to goals and to anti-goals. In a discrepancyreducing loop, the output function is aimed at diminishing or eliminating
discrepancies between input and reference value. This is seen in attempts

Standard,
Reference value

V
Input function

Output function

Effect on
environment
Disturbance
Figure 2.1. Schematic depiction of a feedback loop. In such a loop, a sensed value
is compared to a reference value, or standard, and adjustments are made in an output
function (if necessary) to shift the sensed value in the direction of the standard.

OPTIMISM, PESSIMlSM, AND SELF-REGULATION

33

to approach or attain a valued goal. The goal can be a relatively constant

one (e.g., being an honest person), or it can be more of a moving target
(e.g., developing through the stages of a career). In either case, a discrepancyreducing system tries to make reality match the goal value.
This isnt behavior for the sake of behavior, but behavior to create
and maintain conformity between input and standard. That is, the value
sensed by the input function depends on more than the output (see Figure
2.1). Disturbances from outside can change conditions adversely (increasing
a discrepancy from the reference value) or favorably (diminishing a discrepancy). In the first case, recognition of a discrepancy prompts a change in
output, as always. In the second case, though, the disturbance preempts the
need for an adjustment in output, because the system sees no discrepancy.
The second kind of feedback loop is a discrepancy-enlarging loop.
The value here isnt one to approach, but one to avoid. Its an anti-goal.
Psychological examples of anti-goals are receiving traffic tickets, being subjected to public ridicule, and being fired from your job. A discrepancyenlarging loop senses present conditions, compares them to the anti-goal,
and tries to enlarge the discrepancy between the two. T o use another
example, a rebellious adolescent who wants to be different from his or her
parents (thus treating the parents values as an anti-goal) senses his or her
own behavior, compares it to the parents behavior, and tries to make his
or her behavior as different from theirs as possible.
The action of discrepancy-enlarging processes in living systems is typically constrained in some way by discrepancy-reducing loops (see Carver
& Scheier, 1998).T o put it differently, avoidance behaviors often lead into
approach behaviors. An avoidance loop creates pressure to increase distance
from the anti-goal. The movement away occurs until the tendency to move
away falls within the sphere of influence of an approach loop. This loop
then serves to pull the sensed input into its orbit. Rebellious adolescents,
trying to be different from their parents, soon find a group of other adolescents
to conform to, all of whom are remaining different from their parents.
Feedback Loops and Goals
The view we take on behavior may be viewed by some as somewhat
esoteric in its reliance on the feedback loop as the essential building block
of behavior. In its reliance on the goal concept, however, this view is very
much in accord with contemporary personality-social psychology. That is,
the reference value functions very much as a goal (or anti-goal) for the
operation of that feedback process, and todays personality-social psychological theory and research incorporates a great many discussions of goal constructs (e.g., Austin & Vancouver, 1996; Carver & Scheier, 1998; Elliott
& h e c k , 1988; Pervin, 1982, 1989). Our position is that when these

34

CARVER AND SCHEIER

various goals are being used, they are serving as reference points for feedback
processes involved in the creation of the resulting behavior.
A variety of labels are used in the literature of goals, reflecting differences in emphasis among various writers. Theorists who use different terms
have their own emphases (for discussions, see Austin & Vancouver, 1996;
Carver & Scheier, 1998), but many of the points they make are the same.
All the theories include the idea that goals energize and direct activities
(Pervin, 1982). They implicitly (and sometimes explicitly) convey the sense
that goals give meaning to peoples lives. Each theory places an emphasis
on the idea that to understand a person means understanding the persons
goals. Indeed, in the view represented by these theories, its often implicit
that the self consists partly of the persons goals and the organization
among them.

Hierarchical Organization

An issue to be addressed in this context stems from the obvious fact

that some goals are broader in scope than others. How to think about the
difference in breadth isnt always easy to put your finger on. Sometimes its
a difference in how much time is involved in the pursuit of the goal.
Sometimes, though, its more than that. Its a difference in the goals level
of abstraction.
The notion that goals differ in their levels of abstraction is easy to
illustrate. For example, you might have the goal, at a relatively high level
of abstraction, of being a self-sufficient person. You may also have the goal,
at a lower level of abstraction, of preparing dinner for yourself, or of replacing
the spark plugs in your car. The first goal is to be a particular kind of person,
the second set concerns completing particular kinds of action. You could also
think of goals that are even more concrete than the latter ones, such as
the goal of slicing vegetables into a pan, or the goal of turning a wrench
to a particular degree of tightness. These goals (which some theorists would
call strategies) are closer to specifications of individual acts than were the
second set, which were more summary statements about the desired outcomes
of intended action patterns.
You may have noticed that the examples we used to illustrate concrete
goals relate directly to our example of an abstract goal. We did this to point
out that abstract goals connect to concrete goals. In 1973, William Powers
argued that the self-regulation of behavior occurs by way of a hierarchical
organization of feedback loops. Because feedback loops imply goals, Powerss
argument also constituted a model of hierarchical structuring among the
goals involved in creating action.
Powerss general line of thinking ran as follows: In a hierarchical
organization of feedback systems, the output of a high level system consists
OPTIMISM, PESSIMlSM, AND SELF-REGULATION

35

of resetting reference values at the next lower level of abstraction. T o put

it differently, higher order or superordinate systems behave by providing
goals to the systems just below them. The reference values are more concrete
and restricted as one moves from higher to lower levels. Control at each
level regulates a quality that contributes to the quality controlled at the
next higher level. Each level monitors input at the level of abstraction of
its own functioning, and each level adjusts output to minimize its discrepancies. One processor is not presumed to handle functions at various levels
of abstraction. Rather, structures at various levels handle their concerns
simultaneously.
This hierarchical model per se has not been tested empirically. However, Vallacher and Wegners ( 1985) action-identification theory strongly
resembles the Powers hierarchy in some respects, and their theory has been
tested. The Vallacher and Wegner (1985) model is framed in terms of how
people think about the actions theyre taking, but it also conveys the sense
that how people think about their actions is informative about the goals
by which theyre guiding the actions. Work done within the framework of
action-identification theory has developed considerable support for the notion of hierarchicality as proposed by Powers.
Hierarchicality and the Complexity of Behavior
Although the hierarchy were discussing is in some ways simple, it has
implications for several issues in thinking about behavior (see Carver &
Scheier, 1998). Implicit here is that goals at any given level can often be
achieved by a variety of means at lower levels. This permits one to address
the fact that people sometimes shift radically the manner in which they try
to reach a goal, when the goal itself has not changed. This is common when
the quality that constitutes the higher order goal is implied in several
different kinds of lower order activities. For example, you can be helpful
(fulfilling an abstract goal) by writing a donation check, picking up discards
for a recycling center, volunteering at charity, or holding a door open for
someone else.
Just as a given goal can be obtained by way of multiple pathways, so
can a specific act be performed in the service of diverse goals. For example,
you could buy someone a gift to make him feel good, to repay a kindness,
to put him in your debt, or to satisfy a perceived holiday-season role. Thus,
a given act can have strikingly different meanings, depending on the purpose
its intended to serve. This is an important subtheme of this view on behavior:
Behavior can be understood only by identifying the goals to which it is
addressed.
A related point made by the notion of hierarchical organization concerns the fact that goals are not equal in their importance. The higher you

36

CARVER AND SCHEIER

go into the organization, the more fundamental to the overriding sense of

self are the qualities encountered. Thus, goal qualities at higher levels appear
to be intrinsically more important than those at lower levels.
Even two goals at a lower level arent necessarily equivalent in importance. In a hierarchical system there are two ways for importance to accrue
to a concrete (lower level) goal. The more directly the attainment of a
concrete goal contributes to the attainment of a valued abstract goal, the
more important is the concrete goal. Second, a concrete act that contributes
to attaining several higher level goals at once is more important than an
act that contributes to the attainment of only one such goal. With respect
to the value aspect of the expectancy-value approach, then, value accrues
(a) as one considers goals higher in a persons hierarchy of goals, (b) as a
concrete goal relates more directly to the attainment of a higher level goal,
or (c) as a concrete goal represents a step to satisfying multiple higher order
goals simultaneously.
Feelings
The model described in the preceding sections addresses control of
action. We have also suggested a way of thinking about how feelings arise
in the course of behavior, by way of another feedback process (Carver &
Scheier, 1990). This process operates along with the behavior-guiding one
and in parallel to it. One way to describe what this system does is to
say its checking on how well the behavior loop is doing at reducing its
discrepancies. Thus, the perceptual input for the affect-creating loop is a
representation of the rate of discrepancy reduction in the action system over
time. (We focus first on discrepancy-reducing loops, turning later to enlarging loops.)
We find an analogy useful here. Because action implies change between
behavioral states, consider behavior as analogous to distance (change from
one physical position to another). If the action loop deals with distance,
and if the affect loop assesses the progress of the action loop over time, then
the affect loop is dealing with the psychological equivalent of velocity
(change in physical distance over time).
We dont believe the velocity input creates affect by itself, because
a given rate of progress has different affective consequences under different
circumstances. As in any feedback system, this input is compared to a
reference value (cf. Frijda, 1986):an acceptable or desired rate of behavioral
discrepancy reduction. As in other feedback loops, the comparison checks
for a deviation from the standard.
We suggest that the result of the comparison process at the heart
of this loop (the error signal generated by the comparator) is manifest
phenomenologically in two forms. One is a hazy and nonverbal sense of
OPTIMISM, PESSIMlSM,

AND SELF-REGULATION

37

confidence or doubt. The other is affect, feeling, a sense of positiveness or

negativeness. Several studies have yielded evidence that tends to support
this view (for review see Carver & Scheier, 1998).
It took us a while to realize it, but this is essentially a cruise control
model of affect and action. That is, the system weve argued for functions
much the same as the cruise control on your car. If youre going too slowly
toward some goal in your behavior, negative affect arises. You respond to
the situation by putting more effort into your behavior, trying to speed up.
If youre going faster than you need to, positive affect arises, and you coast.
The cars cruise control is very similar. You come to a hill, which
slows you down. Your cruise control responds by feeding the engines cylinders more gas, to bring the speed back up. If you come across the crest of
a hill and are rolling downhill too fast, the system pulls back on the gas
and drags the speed back down.
When we began this section we said wed restrict ourselves at first to
discrepancy-reducing loops. Thus far weve done that, dealing only with
the context of approach. Now we turn to attempts to distance oneself from
a point of comparison, that is, discrepancy-enhrging loops.
As indicated in our earlier discussion, behavior toward anti-goals is
just as intelligible as behavior toward approach goals. But what are the
affective accompaniments to avoidance loops? The affect theory described
above rests on the idea that positive feelings arise when a behavioral system
is doing well at doing what its organized to do. The systems weve considered
thus far are organized to reduce discrepancies. Theres no obvious reason,
however, why the principle shouldnt apply just as well to systems organized
to increase discrepancies. If the system is doing well at what its organized
to do, the result should be positive affect. If it is doing poorly at what its
organized to do, the result should be negative affect.
That much seems the same across the two types of systems, but we
see a difference in the affects involved (see Figure 2.2). For both approach
and avoidance systems theres a positive pole and a negative pole, but the
positives arent quite the same, nor are the negatives. Our view of this
difference derives partly from insights of Higgins (Higgins, 1987). Following
their lead, we suggest that the affect dimension relating to discrepancyreducing loops is (in its purest form) the dimension that runs from depression
to elation. The affect dimension that relates to discrepancy-enlarging loops
is (in its purest form) the dimension running from anxiety to relief or
contentment.
As Higgins and his colleagues note, these two dimensions capture the
core qualities behind dejection-related and agitation-related affects. The
link between affect quality and type of system shown in Figure 2.2 is compatible not only with the Higgins model, but also with other theories. For
example, Roseman (1984, p. 31) has argued that joy and sadness are related

38

C A R V E R A N D SCHEIER

Discrepancyreducing
loop

well

Discrepancyenlarging
loop
well

Figure 2.2. Two sorts of affect-generating systems and the affective dimensions
we believe arise from the functioning of each. Discrepancy-reducing systems are
presumed to yield affective qualities of sadness or depression when progress is below
standard, and happiness or elation when progress is above standard. Discrepancyenlarging systems are presumed to yield anxiety when progress is below standard, and
relief or contentment when progress is above standard. From On the Self-Regulationof
Behavior, p. 138, by C. S. Carver and M. F. Scheier, 1998, New York: Cambridge
University Press. Copyright 1998 by Cambridge University Press. Reprinted with
permission.

to appetitive (moving-toward) motives, whereas relief and distress are related

to aversive (moving-away-from) motives. Davidson (1992, 1995) has argued
that the left frontal cerebral cortex is part of the neural substrate for approach
and for feelings of eagerness and depression, and that the right frontal cortex
serves as part of the neural substrate for avoidance and for feelings of anxiety
and relief.
Confidence and Doubt

In describing our view of the origin of affect, we suggested that one

mechanism yields as subjective readouts both affect and a hazy sense of
confidence versus doubt about the immediate future. This immediate sense
of confidence and doubt can have an important influence on subsequent
action. However, its influence usually doesnt occur in a psychological
vacuum.
Weve often suggested that when people experience adversity in trying
to move toward their goals, they periodically experience an interruption of
their efforts, to assess in a more deliberative way the likelihood of a successful
outcome (e.g., Carver & Scheier, 1981, 1990, 1998). In effect, people
suspend the behavioral stream, step outside it, and evaluate in a more
thoughtful way than occurs while acting. This may happen once, or often.
It may be brief, or it may take a long time. In this assessment people
presumably depend heavily on memories of prior outcomes in similar
OPTIMISM, PESSIMISM, AND SELF-REGULATION

39

situations, and they consider such things as additional resources they might
bring to bear or alternative approaches to the problem.
How do these thoughts influence the expectancies that result? In
some cases, when people retrieve chronic expectancies from memory,
the information already is expectancies, summaries of products of previous
behavior. These chronic expectancies may simply substitute for those that
were derived from immediate experience, or they may blend with and color
those immediate expectancies to a greater or lesser degree.
In some cases, however, people think more expansively about possibilities for the situations potential evolution. For these possibilities to influence
expectancies, their consequences must be evaluated. The possibilities probably are briefly played through mentally as behavioral scenarios (cf. Taylor
& Pham, 1996), leading to conclusions that influence the expectancy (e.g.,
If I try approaching it this way instead of that way, it should work better.
This is the only thing I can see to do, and it will just make the situation worse.).
It seems reasonable that this mental simulation engages the same
mechanism as handles the process that creates the sense of affect and
confidence during actual behavior. When your progress is temporarily stalled,
playing through a scenario thats confident and optimistic yields a higher
rate of perceived progress than is currently occurring. The confidence loop
thus yields a more optimistic outcome assessment than is being derived from
current action. If the scenario is negative and hopeless, it indicates a further
reduction in progress, and the confidence loop yields further doubt.

Variations in Breadth of Expectancies

The fact that goals vary in specificity-from the very concrete and
specific, to those that pertain to a particular domain of life, to the very
general-suggests that people have a comparable range of variations in
expectancies (Armor & Taylor, 1998; Carver & Scheier, 1998). T o put it
differently, you can be confident or doubtful about tying your shoes, about
finding healthy food for dinner, about winning a particular tennis match,
about performing well in socially evaluative circumstances, or about having
a good life.
Which of these sorts of expectancies matter? Probably all of them.
Expectancy-based theories tend to imply that behavior is predicted best
from expectancies when the level of specificity of the expectancy matches
that of the behavior. T o predict a specific performance, you measure a
specific, focalized expectancy. T o predict an index of many kinds of performances in a given domain, you measure a broader sort of expectancy. This
logic also suggests that to predict behavior over the broadest ranges, you
should measure a generalized expectancy. This is what we think of as disposi-

40

CARVER AND SCHEIER

tional optimism versus pessimism: the generalized expectancy of good versus

bad outcomes in life (Scheier & Carver, 1985, 1992).
Many outcomes in life have multiple causes. People also often face
situations theyve never experienced before, or situations that unfold and
change over time. In circumstances such as these, generalized expectations
may be particularly useful in predicting behavior and emotional reactions
(Scheier & Carver, 1985).
The principles that apply to a focused sense of confidence versus doubt
also apply to the generalized sense of optimism and pessimism. The difference
between these cases is simply the breadth of the goals and the sense of
confidence or doubt. When we talk about generalized optimism and pessimism, the sense of confidence at issue is more diffuse and broader in scope
than when we talk about specifics such as test anxiety or social anxiety.
Thus, when confronting a challenge or threat, optimists should tend
to take a posture of confidence (even if progress is presently difficult or
slow). Pessimists should be more doubtful and hesitant. This divergence
should also be displayed-and maybe even be amplified-under conditions
of serious adversity. Optimists are likely to assume the adversity can be
handled successfully, in one fashion or another, whereas pessimists are likely
to anticipate disaster.
Our basic conceptualization of behavioral self-regulation addresses both
approaching goals and avoiding anti-goals. This suggests that both approach
and avoidance processes should involve expectancies (as well as affects).
We have done no work that explicitly addresses this distinction, but this
distinction may well bear on a measurement issue that has arisen in this
literature. Our measure of optimism (the Life Orientation Test, or LOT)
typically forms two factors. One factor is composed of items referencing
good outcomes, the other is composed of items referencing bad outcomes
(see, Marshall, Wortman, Kusulas, Hervig, & Vickers, 1992; Scheier &
Carver, 1985, 1992; Scheier, Carver, & Bridges, 1994). It may be that
one component of the measure reflects confidence about attaining positive
outcomes, and the other component reflects confidence about avoiding
undesired outcomes. This is an interesting possibility for future exploration.
Behavioral Consequences of Confidence and Doubt

As indicated earlier, the expectancies that people hold while attempting to attain goals can be influenced by many sources of information, an
important source being the consolidated set of expectancies they have in
memory. Whatever their source, the expectancies with which people return
to action are reflected in subsequent behavior. If expectations are for a
successful outcome, the person returns to effort toward the goal. If doubts
are strong enough, the result is an impetus to disengage from effort, and
OPTIMISM, PESSIMISM, AND SELF-REGULATION

41

potentially from the goal itself (Carver & Scheier, 1981,1990, 1998;Klinger,
1975; Kukla, 1972; Wortman & Brehm, 1975).
This theme-divergence in response as a function of confidence versus
doubt-is a very important one, applying to a surprisingly broad range of
literatures (see chapter 11 of Carver & Scheier, 1998). One of the more
obvious of these literatures is that of stress and coping. The differences in
how people approach stressful circumstances have important implications
for the manner in which people cope with adversities. Optimists take a
goal-engaged approach to coping, whereas pessimists tend to show signs of
coping in ways implying disengagement (e.g., Carver et al., 1993; Scheier
et al., 1989).
Sometimes the disengagement that follows from doubt is overt, but
disengagement also can take the form of mental disengagement-off-task
thinking, daydreaming, and so on. Although this can sometimes be useful
(self-distraction from a feared stimulus may permit anxiety to abate), it can
also create problems. If theres time pressure, mental disengagement impairs
performance, as time is spent on task-irrelevant thoughts. Further, mental
disengagement often cant be sustained, as situational cues force a reconfronting of the task. In such cases, the result is a phenomenology of repetitive
negative rumination, which often focuses on self-doubt and perceptions of
inadequacy. This cycle is both unpleasant and performance-impairing.

Is Disengagement Good or Bad?

A question that needs to be considered is whether the disengagement
tendency is good or bad. The answer is that it depends. O n the one hand,
disengagement (at some level, at least) is a necessity. Disengagement is a
natural and indispensable part of self-regulation. If we are ever to turn away
from efforts at unattainable goals, if we are ever to back out of blind alleys,
we must be able to disengage, to give up and start over somewhere else.
The importance of disengagement is particularly obvious with regard
to concrete, low level goals: People must be able to remove themselves
from literal blind alleys and wrong streets, give up plans that have become
disrupted by unexpected events, even spend the night in the wrong city if
theyve missed the last plane home. The tendency is also important, however,
with regard to some higher level goals. A vast literature attests to the
importance of disengaging and moving on with life after the loss of close
relationships (e.g., Cleiren, 1993; Orbuch, 1992; Stroebe, Stroebe, & Hansson, 1993). People sometimes must even be willing to give up values that
are deeply embedded in the self, if those values create too much conflict
and distress in their lives.
The choice between continuing effort and giving up presents a variety
of opportunities for things to go awry, however. Its possible to stop trying
42

CARVER AND SCHEIER

prematurely, thereby creating potentially serious problems for oneself; its

also possible to hold onto unattainable goals for too long, and prevent
oneself from taking adaptive steps toward new goals (Carver & Scheier,
1998). But both continuing effort and giving up are necessary parts of the
experience of adaptive self-regulation. Each plays an important role in the
continuing flow of behavior.
More specifically, giving up is a functional and adaptive response when
it leads to the taking up of other goals, whether these are substitutes for the
lost goal or simply new goals in a different domain. Giving up thus can
provide an opportunity to re-engage and move ahead again. In such cases
(which appear to include all of the examples in the preceding paragraph),
giving up is occurring in service to the broader function of returning the
person to an engagement with life (Scheier & Carver, in press).
Thus, giving up a goal because of doubts about its attainment can be
adaptive, when doing so leads to engagement with another goal. This principle appears to apply to goal values and doubts that extend fairly deeply into
the sense of self. It probably does not extend, however, to the sense of
generalized pessimism. Generalized pessimism is a sense of doubt that focuses
not on just a single behavioral goal or even a broader behavioral domain,
but on ones entire life space. A person who feels a deep sense of pessimism
about all of life is a person who has nothing else to turn to. If this person
gives up on that life space, there is nothing else to take its place. This absence
of substitute goals-which seems to exemplify this persons situation-is
a very serious problem (see chap. 18 of Carver & Scheier, 1998; Scheier
& Carver, in press).

ENGAGEMENT VERSUS GIVING UP:

EXPLORATION IN A NEW DIRECTION
An issue that bears further discussion is the divergence we postulate
among responses to confidence and doubt. Weve long argued for the existence of a psychological watershed between two sets of responses to adversity
(Carver & Scheier, 1981).One set of responses consists of continued comparisons between present state and goal, and continued efforts. The other set
consists of disengagement from such comparisons, and quitting of efforts.
Just as rainwater falling on a mountain ridge ultimately flows to one side
of the ridge or the other, so do behaviors ultimately flow to one or the
other of these sets.
Our initial reason for taking this position stemmed largely from several
demonstrations that self-focused attention creates diverging efforts on both
information seeking and behavior, as a function of confidence versus doubt
about success (Carver & Scheier, 1981). We arent the only ones to have
OPTIMISM, PESSIMISM, AND SELF-REGULATION

43

emphasized a disjunction among these two classes of responses, however.

A number of other theorists have done so, for reasons of their own.
An early model incorporating such a disjunction in behavior was
proposed by Kukla (1972). Another is the integration by Wortman and
Brehm (1975) of reactance and helplessness: the argument that threats to
control produce attempts to regain control and that perceptions of loss of
control produce helplessness. Brehm and his collaborators (Brehm & Self,
1989; Wright & Brehm, 1989) have more recently developed an approach
to task effort that resembles that of Kukla (1972), although their way of
approaching the conceptual problem is somewhat different.
Over the past several years, we have become interested in the possibility
of relating the dichotomy between effort and giving up to the principles of
catastrophe theory. Catastrophe theory is a mathematical model that addresses the creation of discontinuities, bifurcations, or splittings (Brown,
1995; Stewart & Peregoy, 1983; Thom, 1975). A catastrophe occurs when
a small change in one variable produces an abrupt (and usually large) change
in another variable.
Although several types of catastrophe exist, the one thats been examined most frequently regarding psychological issues is the cusp catastrophe,
in which two control parameters (variables that change the dynamics of
the system-roughly akin to the independent variables of an experiment)
influence an outcome. Figure 2.3 portrays its three-dimensional surface. The
control parameters are x and z,y is the outcome. At low values of 7 , the
surface of the figure expresses a roughly linear relationship between x and
y, such that as x increases, so does y. As s (the second control parameter)
increases, the relationship between x and y becomes less linear. It first shifts
toward something like a step function. With further increase in 5, the
x-y relationship becomes even more clearly discontinuous-the outcome
is either on the top surface or on the bottom. Thus, changes in z cause a
change in the way x relates to y.
An important feature displayed by a catastrophe is hysteresis. A simple
characterization of hysteresis is that in some range of z theres a foldover
in the middle of the x-y relationship. At some range of 5, a region of x
exists with more than one value of y. The upper level and lower levels of
possible values of y can be seen at the front of the surface of Figure 2.3,
between points a and b. (The area where the fold goes backward is unstable;
only the top and bottom surfaces have meaning.)
Another way of characterizing hysteresis is captured by the statement
that the systems behavior depends on the systems recent history (Brown,
1995). That is, as you move into the middle zone of the front of the
catastrophe surface, it matters which side youre coming from. If the system
is moving from area c into the zone of hysteresis, it stays on the bottom
surface until it reaches point b, where it jumps to the top surface. If the

44

CARVER AND SCHEZER

Figure 2.3. Three-dimensional depiction of a cusp catastrophe. Variables x and z

are control parameters, y is the systems behavior, the dependent variable. A cusp
catastropheexhibits a region of hysteresis (between Points a and b), in which a given
x has two values of y (the top and bottom surfaces). Entering the zone of hysteresis
from area c results in an abrupt shift (at Point b on the x axis) from the lower to the
upper portion of the surface (right arrow). Entering from area d results in an abrupt
shift (at value a on the x axis) from the upper to the lower portion of the surface (left
arrow). Thus, the disjunction between portions of the surface occurs at two different
values of x, depending on the starting point. From On the Self-RegulationofSehavior
(pp. 267, 279), by C. S. Carver and M. F. Scheier, 1998, New York: Cambridge
University Press. Copyright 1998 by Cambridge University Press. Adapted with
permission.

system is moving from area d into the zone of hysteresis, it stays on the top
surface until it reaches point a, where it jumps to the bottom surface.
Earlier in this section we said that several theories assume a disjunction
between engagement of effort and giving up. In those models (including
ours), theres a point at which effort seems fruitless and the person stops
trying. Earlier we simply emphasized that these models all assumed a discontinuity. Now we look at the discontinuity more closely and suggest that the
phenomena addressed by these theories may embody a catastrophe.
Figure 2.4 shows a cross-section of a cusp catastrophe, similar to the
front edge of Figure 2.3. This figure displays a hypothesized region of
hysteresis in the engagement versus disengagement function. In that region,
where task demands are close to peoples perceived limits to perform, there
should be greater variability in effort or engagement, as some people are on
the top surface of the catastrophe, and others are on the bottom surface.
Some people would be continuing to exert efforts, at the same point where
others would be exhibiting a giving-up response.
OPTIMISM, PESSIMISM, AND SELF-REGULATION

45

Figure 2.4. A catastrophe model of effort versus disengagement. From On the SelfRegulation of Behavior (p. 291), by C. S. Carver and M. F. Scheier, 1998, New
York: Cambridge University Press. Copyright 1998 by Cambridge University Press.
Reprinted with permission.

Recall that the catastrophe figure also conveys the sense that behavioral
history matters. A person who enters the region of hysteresis from the
direction of high confidence (who starts out confident but confronts many
cues indicating the confidence is unfounded) will continue to display efforts
and engagement, even as the cues imply increasingly less basis for confidence.
A person who enters the region of hysteresis from low confidence (who
starts out doubtful but confronts cues indicating the doubt is unfounded) will
continue to display little effort, even as cues imply more basis for confidence.
This model helps indicate why it can be so difficult to get someone
with strong and chronic doubt about success in some domain of behavior
to exert real effort and engagement in that domain. It also suggests why a
confident person is so rarely put off by encountering difficulties in the
domain where the confidence lies. To put it more broadly, it helps show
why optimists tend to stay optimistic and pessimists tend to stay pessimistic,
even when the current circumstances of these two sorts of people are identical
(i.e., in the region of hysteresis).
We have suggested elsewhere that the Wortman and Brehm (1975)
model, the Brehm and Self (1989), and the Kukla (1972) models can be
fit to this curve (Carver & Scheier, 1998).Although none of those authors
proposed that their models have a region of hysteresis, we think a plausible
case can be made for the presence of that feature in the phenomena addressed
by their theories. We should acknowledge, however, that our position on
this matter is speculative. No one has studied the processes of effort and
disengagement in a truly parametric manner that would allow careful investigation of this question.
In contemplating such studies, its important to keep in mind that the
catastrophe cross-section (see Figure 2.4) is the picture that emerges under

46

CARVER AND SCHEIER

catastrophe theory only once a clear region of hysteresis has begun to develop.
Further back, the catastrophe model more closely resembles a step function.
Even further back, the relation is more linear (see Figure 2.3). An implication
is that in trying to show a catastrophe, researchers need to engage the
control variable thats responsible for bringing out the bifurcation in the
surface (that is, Axis z in Figure 2.3). If this isnt done, the hysteresis would
be less observable, even if the research procedures were otherwise suitable
to observe it.
What is the control variable that induces the bifurcation? We think
that in the motivational models under discussion-and perhaps much more
broadly-the control parameter is importance. The subjective sense of importance has several sources. As we noted earlier, the more deeply an activity
connects to the abstract sense of self, the more important the activity. The
sense of importance also arises from social pressure from others (Tesser,
1980),from time pressure, and from self-imposed pressures. Theres a common thread among important events: They demand mental resources. We
suspect that almost any strong pressure that demands resources will induce
similar bifurcating effects.
This hypothesis has a number of interesting implications. For example,
it suggests that when a person is currently pessimistic about outcomes in
some domain of life, that pessimism will exert a more profound influence
when the situation is experienced subjectively as important. When pressure
is high, when the behavior is important (when z is large), a region of
hysteresis emerges. For the person on the bottom part of the surface, effort
is potentially possible, but for it to happen now requires a high degree of
confidence (see Figure 2.5). That is, as importance increases (the bottom
portion of Figure 2.5), the region of hysteresis spreads increasingly wider,
causing its edge (where the shift to effort actually takes place) to be more
extreme on the confidence dimension.
Many problems in living seem to fit this picture, including a variety
of all-or-none, black-and-white thinking. Of greatest importance at present
is the fact that people who are crippled by doubts in an important domain
of life would appear to live out that domain of life on the lower portion of
the front of the catastrophe surface. Given their doubt, its hard for them
to make efforts in whatever domain the doubts exist precisely because that
domain matters so much. That is, theyre facing a huge challenge. T o get to
the point of real effort (the point where they jump from the lower to the
upper surface), they need to be very confident (see bottom of Figure 2.5).
Because theyre not, the effort doesnt happen. Theyve lost the sense that
they can make their lives better, and feel stranded miles away from the
possibility of re-engagement with life.
Near the back of the surface, the relation between the other two
variables is more gradual and more linear. At the back, a slight increment
OPTIMISM, PESSIMISM, AND SELF-REGULATION

47

Corddence
12

>

Importance

Figure 2.5. The so-called bifurcation set of the cusp catastrophe in Figure 2.3. This
is essentially a downward projection of the two edges of the hysteresis onto a flat
plane. It shows that as importance increases and the degree of hysteresis increases
correspondingly, a person on the lower surface must move farther along the confidence
dimension in order to shift to the top surface. With relatively low importance, the
person must reach only Point 1, but as importance increases, the confidence needed
continues to increase (Points 2 and 3). From On the Self-Regulation of Behavior
(p. 298), by C. S. Carver and M. F. Scheier, 1998, New York: Cambridge University
Press. Copyright 1998 by Cambridge University Press. Adapted with permission.

in confidence gains something in engagement. The cost of a doubt is gradual

rather than precipitous. The person is able to stay engaged in the process
of goal-pursuit, and staying engaged is a critical determinant of success.
The key to handling some kinds of problems, then, may be to reduce
the pressure, to get some psychological distance. We think that creating
distance, or reducing pressure, implies moving toward the back of the catastrophe plane. This diminishes the emotion-evoking potential of the event.
It also ameliorates the impact of doubts on actions, by making the impact
of doubt less abrupt.

CONCLUDING COMMENT
In this chapter weve provided a brief overview of the broad model of
behavior from which our own interest in optimism first grew. Certainly
thinking about optimism and pessimism per se does not require all of the
complexities of that model (for example, the notion of feedback control
of action). However, we want to make apparent the fact that our view of
optimism connects to a dynamic and very broad model of the creation of
behavior and feelings. This connection means that optimism need not be
a topic that stands apart from the rest of psychology. This connection

48

CARVER AND SCHElER

also provides a broader context for studies of behavioral and emotional

consequences of optimism and pessimism, topics that are pursued in later
chapters of this book.

REFERENCES
Armor, D. A., & Taylor, S. E. (1998). Situated optimism: Specific outcome expectancies and self-regulation. In M. Zanna (Ed.), Advances in experimental social
psychology (Vol. 29, 309-379). San Diego, CA: Academic Press.
Austin, J. T., & Vancouver, J. B. (1996). Goal constructs in psychology: Structure,
process, and content. Psychological Bulktin, 120, 338-375.
Brehm, J. W., & Self, E. A. (1989). The intensity of motivation. Annual Rewiew
of Psychology, 40, 109-131.
Brown, C. (1995). Chaos and catastrophe theories (Quantitative applications in the
social sciences, No. 107). Thousand Oaks, CA: Sage.
Carver, C. S., Pozo, C., Harris, S. D., Noriega, V., Scheier, M. F., Robinson, D. S.,
Ketcham, A. S., Moffat, F. L., & Clark, K. C. (1993). How coping mediates
the effect of optimism on distress: A study of women with early stage breast
cancer. Journal of Personality and Social Psychology, 65, 375-390.
Carver, C. S.,& Scheier, M. F. (1981). Attention and self-regulation: A control-theory
approach to human behawior. New York: Springer-Verlag.
Carver, C. S., & Scheier, M. F. (1990). Origins and functions of positive and
negative affect: A control-process view. Psychological Rewiew, 97, 19-35.
Carver, C. S., & Scheier, M. F. (1998). O n the self-regulation of behawior. New York:
Cambridge University Press.
Cleiren, M. (1993). Bereawement and adaptation: A comparative study of the aftermath
of death. Washington, DC: Hemisphere.
Davidson, R. J. (1992). Anterior cerebral asymmetry and the nature of emotion.
Brain and Cognition, 20, 125-151.
Davidson, R. J. (1995). Cerebral asymmetry, emotion, and affective style. In R. J.
Davidson & K. Hugdahl (Eds.), Bruin asymmetry (pp. 361-387). Cambridge,
MA: MIT Press.
Elliott, E. S., 61 Dweck, C. S. (1988). Goals: An approach to motivation and
achievement, Journal of Personality and Social Psychology, 54, 5-12.
Frijda, N. H. (1986). The emotions. Cambridge, UK: Cambridge University Press.
Higgins, E. T. (1987). Self-discrepancy:A theory relating self and affect. Psychological
Review, 94, 319-340.
Klinger, E. (1975). Consequences of commitment to and disengagement from incentives. Psychological Review, 82, 1-25.
Kukla, A. (1972). Foundations of an attributional theory of performance. Psychological Review, 79, 454-470.
OPTlMlSM, PESSlMlSM, AND SELF-REGULATION

49

MacKay, D. M. (1966). Cerebral organization and the conscious control of action.

In J. C. Eccles (Ed.), Brain and conscious experience (pp. 422-445). Berlin,
GDR: Springer-Verlag.
Marshall, G. N., Wortman, C. B., Kusulas, J. W., Hervig, L. K., & Vickers, R. R.,
Jr. (1992). Distinguishing optimism from pessimism: Relations to fundamental
dimensions of mood and personality. journal of Personality and Social Psychology,
62, 1067-1074.
Miller, G. A., Galanter, E., & Pribram, K. H. (1960). Plans and the structure of
behavior. New York: Holt, Rinehart, & Winston.
Orbuch, T. L. (Ed.). (1992). Close relationship loss: Theoretical approaches. New York:
Springer-Verlag.
Pervin, L. A. (1982). The stasis and flow of behavior: Toward a theory of goals.
In M. M. Page & R. Dienstbier (Eds.), Nebraska symposium on motivation (Vol.
30, pp. 1-53). Lincoln, NE: University of Nebraska Press.
Pervin, L. A. (Ed.). (1989). Goal concepts in personality and socialpsychology. Hillsdale,
NJ: Erlbaum.
Powers, W. T. (1973). Behavior: The control of perception. Chicago: Aldine.
Roseman, I. J. (1984). Cognitive determinants of emotions: A structural theory.
In P. Shaver (Ed.), Review of personality and social psychology (Vol. 5, pp.
11-36). Beverly Hills, CA: Sage.
Scheier, M. F., & Carver, C. S. (1985). Optimism, coping and health: Assessment
and implications of generalized outcome expectancies. Health Psychology, 4,
2 19-247.
Scheier, M. F., & Carver, C. S. (1992). Effects of optimism on psychological
and physical well-being: Theoretical overview and empirical update. Cognitive
Therapy and Research, 16, 201-228.
Scheier, M. F., & Carver, C. S. (in press). Adapting to cancer: The importance of
hope and purpose. In A. Baum & B. L. Andersen (Eds.), Psychosocial interventions for cancer. Washington, DC: American Psychological Association.
Scheier, M. F., Carver, C. S., & Bridges, M. W. (1994). Distinguishing optimism
from neuroticism (and trait anxiety, self-mastery, and self-esteem): A reevaluation of the Life Orientation Test. Journal of Personality and Social Psychology,
67, 1063-1078.
Scheier, M. F., Matthews, K. A., Owens, J. F., Magovem, G. J., Sr., Lefebvre, R. C.,
Abbott, R. A., & Carver, C. S. (1989). Dispositional optimism and recovery
from coronary artery bypass surgery: The beneficial effects on physical and
psychological well being. Journal of Personality and Social Psychology, 57, 10241040.
Stewart, I. N., & Peregoy, P. L. (1983). Catastrophe theory modeling in psychology.
Psychological Bulletin, 94, 336-362.
Stroebe, M. S., Stroebe, W., & Hansson, R. 0. (Eds.). (1993). Handbook ofbereavement: Theory, research, and intervention. Cambridge, UK: Cambridge University Press.

50

CARVER AND SCHEIER

Taylor, S. E., & Pham, L. B. (1996). Mental stimulation, motivation, and action.
In P. M. Gollwitzer & J. A. Bargh (Eds.), The psychology of action: Linking
cognition and motivation to behavior (pp. 219-235). New York: Guilford.
Tesser, A. (1980). When individual dispositions and social pressure conflict: A
catastrophe. Human Relations, 33, 393-407.
Thom, R. (1975). Structural stability and morphogenesis. Reading, MA: Benjamin.
Vallacher, R. R., & Wegner, D. M. (1985). A theory ofaction identification. Hillsdale,
NJ: Erlbaum.
Wortman, C. B., & Brehm, J. W. (1975). Responses to uncontrollable outcomes:
An integration of reactance theory and the learned helplessness model. In
L. Berkowitz (Ed.), Advances in experimental social psychology (Vol. 8, pp. 277336). New York: Academic Press.
Wright, R. A., & Brehm, J. W. (1989). Energization and goal attractiveness. In L. A.
Pervin (Ed.), Goal concepts in personality and social psychology (pp. 169-210).
Hillsdale, NJ: Erlbaum.

OPTIMISM, PESSIMISM, AND SELF-REGULATION

51

3
OPTIM SM, PESSIM SM, A
EXPLANATORY STYLE
JANE E. GILLHAM, ANDREW J. SHATTE, KAREN J. REIVICH,
AND MARTIN E. P. SELIGMAN

Dictionary definitions of optimism encompass two related concepts.

The first is a hopeful disposition or a conviction that good will ultimately
prevail. The second broader conception refers to the belief, or the inclination
to believe, that the world is the best of all possible worlds. In psychological
research, optimism has referred to hopeful expectations in a given situation
(Scheier & Carver, 1988) and recently has referred to general expectancies
that are positive (Scheier & Carver, 1993). This more generalized expectancy, or dispositionaloptimism, is related to a variety of indices of psychological and physical health. Individuals who score high on measures of dispositional optimism report fewer depressive symptoms, greater use of effective
coping strategies, and fewer physical symptoms than do pessimistic individuals (for reviews, see Scheier & Carver, 1992, 1993).
Perhaps consistent with the second, broader definition of optimism,
the terms optimism and pessimism have recently been applied to the ways
in which people routinely explain events in their lives (Seligman, 1991).

53

People are optimistic when they attribute problems in their lives to temporary, specific, and external (as opposed to permanent, pervasive, and internal)
causes. A n optimistic explanatory style is associated with higher levels
of motivation, achievement, and physical well-being and lower levels of
depressive symptoms (for a recent review, see Buchanan & Seligman, 1995).
Psychologists interested in optimism tend to reside in one of two
parallel universes. In each, similar terms apply and similar findings are
obtained. Until recently, however, there has been surprisingly little discussion of the relationship between dispositional optimism and explanatory style. Scheier and Carver (1993) argued that dispositional optimism
and explanatory style theories are conceptually linked. However, several
researchers caution that causal attributions and predictions can be unrelated (Abramson, Alloy, & Metalsky, 1989; Hammen & Cochran, 1981;
Zullow, 1991). Given these conflicting views, it is important to clarify the
link between explanations and expectations both theoretically and empirically.
Research on dispositional optimism is reviewed in chapter 2 of this
volume. In this chapter, we describe the explanatory style construct of
optimism, present some of the major research findings from this literature,
and discuss the theoretical relationship of explanatory style to dispositional
optimism. Research exploring the association of these constructs is reviewed.
Finally, we discuss questions for future research and ways in which the
investigators from one research tradition can learn from the other.

EXPLANATORY STYLE THEORIES

Reformulated Learned Helplessness Theory

When events, particularly negative ones, occur in our lives, we search

for an explanation. According to the reformulated learned helplessness
theory (RLHT; Abramson, Seligman, & Teasdale, 1978), the manner in
which we routinely explain events in our lives can drain our motivation,
reduce our persistence, and render us vulnerable to depression. Alternatively,
our explanatory style can inspire us to problem solve and make us resilient
in the face of adversity. The RLHT describes three dimensions on which
explanations can vary: (a) internal versus external, (b) stable versus unstable,
and (c) global versus specific. Pessimistic explanations for negative events
are those that are more internal, stable, and global, that is, adversity is
attributed to characteristics of ones self, factors that are likely to endure
through time, and circumstances that affect many domains of our lives.
Optimistic explanations show the reverse pattern. Negative events are
ascribed to environmental or situational factors that are temporary and

54

GILLHAM ET AL.

affect few domains of ones life. In explaining a conflict in an intimate

relationship, for example, a pessimistic person might tell himself or herself,
Im not loveable (internal, stable, and global), whereas an optimistic
person may speculate along the following lines: Weve both been under a
great deal of stress at work lately (an external, unstable, and specific attribution). When explaining positive events, pessimistic and optimistic patterns
reverse. Pessimistic explanations for positive events are external, unstable,
and specific. That is, the source of success and good fortune is seen as
fleeting, of limited influence, affecting few areas of life, and as caused by
other people or circumstances. In contrast, optimistic explanations are internal, stable, and global.
The RLHT predicts that pessimistic and optimistic explanations will
lead to different expectations about the future. Individuals who attribute
negative events to stable and global causes will expect outcomes to be
uncontrollable in the future. These individuals will be vulnerable to helplessness in the face of adversity. In contrast, individuals who attribute negative
events to unstable or specific causes will expect to exert control in the
future, and hence, will be more resilient. The RLHT proposes that the
stability of the cause is related to the duration of helplessness symptoms,
the pervasiveness (globality) of the cause is related to the generalization of
helplessness across multiple situations, and the internality of the cause is
related to the occurrence of self-esteem deficits in depression.
Hopelessness Theory of Depression
Abramson et al. (1989) argued that the stable and global dimensions
of explanatory style have a stronger impact on motivation and depression
than does the internal dimension. Thus, blaming the conflict described
above on the belief that love never endures (an external, stable, and
global attribution) also will lead to helplessness. According to the hopelessness theory (HT), a revision of the RLHT, three types of interpretations
can put one at risk for depression following a negative event. First, the
event may be attributed to stable and global causes. Second, negative or
catastrophic consequences of the event may be inferred. Third, negative
characteristics about the self may be inferred. When these interpretations
are made frequently, they lead to negative expectations about the occurrence
of highly valued outcomes (a negative outcome expectancy) and to negative
expectations about ones ability to change the likelihood of these outcomes
(a helplessness expectancy). According to the HT, these negative expectations are the proximal cause of a subtype of depression characterized by
retarded initiation of voluntary response, sad affect, lack of energy, apathy,
psychomotor retardation, sleep disturbance, difficulty in concentration, negative thinking, and suicidal ideation.
EXPLANATORY S m E

55

EXPLANATORY STYLE RESEARCH

Several hundred published studies have investigated the relationship
between explanatory style and various aspects of psychological and physical
health. We present some of the major findings from this literature (for a
recent review of research on explanatory style, see Buchanan & Seligman, 1995).
Measurement
As noted in the introduction of this volume, researchers interested in
explanatory style have used several measures in their investigations (see
also Reivich, 1995). The majority of studies of adults have used the Attributional Style Questionnaire (ASQ; Peterson, Semmel, von Baeyer, Abramson,
Metalsky, & Seligman, 1982; Seligman, Abramson, Semmel, & von Baeyer,
1979). The ASQ yields composite scores for explanatory style for positive
events (CP) and negative events (CN), as well as scores for 6 subscales
(internal, stable, and global for positive and negative events). A n overall
composite (CP-CN) is calculated by subtracting the negative-event composite from the positive-event composite.
Over the past 10 years, investigations of explanatory style have frequently used one of the two expanded ASQs (E-ASQ;Metalsky, Halberstadt,
& Abramson, 1987; Peterson & Villanova, 1988). The E-ASQs have been
the favorite measures in investigations of the HT (Metalsky &Joiner, 1992;
Metalsky, Joiner, Hardin, & Abramson, 1993). These scales contain more
negative events, and are, therefore, more reliable than the original ASQ.
Because the Extended-ASQs do not sample positive events, only explanatory
style for negative events can be assessed with these instruments. Recently,
Abramson, Alloy, and Metalsky ( 1998)developed the Cognitive Style Questionnaire (CSQ) to assess the tendency to infer stable and global causes,
negative consequences, and negative characteristics about the self.
The Childrens Attributional Style Questionnaire (CASQ; Kaslow,
Tannenbaum, & Seligman, 1978) is the most widely used measure of explanatory style in children. The CASQ presents 48 hypothetical events (24
positive and 24 negative events) in a forced-choice format. This instrument
yields the same composite and subscale scores as the original ASQ.
All of these measures are self-report measures. However, some studies
have assessed explanatory style by analyzing speeches, statements, journal
entries, and other written materials using the Content Analysis of Verbatim
Explanations (CAVE) technique (Peterson, Bettes, & Seligman, 1985). In
CAVE, causal explanations for positive and negative events are extracted
and then coded for their internality, stability, and globality (for additional
discussion of the CAVE technique, see chapter 6 in this volume).

56

GILLHAM ET AL.

Depression

Researchers have evaluated two hypotheses regarding the link between

explanatory style and depression. The weak hypothesis, that a pessimistic
explanatory style is associated with depressive symptoms, is supported by
many studies of adults (for a review, see Robins & Hayes, 1995) and children
(for reviews, see Gladstone & Kaslow, 1995; Joiner & Wagner, 1995a).
The stronger hypothesis, that a pessimistic explanatory style is a risk
factor for depressive symptoms, has met with conflicting findings. Several
studies have investigated the ability of explanatory style to predict changes
in depressive symptoms over time. In some of these studies, CP-CN and
CN scores predict depressive symptoms, with pessimistic explanatory style
predicting increases in symptoms over time (e.g., Golin, Sweeney, & Schaeffer, 1981; Seligman et al., 1984). In other investigations, however, explanatory style does not significantly predict changes in symptoms (e.g., Bennet
& Bates, 1995; Hammen, Adrian, & Hiroto, 1988; Tiggemann, Winefield,
Winefield, & Goldney, 1991).
Few studies have investigated the link between explanatory style and
clinical depression. Recently, Abramson, Alloy, Hogan et al. (1998) found
that episodes of clinical depression in college students were predicted by
their scores on the CSQ and a measure of dysfunctional attitudes central
to Beck's theory of depression (DAS; Weissman, 1979).
In adults, recovery from depression during cognitive therapy is linked
to improvement in explanatory style (Seligman et al., 1988). In addition,
a pessimistic explanatory style predicts relapse following the termination of
therapy (DeRubeis & Hollon, 1995; Ilardi, Craighead, & Evans, 1997).
Programs that improve explanatory style help prevent depressive symptoms
in adults and children (Gillham, Reivich, Jaycox, & Seligman, 1995; Seligman, Schulman, DeRubeis, & Hollon, 1999).
Most studies investigating the relationship between explanatory style
and depressive symptoms have examined the overall (CP-CN ) composite
or the CN composite. Explanatory style for positive events is less consistently
related to depressive symptoms (Peterson & Seligman, 1984; Robins &
Hayes, 1995). For example, Seligman and colleagues (1984) found that the
CN but not the CP composite predicted depressive symptoms in children.
Achievement and Productivity

Optimistic explanations for negative events are linked to higher academic achievement in college students and increased job productivity (for
a review, see Schulman, 1995). Students who explain events in an optimistic
manner are more likely than those who explain them in a pessimistic one
to exceed the level of academic performance predicted by their high school
EXPLANATORY STYLE

57

class rank, SAT scores, and achievement test scores (Schulman, 1995). In
children, attributions of academic failure to stable and global factors, such
as lack of ability, are associated with decreased persistence and more negative
expectations for future success (e.g., Dweck, 1975). In contrast to these
findings, however, Satterfield, Monahan, and Seligman (1997) found that
law students with a pessimistic explanatory style actually outperformed those
who were considered optimists.
Physical Health
Explanatory style also appears to be linked to physical health (for a
review, see Peterson & Bossio, 1991). Optimistic college students report
fewer physical illnesses, make fewer doctor visits, and feel more able to
prevent health problems than their pessimistic peers (Peterson, 1988; Peterson & De Avila, 1995). Explanatory style in young adulthood has been
found to predict physical health in middle age (Peterson, Seligman, &
Vaillant, 1988). More recently, Buchanan ( 1995) found that explanatory
style predicted long-term survival among men who had had one heart attack.
Improving explanatory style may enhance physical well-being. Selig
man and colleagues tracked visits to a student health center by university
students who participated in a program designed to improve explanatory
style. Compared to control participants, these students made more preventative health care visits and fewer illness-related visits (Buchanan, Gardenswartz, & Seligman, 1999).
Other Life Arenas

An optimistic explanatory style has also been linked to many other

outcomes. For example, studies suggest that explanatory style is linked to
athletic performance (Rettew & Reivich, 1995; Seligman, Nolen-Hoeksema,
Thornton, & Thornton, 1990). Basketball teams whose members gave more
optimistic explanations for losses were more likely than those with pessimistic explanatory styles to come back from a loss by winning the next game
(Rettew & Reivich, 1995).In another study, members of a university varsity
swim team were given false feedback after they performed their best event.
Each swimmer was given a time for the event that was slightly longer
than his or her actual time (Seligman et al., 1990). Although the actual
discrepancies were small, they were large enough to make the difference
between good and bad times. A n optimistic explanatory style was associated
with greater resilience and predicted greater performance following this
negative feedback.
Explanatory style is also linked to marital satisfaction. Couples who
attribute marital events to factors that are external to their spouse, unstable,

58

GILLHAM ET AL.

and specific report higher levels of marital satisfaction than those with the
reverse explanatory style. An optimistic explanatory style for marital events
also appears to predict future marital satisfaction (e.g., Fincham & Bradbury, 1993)
Finally, explanatory style has been linked to political victory (Zullow,
1995), and military assertiveness (Satterfield & Seligman, 1994). For example, Zullow and Seligman (1990) content-analyzed (using CAVE) nomination acceptance speeches made by presidential candidates from 1948 to
1984. They found that defeat in the presidential elections was predicted by
a combination of explanatory style and rumination (the tendency to dwell
on negative events). Specifically, in 9 of the 10 elections, the candidate
whose style was scored as more pessimistically ruminative lost the election.

Related Findings From the Dispositional Optimism Literature

For the most part, findings from the dispositional optimism literature
parallel those from the explanatory style literature. For example, dispositional
optimists report fewer depressive symptoms (Carver & Gaines, 1987) and
fewer physical health problems than pessimistic people (Scheier & Carver,
1992). Dispositional optimism predicts greater psychological well-being and
faster physical recovery after coronary artery bypass surgery (Scheier &
Carver, 1992).
A few discrepancies exist across the literatures. For example, dispositional optimism predicts postpartum depression (Carver & Gaines, 1987).
In contrast, Manly, McMahon, Bradley, and Davidson (1982) found that
explanatory style was not related to postpartum depression. Cutrona (1983)
found that explanations regarding daily events and child care stress were
not related to postpartum depressive symptoms, but causal attributions for
so-called maternity blues were significantly related to depressive symptoms.

EXPLANATORY STYLE AND OPTIMISM

Expectations about the future are a crucial component of both the

RLHT and HT because they provide the link between causal explanations

and helplessness deficits. Thus, according to Abramson and colleagues

(1978), the attribution merely predicts the recurrence of the expectations
but the expectation determines the occurrence of the helplessness deficits
(P. 58).

Research on Explanatory Style and Expectations

Despite their theoretical importance, few investigations of explanatory
style have assessed expectations. One recent exception is a study by Metalsky
EXPLANATORY STYLE

59

and colleagues (1993). Among students who received a low grade, those
who attributed negative achievement events to stable and global factors
expected to perform poorly in the future. These expectations predicted
changes in mood.
A handful of studies have assessed both explanatory style and dispositional optimism, using the Life Orientation Test (LOT; Scheier & Carver,
1985).These studies have yielded inconsistent findings. Scheier and Carver
(1992) reported correlations between the ASQ CP-CN composite and the
LOT that are in the high teens and low twenties. Kamen (1989) reported
a correlation of -25 between the LOT and the ASQ CN composite. In
contrast, Hjelle, Belongia, and Nesser (1996) reported a correlation of .41
between the LOT and the ASQ CP-CN composite. Gillham, Tassoni,
Engel, DeRubeis, and Seligman (1998) found correlations of .63 and .41
between the LOT and the ASQ CP-CN composite in a sample of college
students at two assessment points. These correlations rose to .77 and .49,
respectively, after they were corrected for attenuation. Thus, across these
studies, correlations have ranged from below .20 to .77. Clearly more research
is needed that directly examines the relationship between explanatory style
and dispositional optimism.
Relationship Between Explanation and Expectation
How strong a relationship should we expect to find between explanatory
style and expectations? Several researchers have pointed out that the inferred
consequences of an event may be quite negative even when an optimistic
explanation is given (Abramson et al., 1989; Hammen & Cochran, 1981;
Zullow, 1991). For example, Abramson and colleagues (1989) gave the
example of a college student who performs poorly on the graduate record
exam. This student could attribute his poor performance to noises in the
testing room (an unstable and specific cause) while making catastrophic
predictions (e.g., Ill never be admitted to a graduate program.). Similarly,
Zullow (1991) suggested that a person who suffers permanent injury in an
automobile accident can give an optimistic explanation for the event (e.g.,
The other driver was distracted.) and still make bleak predictions about
his or her future. According to the HT, a pessimistic explanatory style is
only one of several pathways to hopelessness. Individuals who catastrophize
and blame themselves following negative events are also vulnerable to
hopelessness, even if they do not have a pessimistic explanatory style.
Interaction With Other Variables
Explanations and expectations may interact with other cognitive variables in predicting psychological well-being. For example, explanations and

60

GILLHAM ET AL

expectations should affect us the most when they concern events that are
important to us (Feather & Tiggemann, 1984; Turner & Cole, 1994). Zullow
(1991) argued that explanatory style and expectancies may interact with
rumination. The combination of an optimistic explanatory style, optimistic
expectancies, and rumination may be particularly motivating. That is, individuals may be most motivated when they attribute setbacks to unstable
specific factors and expect their efforts to pay off, but focus on the negative
consequences that they are trying to prevent. Consistent with this hypothesis, Zullow (1991) reported that optimistic rumination was associated with
vigorous campaigning in political candidates who were behind in the polls.
Zullow (1991) also indicated that motivation will be affected by the
magnitude of the difference between two types of expectancies: (a) outcome
expectancies given action and (b) outcome expectancies given inaction. A
large difference in these expectancies may be especially motivating for
individuals with an optimistic explanatory style who are optimistic about
their ability to control outcomes. These individuals may pursue goals arduously when the perceived consequence of inaction is stable and global.
Zullow (1991) illustrated this point using an environmental example. He
suggested that the belief that deforestation of the rain forest will imperil
the entire planet is more motivating than the belief that deforestation will
deplete the Amazon basins soil of nutrients, which will take years to
replenish (p. 48).

BRIDGING THE RESEARCH LITERATURES

In the remainder of this chapter, we review some of the newer research
questions within the explanatory style and dispositional optimism literatures.
Although some questions are common to both research traditions, in many
cases researchers within each tradition can benefit from considering the
questions posed by the other. Further research is needed to clarify and better
define the constructs of optimism, to understand the mechanisms by which
optimism produces its effect on physical and psychological well-being, and
to understand its origins.
Clarifying the Optimism Construct

Although explanatory style and dispositional optimism have been described and researched extensively for the past 20 years, fundamental questions about the definition of these constructs still remain. For example, are
pessimism and optimism opposites? What are the important dimensions of
optimism? What is the relationship between optimism and control?
EXPLANATORY S T n E

61

Optimism

PIS. Pessimism

In the dispositional optimism literature, optimism and pessimism are

often viewed as two poles of the same continuum. The terms optimism,
pessimism, hope, and hopelessness are often used interchangeably. The underlying assumption is that the more individuals expect positive events, the less
they expect negative events to occur. The explanatory style literature is
similarly confused. Researchers using the CP-CN composite assume that
people who give negative explanations for problematic events will also give
negative explanations for good events. This assumption is complicated with
regard to explanatory style, because pessimistic explanations for negative
and positive events reflect opposite patterns of attribution (the former are
internal, stable, and global, whereas the latter are external, unstable, and
specific). T o make matters more confusing, within the explanatory style
literature, the term optimism is used to refer to several measures of attribution,
including the CP-CN, CP, CN, and hopelessness composite scores.
Some researchers have suggested that by regarding optimism and pessimism as poles on one continuum, important distinctions between these
constructs are blurred (Chang, 1998; Chang, DZurilla, & Maydeu-Olivares,
1994; Nunn, 1996). An individual who expects his or her wishes will go
unfulfilled does not necessarily expect catastrophes to occur. Chang and
colleagues (1994) suggested using the term pessimism to refer to expectations
of negative outcomes and optimism to refer to expectations of positive outcomes. Thus, individuals can be high or low on optimism and high or low
on pessimism.
Recently, researchers have begun to explore differences between dispositional optimism and pessimism. Chang and colleagues (Chang, 1998;
Chang et al., 1994) found evidence that measures of dispositional optimism
contain two factors: one that includes positive expectancy items and a
second that includes negative expectancy items. Positive and negative factors
were correlated, however (with r s ranging from -.43 to -.60).
In the explanatory style literature, relatively little work has been conducted on the distinction between attributions for positive versus negative
events. Peterson ( 1991) noted that in several investigations, explanatory
style for positive events is weakly or not significantly correlated with explanatory style for negative events. Clearly, further research is needed that explores
the distinction between explanatory style for positive versus negative events.
As in the dispositional optimism literature, more thoughtful and precise use
of terminology is needed.
One important reason for distinguishing between expectations for good
versus bad events (optimism vs. pessimism) and explanatory style for good
versus bad events is that these constructs may have different correlates and
may be related to different outcomes. Within the explanatory style literature,

62

GILLHAM ET AL.

for example, explanatory style for negative events may predict increases in
depressive symptoms, whereas explanatory style for positive events may
predict reductions in symptoms. Consistent with this hypothesis, Needles
and Abramson (1990) found that the interaction of positive life events with
a hopeful explanatory style (stable and global explanations for positive
events) predicted recovery from depressive symptoms in college students
(see also Edelman, Ahrens, & Haaga, 1994). Similarly, Johnson, Crofton,
and Feinstein (1996) found a hopeful explanatory style interacted with
positive life events to predict symptom reduction in psychiatric inpatients
with depression.

Optimism 01s. Explanatory Style

Another obvious area of confusion results from the use of the terms
optimism and pessimism to refer to explanatory style. Peterson (1991) presented the following arguments for the use of these terms by researchers
investigating explanatory style: First, optimism and pessimism are terms
from ordinary language. Second, it is optimistic to view the causes of negative
events as external, temporary, and affecting few areas of life. Finally, explanatory style is associated with many outcomes, which are linked to pessimism,
such as depression, lowered expectancies, passivity, poor achievement, and
poor health.
In contrast, other researchers have argued that the use of optimism and
pessimism to denote causal attributions is potentially misleading, because our
intuitive notion of optimism concerns expectations (Abramson, Dykman, &
Needles, 1991; Zullow, 1991). Although optimism and pessimism may be
useful modifiers of terms such as attributions and explanatory style, we agree
with the suggestion by Abramson and colleagues (1991) that these terms
in isolation be reserved for the expectational components of the RLHT and
HT. This suggestion also applies to the terms hopeful and hopeless, which
have recently been applied to explanatory style (Abramson et al., 1989;
Needles & Abramson, 1990).
Although explanatory style and pessimism are associated with many
of the same outcomes, this does not mean we should equate the constructs.
In most studies, measures of explanatory style are only weakly to moderately
correlated with measures of dispositional optimism (Hjelle et al., 1996;
Kamen, 1989; Scheier & Carver, 1992). Explanatory style also is correlated
with self-esteem, extroversion, and neuroticism (Gillham et al., 1998) and
is associated with many of the same outcomes (Clark & Watson, 1991).
However, few researchers advocate using these terms interchangeably.

Dimensions
Most research on explanatory style uses composite scores that sum
across three dimensions (internality, stability, and globality) of explanatory
EXPLANATORY S T Y L E

63

style. There is considerable debate among researchers regarding the validity

of this practice. Correlations between the different explanatory style dimensions, particularly correlations between the internal dimension and other
dimensions, are often quite low. This raises questions about whether these
dimensions (a) reflect a single construct (explanatory style), and (b) should
be weighted equally (Carver, 1989; Perloff & Persons, 1988). In contrast
to measures of explanatory style, measures of outcome expectancies (such
as the LOT) do not assess dimensions of optimism. Instead, these measures
simply yield a composite score (or sometimes one score for optimism and
one for pessimism). Nevertheless, there may be important dimensions of
expectation that predict psychological and physical well-being. Further research is needed on the use of dimensions within the explanatory style and
dispositional optimism literatures.

The Role of Control

An important component of the RLHT and HT of depression is the
expectation that one can control events in the environment. Individuals who
feel powerless are more vulnerable to motivational deficits and depression.
Despite the importance of control to explanatory style theories, few investigations of explanatory style investigate perceptions of control. Peterson
(1991) observed that perceptions of control are usually inferred from the
causal attributions people give. Thus, when attributions for negative events
are internal, stable, and global, the event will arguably be regarded as
uncontrollable. Few studies have investigated this assumption, however.
Peterson (1991) found that the stability and globality of explanations did
not load onto the same factor with perceptions of control.
The relationship of control to dispositional optimism is also confusing.
Scheier and Carver (1993) wrote that their approach to optimism intentionally de-emphasizes the role of personal efficacy or agency. At the same time,
they propose that peoples actions are greatly influenced by their expectations
about the future (Scheier & Carver, 1992). If individuals perceive desired
outcomes to be attainable, they continue to strive toward those outcomes.
Conversely, if outcomes are perceived to be unattainable, effort is withdrawn.
We have assumed that positive expectancies cause the person to continue
to work toward the attainment of goals. Implicit in this view is the notion
that the optimistic person views the positive outcome as at least partially
contingent on continued effort (Scheier & Carver, 1992, p. 218). Thus,
the dispositional optimism construct blurs two distinct types of expectations
that are differentiated in the HT: (a) expectations of positive outcomes
(or hopeless expectancies) and (b) expectations of control (or helpless
expectancies). An optimistic score on the LOT could reflect either of two
different outlooks: (a) the belief that the world is a benevolent place or (b)

64

GILLHAM ET AL.

the belief that one possesses the skills to achieve ones goals. Research that
investigates the interrelationship between these two types of expectancies
and that investigates the relationship of each of these to psychological wellbeing is needed.
The relationship between beliefs about control and well-being is likely
to be complicated. First, the relationship between control and motivation
may not be linear. Motivation may be highest when individuals have somebut not complete-control over an outcome. Zullow (1991) argued that
the most motivating situations are those for which the perceived probability
of success is high but not 100%. In other words, challenges are motivating.
This hypothesis is consistent with writings on the flow experience (e.g.,
Csikszentmihalyi & Csikszentmihalyi, 1988), which propose that people are
highly motivated (and experience flow) when the challenges of a task
closely match (or are slightly above) the skills that they bring to it. When
the challenges exceed skills, people feel anxious. When the challenges are
too low, people are bored. Second, in some situations, a belief in control
may actually be harmful. Seligman (1993) and Weisz and colleagues (Weisz,
Rothbaum, & Blackburn, 1984) proposed that psychological health requires
the ability to accept that which cannot be controlled. Third, the perceived
controllability of past events may differ from that for future events (e.g.,
Brickman et al., 1982). Finally, there may be important individual and
cultural differences in the need for perceived control and the relationship
of perceptions of control to hopefulness about the future (Nunn, 1996;
Weisz et al., 1984).
Consistency
Several researchers have questioned whether an explanatory style exists. That is, are explanations consistent enough across time and situations to
warrant the designation style?Thesame question can be asked of dispositional
optimism. Research suggests that there is some stability in explanatory style
over time. Tiggemann and colleagues (1991) measured explanatory style in
young adults across 3 years and found a moderate correlation ( r = .44)
between CP-CN scores across this time period. Bums and Seligman (1989)
found explanatory style during late adolescence and young adulthood was
related ( r = .54) to explanatory style 50 years later. Cutrona, Russell, and
Jones (1984) found that explanatory style was not highly consistent across
situations. Abramson and colleagues ( 1989) suggested that explanatory style
may be consistent within domains (e.g., interpersonal, achievement).
Overlap With Other Psychological Constructs
Recently, researchers have become interested in the uniqueness of
optimism as a psychological construct. Watson and Clark proposed that
EXPLANATORY STYLE

65

several seemingly diverse personality and cognitive constructs actually reflect

facets of two broad underlying constructs: positive affectivity (PA) and
negative affectivity (NA) (Clark & Watson, 1991; Watson & Clark, 1984).
Thus, constructs such as neuroticism, self-esteem, optimism, and explanatory
style may correlate with each other and predict depressive symptoms
simply because they each reflect NA. In support of this view, Smith, Pope,
Rhodewalt, and Poulton (1989) found that the correlation between optimism and depressive symptoms disappeared when NA was partialled out.
In contrast, Chang, Maydeu-Olivares, and DZurilla ( 1997) found that optimism and pessimism remained significant predictors of psychological wellbeing even after controlling for PA and NA. Lucas, Diener, and Suh (1996)
found that optimism could be discriminated from negative affectivity and life
satisfaction using a multitrait-multimethod matrix analysis. More research
is needed to evaluate this hypothesis with regard to both optimism and
explanatory style.

Mechanisms
Over the past decade, the focus on the pathway between optimism or
explanatory style and well-being has increased. Three questions receiving
increasing attention are (a) How do life events interact with optimism (or
explanatory style) to affect adjustment? (b) W h y does optimism (or an
optimistic explanatory style) lead to greater well-being? And (c) Does optimism (or explanatory style) interact in important ways with other variables?

Diathesis-Stress Models

The RLHT and HT are diathesis-stress models of depression. According to these models, individuals with a pessimistic explanatory style (the
diathesis) are more likely than those with an optimistic style to become
depressed when they experience negative life events (stress). Most studies
of explanatory style have examined the link between pessimistic explanatory
style and depressive symptoms, without considering the role of negative life
events. Several researchers have argued that these studies are, therefore,
inadequate tests of the RLHT and HT (Abramson et al., 1989; Metalsky
et al., 1987; Robins & Hayes, 1995).
A few studies explicitly test the diathesis-stress component of explanatory style theories. For example, Metalsky, Halberstadt, and Abramson
(1987) studied depressive affect in students following receipt of a poor exam
grade. These researchers found that most students who received poor grades
reported immediate depressed mood, but only those students with a hopeless
(stable and global) explanatory style for negative events continued to report
depressed mood several days later. Thus, enduring depressed mood was

66

GILLHAM ET AL.

predicted by the interaction of a pessimistic explanatory style with the

experience of a negative life event. In general, studies of the diathesis-stress
model have produced mixed results. Some show a significant interaction
between negative life events and explanatory style (Hilsman & Garber,
1995; Houston, 1995), whereas others show that this interaction is not
significant (Follette & Jacobson, 1987; Hammen, Adrian, & Hiroto, 1988;
Tiggemann et al., 1991).
The diathesis-stress model also may be applicable to dispositional
optimism or pessimism. Dispositional pessimists may be particularly vulnerable to making catastrophic predictions about the future when adversity
strikes. This may further increase their hopelessness and render them vulnerable to depression. In support of this hypothesis, Dixon, Heppner, Burnett,
and Lips (1993) found the interaction between stress and hopelessness was
associated with depressive symptoms. Hopelessness was more strongly related
to depression scores under high levels of stress.

Coping
Although optimism and explanatory style are related to a variety of
psychological, behavioral, and physical outcomes, relatively little is known
about the mechanisms involved. Steps have been made recently within
both literatures to delineate the pathway from optimism or explanatory
style to outcome. A major prediction of the RLHT is that an optimistic
explanatory style enables individuals to maintain their motivation in the
face of adversity and, thus, allows them to cope more effectively. Despite
this prediction, few studies have explored the relationship between explanatory style and coping strategies. Follette and Jacobson (1987) found that
contrary to the RLHT students who made internal, stable, and global attributions for poor exam performance reported more plans to study for the next
examination than did their optimistic peers.
More research exists on the relationship between dispositional optimism and coping. Dispositional optimism correlates positively with problemfocused coping, the positive reinterpretation of a problem, and the attempt
to accept the reality of situations that are perceived to be uncontrollable.
In addition, optimism correlates negatively with the use of denial and the
attempt to distance oneself from problems (Amirkhan, Risinger, & Swickert,
1995; Aspinwall & Taylor, 1992; Scheier & Carver, 1993).
Origins of Optimism and Explanatory Style
Little is known about the origins of explanatory style or dispositional
optimism. Within the explanatory style literature, a variety of environmental
causes have been suggested. Negative life events, particularly the experience
EXPLANATORY STYLE

67

of abuse (Kaufman, 1991; Gold, 1986) or parental conflict (NolenHoeksema, Girgus, & Seligman, 1986), are linked to a pessimistic explanatory style in some studies. Seligman, Reivich, Jaycox, and Gillham (1995)
proposed that parents can influence their childrens explanatory style
through modeling and through the attributions that they make for events
in their childrens lives. A child who is continually criticized as stupid or
lazy following mistakes may internalize these internal, stable, and global
explanations in the future. Thus far, research on the relationship between
parent and child explanatory style has yielded conflicting results (for a recent
review, see Joiner & Wagner, 199513). Similar environmental theories have
been proposed for dispositional optimism. For example, Scheier and Carver
( 1993)suggested that the development of dispositional optimism in children
is influenced by experience with success and failure, optimism and pessimism
modeled by parents, and coping strategies taught by parents.
There is some evidence for a genetic component of dispositional optimism and explanatory style. Schulman, Keith, and Seligman (1991) found
that explanatory style was significantly correlated in monozygotic twins. In
contrast, the correlation in dizygotic twins was not significant. Plomin and
colleagues (1992) estimated the heritability of optimism and pessimism to
be about 25%.
Understanding the origins of optimism and explanatory style is one
of the most interesting and potentially valuable goals for future research.
There is growing evidence that depressive symptoms, anxiety, and perhaps
even physical health problems can be prevented with interventions that
focus on improving explanatory style (Gillham et al., 1995; Seligman et
al., 1995;Seligman et al., 1999).Knowledge about the origins of explanatory
style will enable clinicians to develop even more powerful intervention
programs. Eventually, such knowledge may lead to techniques that prevent
a pessimistic explanatory style from developing in children.

CONCLUSION
Explanatory style is related to a variety of psychological and physical
health indices, including academic achievement, depression, and physical
illness. Although findings in the explanatory style literature parallel those
from the dispositional optimism literature, these two constructs have largely
been studied in isolation from each other. We know surprisingly little about
the relationship between explanations and expectations. Studies that bridge
these separate research traditions are sorely needed. We have learned a
tremendous amount about optimism and explanatory style over the past 20
years, but important questions remain. Optimism and explanatory style must
be more precisely defined and differentiated from each other and from other

68

GILLHAM ET AL.

constructs. The mechanisms through which optimism and explanatory styles

affect well-being need to be identified. The sources of optimism and explanatory style need to be discovered. These questions are particularly fascinating
because of their practical value. If optimism and explanatory style are causally
related to well-being, the answers to these questions may enable clinicians
to improve the quality of life for many people.

REFERENCES
Abramson, L. Y., Alloy, L. B., Hogan, M. E., Whitehouse, W. G., Cornette, M.,
Akhavan, S., & Chiara, A. (1998). Suicidality and cognitive vulnerability to
depression among college students: A prospective study. Journal of Adolescence,
21, 157-171.
Abramson, L. Y., Alloy, L. B., & Metalsky, G. I. (1989). Hopelessness depression:
A theory-based subtype of depression. Psychological Review, 96, 358-372.
Abramson, L. Y., Alloy, L. B., & Metalsky, G. I. (1998). The Cognitive Style
Questionnaire: A measure of the vulnerability featured in the hopelessness
theory of depression. Unpublished manuscript. University of WisconsinMadison.
Abramson, L. Y., Dykman, B. M., & Needles, D. J. (1991). Attributional style and
theory: Let no one tear them asunder. Psychological Inquiry, 2, 11-49.
Abramson, L. Y., Seligman, M. E. P., & Teasdale, J. D. (1978). Learned helplessness
in humans: Critique and reformulation. Journal of Abnormal Psychology, 87,
49-74.
Amirkhan, J. H., Risinger, R. T., & Swickert, R. J. (1995). Extraversion: A hidden
personality factor in coping. Journal of Personality, 63, 189-212.
Aspinwall, L. G., &Taylor, S. E. (1992). Modeling cognitive adaptation: A longitudinal investigation of the impact of individual differences and coping on
college adjustment and performance.Journal of Personality and Social Psychology,
63, 989-1003.
Bennet, D. S., & Bates, J. E. (1995). Prospective models of depressive symptoms
in early adolescence. Attributional style, stress and support. Journal of Early
Adolescence, 15, 299-315.
Brickman, P., Rabinowitz, V. C., Kaurza, Jr., J., Coates, D., Cohn, E., & Kidder,
L. (1982). Models of helping and coping. American Psychologist, 37, 368-384.
Buchanan, G. M. (1995). Explanatory style and coronary heart disease. In G. M.
Buchanan & M. E. P. Seligman (Eds.), Explanatory style (pp. 225-232). Hillsdale, NJ: Erlbaum.
Buchanan, G. M., Gardenswartz, C. A. R., & Seligman, M. E. P. (1999). Physical
health following a cognitive-behavioralintervention. Prevention and Treatment,
2, 10, joumals.apa.org.
EXPLANATORY STYLE

69

Buchanan, G. M., & Seligman, M. E. P. (1995). Explanatory style. Hillsdale, NJ:

Erlbaum.
Bums, M. O., & Seligman, M. E. P. (1989). Explanatory style across the life span:
Evidence for stability over 52 years. Journal of Personality and Social Psychology,
56, 471-477.
Carver, C. S. (1989). How should multi-faceted personality constructs be tested?
Issues illustrated by self-monitoring, attributional style, and hardiness. Journal
of Personality and Social Psychology, 36, 1501-151 1.
Carver, C. S., & Gaines, J. G. (1987). Optimism, pessimism, and postpartum
depression. Cognitive Therapy and Research, 1 1 , 449-462.
Chang, E. C. (1998). Distinguishing between optimism and pessimism: A second
look at the optimism-neuroticism hypothesis. In R. R. Hoffman, M. F. Sherrik, & J. S. Warm (Eds.), Viewing psychology us a whole: The integrative science
of William N.Dember (pp. 415-432). Washington, DC: American Psychological Association.
Chang, E. C., DZurilla, T. J., & Maydeu-Olivares, A. (1994). Assessing the dimensionality of optimism and pessimism using a multimeasure approach. Cognitive
Therapy and Research, 18, 143-160.
Chang, E. C., Maydeu-Olivares, A., & DZurilla, T. J. (1997). Optimism and
pessimism as partially independent constructs: Relationship to positive and
negative affectivity and psychological well-being. Personality and Individual
Differences, 23, 433-440.
Clark, L. A., & Watson, D. (1991). General affective dispositions in physical and
psychological health. In C. R. Snyder & D. R. Forsyth (Eds.), Handbook of
sociuland clinicalpsychology:The health perspective (pp. 221-245). New York: Pergamon.
Csikszentmihalyi, M., & Csikszentmihalyi, I. S. ( 1988). Optimal experience: Psychological studies offlow in consciousness. New York: Cambridge University Press.
Cutrona, C. E. (1983). Causal attributions and perinatal depression. Journal of
Abnormal Psychology, 92, 161-172.
Cutrona, C. E., Russell, D., & Jones, R. D., (1984). Cross-situational consistency
in causal attributions: Does attributional style exist?Journal of Personality and
Social Psychology, 47, 1043-1058.
DeRubeis, R. J., & Hollon, S. D. (1995). Explanatory style in the treatment of
depression. In G. M. Buchanan & M. E. P. Seligman (Eds.), Explanatory style
(pp. 99-1 12). Hillsdale, NJ: Erlbaum.
Dixon, W. A., Heppner, P. P., Bumett, J. W., & Lips, B. J. (1993). Hopelessness
and stress: Evidence for an interactive model of depression. Cognitive Therapy
and Research, 17, 39-52.
Dweck, C. S. (1975). The role of expectations and attributions in the alleviation
of learned helplessness.Journal of Personality and Socid Psychology, 3 1 ,674-685.
Edelman, R. E., Ahrens, A. H., & Haaga, D. A. F. (1994). Inferences about
the self, attributions, and overgeneralization as predictors of recovery from
dysphoria. Cognitive Therapy and Research, 18, 551-566.

70

GILLHAM ET AL.

Feather, N. T., & Tiggemann, M. (1984). A balance measure of attributional style.

Australian Journal of Psychology, 36, 267-283.
Fincham, F. D., & Bradbury, T. N. (1993). Marital satisfaction, depression, and
attributions: A longitudinal analysis.Journa1of Personality and Social Psychology,
64, 442-452.
Follette, V. M., &Jacobson, N. S. (1987). Importance of attributions as a predictor
of how people cope with failure. Journal of Personality and Social Psychology,
52, 1205-1211.
Gillham, J. E., Reivich, K. J., Jaycox, L. H., & Seligman,M. E. P. (1995). Prevention
of depressive symptoms in schoolchildren: Two year follow-up. Psychological
Science, 6, 343-351.
Gillham, J. E., Tassoni, C. J., Engel, R. A., DeRubeis, R. J., & Seligman, M. E. P.
( 1998). The relationship of explanatory style to other depression relevant
constructs. Unpublished manuscript, University of Pennsylvania, Philadelphia.
Gladstone, T. R. G., & Kaslow, N. J. (1995). Depression and attributions in children
and adolescents: A meta-analytic review. J o u m l of Abnormal Child Psychology,
23, 597-606.
Gold, E. R. (1986). Long-term effects of sexual victimization in childhood: An
attributional approach. Journal of Consulting and Clinical Psychology, 54, 47 1475.
Golin, S., Sweeney, P. D., & Schaeffer, D. E. (1981). The causality of causal
attributions in depression: A cross-lagged panel analysis. Journal of Abnormal
Psychology, 90, 14-22.
Hammen, C., & Cochran, S. (1981).Cognitive correlates of life stress and depression
in college students. Journal of Abnormal Psychology, 90, 23-27.
Hammen, C. L., Adrian, C., & Hiroto, D. (1988). A longitudinal test of the
attributional vulnerability model in children at risk for depression. British
Journal of Clinical Psychology, 27, 37-46.
Hilsman, R., & Garber, J. (1995). A test of the cognitive diathesis-stress model
of depression in children: Academic stressors, attributional style, perceived
competence, and control. Journal of Personality and Social Psychology, 69, 370380.
Hjelle, L., Belongia, C., & Nesser, J. (1996). Psychometric properties of the Life
Orientation Test and Attributional Style Questionnaire. Psychological Reports,
78, 507-515.
Houston, D. M. (1995). Vulnerability to depressive mood reactions: Retesting the
hopelessness model of depression. British Journal of Social Psychology, 34, 293302.
Ilardi, S. S., Craighead, E. W., & Evans, D. D. (1997). Modeling relapse in unipolar
depression: The effects of dysfunctional cognitions and personality disorders.
Journal of Consulting and Clinical Psychology, 65, 381-391.
Johnson, J. G., Crofton, A., & Feinstein, S. B. (1996). Enhancing attributional style
and positive events predict increased hopefulness among depressed psychiatric
inpatients. Motivation and Emotion, 20, 285-297.
EXPLANATORY STYLE

71

Joiner, T. E., & Wagner, K. D. (1995a). Attributional style and depression in

children and adolescents: A meta-analytic review. Clinical Psychology Review,
15, 777-798.
Joiner, T. E., & Wagner, K. D. (1995b). Parental, child-centered attributions
and outcome: A meta-analytic review with conceptual and methodological
implications. Journal of Abnormal Child Psychology, 24, 37-52.
Kamen, L. P. ( 1989). Learned helplessness, cognitive dissonance, and cell-mediated
immunity. Unpublished doctoral dissertation, University of Pennsylvania,
Philadelphia.
Kaslow, N. J., Tannenbaum, R. L., & Seligman, M. E. P. (1978). The KASTAN:
A childrens attributional style questionnaire. Unpublished manuscript, University of Pennsylvania, Philadelphia.
Kaufman, J. (1991). Depressive disorders in maltreated children. Journal of the
American Academy of Child and Adolescent Psychiatry, 30, 257-265.
Lucas, R. E., Diener, E., & Suh, E. (1996). Discriminant validity of well-being
measures. Journal of Personality and Social Psychology, 71 , 616-628.
Manly, P. C., McMahon, R. J., Bradley, C. F., & Davidson, P. 0. (1982). Depressive
attributional style and depression following child birth. Journal of Abnormal
Psychology, 91, 245-254.
Metalsky, G. I., Halberstadt, L. J., & Abramson, L. Y. (1987). Vulnerability to
depressive mood reactions: Toward a more powerful test of the diathesis-stress
and causal mediation components of the reformulated theory of depression.
Journal of Personality and Social Psychology, 52, 386-393.
Metalsky, G. I., &Joiner, T. E. (1992). Vulnerability to depressive symptomatology:
A prospective test of the diathesis-stress and causal mediation components of
the hopelessness theory of depression. Journal of Personality and Social Psychology, 63, 667-675.
Metalsky, G. I., Joiner, T. E., Hardin, T. S., & Abramson, L. Y. (1993). Depressive
reactions to failure in a naturalistic setting: A test of the hopelessness and selfesteem theories of depression. Journal of Abnormal Psychology, 102, 101-109.
Needles, D. J., & Abramson, L. Y. (1990). Positive life events, attributional style, and
hopefulness: Testing a model of recovery from depression. Journal of Abnormal
Psychology, 99, 156-165.
Nolen-Hoeksema, S., Girgus, J. S., & Seligman, M. E. P. (1986). Depression in
children of families in turmoil. Unpublished manuscript, University of Pennsylvania, Philadelphia.
Nunn, K. P. (1996). Personal hopefulness: A conceptual review of the relevance
to the perceived future of psychiatry. British Journal of Medical Psychology,
69, 227-245.
Perloff, J. M., & Persons, J. B. (1988). Biases resulting from the use of indexes:
An application to attributional style and depression. Psychological Bulletin,
103, 95-104.

72

GILLHAM ET AL.

Peterson, C. (1988). Explanatory style as a risk factor for illness. Cognitive Therapy
and Research, 12, 117-130.
Peterson, C. (1991). The meaning and measurement of explanatory style. Psychological Inquiry, 2, 1-10.
Peterson, C., Bettes, B. A., & Seligman, M. E. P. (1985). Depressive symptoms
and unprompted causal attributions: Content analysis. Behavior Research and
Therapy, 23, 379-382.
Peterson, C., & Bossio, L. M. (1991). Health and optimism. New York: Free Press.
Peterson, C., & De Avila, M. (1995). Optimistic explanatory style and the perception of health problems. ]ournu1 of Clinical Psychology, 51, 128-132.
Peterson, C., & Seligman, M. E. P. (1984). Causal explanations as a risk factor for
depression: Theory and evidence. Psychological Review, 41 , 253-259.
Peterson, C., Seligman, M. E. P., & Vaillant, G. E. (1988). Pessimistic explanatory
style is a risk factor for physical illness: A thirty-five year longitudinal study.
Journal of Personality and Social Psychology, 55, 23-27.
Peterson, C., Semmel, A., von Baeyer, C., Abramson, L. Y., Metalsky, G. I., &
Seligman, M. E. P. (1982). The Attributional Style Questionnaire. Cognitive
Therapy and Research, 6, 287-299.
Peterson, C., & Villanova, P. (1988). A n Expanded Attributional Style Questionnaire. Journal of Abnormal Psychology, 97, 87-89.
Plomin, R., Scheier, M. F., Bergeman, C. S., Pederson, N. L., Nesselroade, J., &
McClearn, G. (1992). Optimism, pessimism and mental health: A twin/adoption analysis. Personality and Zndividwll Differences, 13, 921-930.
Reivich, K. (1995). The measurement of explanatory style. In G. M. Buchanan &
M. E. P. Seligman (Eds.), Explanatory styk (pp. 21-48). Hillsdale, NJ: Erlbaum.
Rettew, D., & Reivich, K. (1995). Sports and explanatory style. In G. M. Buchanan
& M. E. P. Seligman (Eds.), Explanatory styk (pp. 173-186). Hillsdale, NJ:
Erlbaum.
Robins, C. J., & Hayes, A. M. (1995). The role of causal attributions in the
prediction of depression. In G. M. Buchanan & M. E. P. Seligman (Eds.),
Explanatory styk (pp. 71-98). Hillsdale, NJ: Erlbaum.
Satterfield, J. M, Monahan, J., & Seligman, M. E. P. (1997). Law school performance
predicted by explanatory style. Behavioral Sciences and the Law, 15, 95-105.
Satterfield, J. M., & Seligman, M. E. P. (1994). Military aggression and risk predicted
by explanatory style. Psychological Science, 5, 77-82.
Scheier, M. F., & Carver, C. S. (1985). Optimism, coping and health: Assessment
and implications of generalized outcome expectancies. Health Psychology, 4 ,
2 19-247.
Scheier, M. F., & Carver, C. S. (1988). A model of behavioral self-regulation:
Translating intention into action. In L. Berkowitz (Ed.), Advances in experimental social psychology (Vol. 21, pp. 303-346). San Diego, CA: Academic Press.
EXPLANATORY STYLE

73

Scheier, M. F., & Carver, C. S. (1992). Effects of optimism on psychological

and physical well-being: Theoretical overview and empirical update. Cognitive
Therapy and Research, 16, 201-228.
Scheier, M. F., & Carver, C. S. (1993). On the power of positive thinking: The
benefits of being optimistic. Current Directions in Psychological Science, 2,26-30.
Schulman, P. (1995). Explanatory style and achievement in school and work. In
G. M. Buchanan & M. E. P. Seligman (Eds.), Explanatory style (pp. 159-171).
Hillsdale, NJ: Erlbaum.
Schulman, P., Keith, D., & Seligman, M. E. P. (1991). Is optimism heritable? A
study of twins. Behavior Research and Therapy, 31, 569-574.
Seligman, M. E. P. (1991). Learned optimism. New York: Knopf.
Seligman, M. E. P. (1993). What you can change and what you cant. New York: Knopf.
Seligman, M. E. P., Abramson, L. Y., Semmel, A., & von Baeyer, C. (1979).
Depressive attributional style. Journal of Abnormal Psychology, 88, 242-247.
Seligman, M. E. P., Castellon, C., Cacciola, J., Schulman, P., Luborsky, L., Ollove,
M., & Downing, R. (1988). Explanatory style change during cognitive therapy
for unipolar depression. Journal of Abnormal Psychology, 97, 13-18.
Seligman, M. E. P., Nolen-Hoeksema, S., Thornton, K. M., & Thornton, N. (1990).
Explanatory style as a mechanism of disappointing athletic performance. Psychological Science, 1 , 143-146.
Seligman, M. E. P., Peterson, C., Kaslow, N. J., Tannenbaum, R. L., Alloy, L. B.,
& Abramson, L. Y. (1984). Attributional style and depressive symptoms among
children. Journal of Abnormal Psychology, 93, 235-238.
Seligman, M. E. P., Reivich, K. J., Jaycox, L. H., & Gillham, J. (1995). The optimistic
child. New York: Houghton-Mifflin.
Seligman, M. E. P., Schulman, P., DeRubeis, R. J., & Hollon, S. D. (1999). The
prevention of depression and anxiety. Prevention B Treatment, 2,8, joumals.apa
.org.
Smith, T. W., Pope, M. K., Rhodewalt, F., & Poulton, J. L. (1989). Optimism,
neuroticism, coping, and symptom reports: An alternative interpretation of the
Life Orientation Test. Journal of Personality and Social Psychology, 56, 640-648.
Tiggemann, M., Winefield, A. H., Winefield, H. R., & Goldney, R. D. (1991).
The prediction of psychological distress from attributional style: A test of the
hopelessness model of depression. AustralianJournal of Psychology, 43, 125-127.
Turner, J. E., &Cole, D. A. (1994). Developmental differences in cognitive diathesis
for child depression. Journal of Abnormal Child Psychology, 22, 15-32.
Watson, D., & Clark, L. A. (1984). Negative affectivity: The disposition to experience aversive emotional states. Psychological Bulletin, 96, 465-490.
Weissman, A. N. (1979). The Dysfunctional Attitudes Scale: A validation study.
Unpublished doctoral dissertation. University of Pennsylvania, Philadelphia.
Weisz, J. R., Rothbaum, F. M., & Blackburn, T. C. (1984). Standing out and
standing in: The psychology of control in American and Japan. American
Psychologist, 39, 955-969.

74

GILLHAM ET AL.

Zullow, H. M. ( 1991). Explanations and expectations: Understanding the doing

side of optimism. Psychological Inquiry, 2, 45-49.
Zullow, H. M. (1995). Pessimistic rumination in American politics and society. In
G. M. Buchanan & M. E. P. Seligman (Eds.), Exphnutory style (pp. 21-48).
Hillsdale, N]: Erlbaum.
Zullow, H. M., & Seligman, M. E. P. (1990). Pessimistic rumination predicts defeat
of presidential candidates, 1900 to 1984. Psychological Inquiry, I , 52-61.

EXPLANATORY S T n E

75

DEFENSIVE PESSIMISM, OPTIMISM,

AND PESSIMISM
JULIE K. NOREM

The term defensive pessimism was coined in the mid-1980s by Nancy

Cantor and her students (Norem & Cantor, 1986a; Norem & Cantor,
198613) and refers to a cognitive strategy in which individuals set low
expectations for an upcoming performance, despite having done well in
similar situations in the past. Theoretically, setting low expectations helps
cushion the potential blow of failure. People using the strategy then
mentally rehearse or extensively reflect on what might happen-with special
attention paid to potential problems that they might encounter. They play
through various mental scenarios, including all of the possible things that
might go wrong, and work hard to prepare for the upcoming performance
or situation. Typically, individuals who use this strategy initially feel anxious
and out of control. The strategy seems to help them gain a feeling of control
and to harness their anxiety as motivation. Individuals using the strategy
perform quite well, contrary to their dire predictions.
A prototypical example of defensive pessimism would be an expert
public speaker who is anxious before every speech, who repeatedly asserts

77

that he or she is going to embarrass himself or herself terribly in the next

talk, despite a multitude of previously successful talks. This individual can
enumerate in detail all of the potential pitfalls that might mar the performance, from the forgotten note cards to the spilled water pitcher on the
podium, to the humiliating trip over the microphone cord. Having envisioned the pending disasters in such vivid detail, however, the anxious
speaker can also plan carefully to prevent them: He or she requests two
copies of the notes for the speech, that the water pitcher be moved out of
range, and that electrical tape secure the microphone cord to the floor.
Although these precautions cannot ensure rousing applause at the end of
the talk, they can at least alleviate anxiety about the worst possible outcomes.
Defensive pessimism has a reputation as a gadfly in the world of optimism and pessimism research, because investigation of the construct started
with the hypothesis that this kind of pessimism served positive functions
for those who used it. This position seems contrary to most research, in
which pessimism is almost invariably related to more negative outcomes
(see Isaacowitz & Seligman, 1998; chap. 16, this volume). In this chapter,
I compare and contrast the construct of defensive pessimism with other
constructs of optimism and pessimism, review past and ongoing defensive
pessimism research, and discuss remaining research questions and promising
future directions. I conclude with consideration of the implications defensive
pessimism has for how people think about pessimism and optimism.

STRATEGIES AS UNITS OF ANALYSIS

The construct of defensive pessimism as a cognitive strategy derives
from a social-cognitive approach to the study of personality and behavior.
Cantor and Kihlstroms ( 1987) social intelligence theory discusses strategies
as units of analysis that describe how people go about pursuing their important goals, or life tasks. Strategies describe coherent patterns of expectations,
appraisals, planning, effort, and retrospection as individuals pursue personally
relevant goals (Norem, 1989).
There are several implications that follow from the description of
defensive pessimism as a strategy. First, strategies are developed in the
context of particular goals, which implies that the same individual may use
different strategies over time and across situations as he or she pursues
different goals. Thus, although strategies describe important individual differences, they are not assumed to be stable over time and consistent across
situations in the same way as traditional traits, motives, or styles. Theoretically they are potentially malleable. Individuals may develop a repertoire
of strategies for different goals or situations, and they may change strategies

78

JULIE K. NOREM

as their appraisals of relevant situations or goals change. This implies that

there is considerable domain specificity in the use of these strategies. In
other words, an individual may use defensive pessimism in one type of
situation, strategic optimism in another, and some other strategy in a third.
Second, strategies describe processes that unfold over time. For defensive pessimism, for example, the deployment of the strategy begins with
setting low expectations in response to anxiety. The strategy then continues
as the individual plays through different scenarios. This implies that components of a strategy may have different effects separately than they do together.
For example, pessimism by itself and increased reflectivity by itself tend to
increase anxiety, but for people accustomed to using defensive pessimism,
they tend to decrease anxiety. Furthermore, an individual's specific vulnerability to disruption of the strategy will vary according to when that disruption
occurs during its use.
It is important to make clear what is not implied by the use of the term
cognitive strategy. We do not assume that people are necessarily conscious of
the strategies they use, although they may be. Just as with other defensive
or self-protective strategies, defensive pessimism and strategic optimism may
be used without awareness of the process, the motivation, or the consequences. People ma,y also be aware that they use a particular strategy without
necessarily being aware of when or why they are using that strategy. Thus,
for example, many people recognize themselves in the description of someone
using defensive pessimism but maintain that using the strategy was not what
they were doing in the particular laboratory performance or situation under
discussion, even though data or observation suggests otherwise. Other times,
individuals will recognize in retrospect that they used a strategy, but they
were not aware of using it at the time.
Describing processes as strategic means only that they are done in the
context or in reference to a particular problem or goal. It does not mean
that a particular strategy will necessarily be effective or lead to success
(Norem, 1998). People may use ineffective strategies, or they may use
strategies inappropriately. Few strategies are inherently effective, independent of when, where, and how they are used. Strategy effectiveness needs
to be evaluated in terms of the costs and benefits most likely to be associated
with particular strategies, with sensitivity to needs of the individual using
the strategy, and by the appropriateness of the strategy to specific goals
and situations.
Strategies are cognitive in that they capture the way that people think
about or construe situations, but they are not cold in the sense of connoting rational processes that are untouched by affect or motivation. Quite the
contrary, their purpose is to regulate emotion and behavior, and they
have emotional and motivational antecedents and consequences. Strategies

DEFENSlVE PESSIMISM

79

emphasize what Adler (1935/1979) called the creative power of the individual (p. 67) in adapting to ones own temperament and the influence of
the environment.
This chapter focuses on defensive pessimism and frequently contrasts
it with strategic optimism. Strategic optimism refers to a strategy used by
individuals who do not typically feel anxious in a particular domain. Instead,
they feel in control of their own outcomes, and they set high expectations
that are generally congruent with their perceptions of themselves and their
past experiences. People using this strategy actively avoid thinking about
negative possible outcomes. Indeed, they generally avoid thinking about
upcoming tasks, although they do tend to do what is necessary to prepare
(e.g., they will study for a test, but they will avoid thinking about what
grade they might get, what trick questions the test might contain, and
so forth). Strategic optimism offers an informative contrast to defensive
pessimism, as the research discussed below illustrates.
Focusing on these two strategies, however, does not assume that they
are the only two strategies that people use, or that everyone is either a
defensive pessimist or a strategic optimist. In the research conducted with
college student samples, up to 40% of the respondents may be classified as
aschematic with respect to defensive pessimism. That means that they do
not appear to use either defensive pessimism or strategic optimism consistently within the domain or situation being assessed. There are potentially
any number of strategies that people may use in different situations. Among
the most studied by psychologists are various self-handicapping strategies,
which can be distinguished from both defensive pessimism and strategic
optimism (Martin, 1999). Self-handicapping involves adopting or claiming
an impediment to performance (e.g., drinking too much at a party, not
practicing before a big game) to obscure attributions in the case of failure
and thereby protect ones sense of self-esteem (see Berglas, 1985; Higgins
& Harris, 1988). It is important to note that although both those who use
self-handicapping and those who use defensive pessimism generally report
high anxiety, defensive pessimists do not withdraw effort or adopt handicaps
in response to their anxiety. Instead, they defend themselves from the
negative impact of failure by lowering their expectations so they are neither
surprised nor as severely disappointed if failure occurs. They then cope
with or harness their anxiety through extensive reflection about possible
outcomes and plans to avoid negative outcomes and to approach positive
outcomes. This contrasts sharply with the effort withdrawal or self-sabotage
characteristic of self-handicapping.
Defensive pessimism and strategic optimism, construed as strategies,
can thus be differentiated from other constructs of optimism and pessimism
by their connection to goals, their domain specificity, and their temporal
frame. Strategic optimism is perhaps most closely related to the kinds of

80

JULIE K. NOREM

positive illusions described by Taylor and Brown (1988). Indeed, early

research referred to illusory glow optimism instead of strategic optimism,
and demonstrated that individuals using this strategy would reliably resort to
self-esteem-protecting attributions after failure, whereas defensive pessimists
seemed to cushion the impact of failure by way of their protective expectations (Norem & Cantor, 1986a, 198613). In contrast, however, to Taylor
and Browns emphasis on the adaptive advantages of positive illusions, work
on defensive pessimism and strategic optimism has focused on the costs and
benefits of each strategy, and on how each strategy makes sense given the
psychological situation faced by those using it, that is, by emphasizing
copers as individuals, as opposed to coping as an abstract process.
Defensive pessimism is also different from the pessimistic attributional
style described by Seligman and his colleagues (see chap. 3 in this volume).
Defensive pessimism does not correlate strongly with an internal, global,
and stable attributional style, nor do defensive pessimists typically make
different attributions for success and failure. Defensive pessimism describes
a strategy used to prepare for stressful events, not a reaction after those
events, and thus is further differentiated from attributional style. Showers
and Ruben (1990) found that defensive pessimists did not use the avoidant
coping methods found among people with depressive disorders, nor did their
anxiety and rumination persist after stressful events. Similarly, Ronan and
his colleagues have shown that individuals using defensive pessimism use
more effective, confrontive problem solving than do mildly depressed individuals (Cuddihy & Ronan, 1990; Hammontree & Ronan, 1992).
Defensive pessimism and strategic optimism are also different from the
trait optimism and pessimism explored by Carver and Scheier (see chap. 2
in this volume). Trait optimism describes a stable tendency to hold positive
expectations and should generally influence expectation setting consistently
across situations and over time. As already noted, strategic optimism and
defensive pessimism are domain specific and are hypothesized to be relatively
malleable. They also describe specific processes that include both expectation
setting and subsequent cognitions. The description of the strategies also
includes an explicit statement of the function of those processes.
MEASUREMENT OF DEFENSIVE PESSIMISM AND

STRATEGIC OPTIMISM

Early attempts to measure defensive pessimism relied on a 9-item, facevalid questionnaire-the Optimism-Pessimism Prescreening Questionnaire
(OPPQ; Norem & Cantor, 1986a). When the items were generated, the
research team focused on elaborating the description of defensive pessimism.
Six of the items on the early questionnaire thus referred to the two hypoDEFENSlVE PESSIMISM

81

thesized components of defensive pessimism: pessimistic expectations and

negative thinking, and their presumed opposites. That is, there were two
questions about expectations intended to measure pessimistic and optimistic
expectations within whatever domain was studied and two questions each
about negative and positive thoughts and feelings. There were also two
questions that referred to feelings after a performance, which reflected the
hypothesis that defensive pessimists may feel more relief than satisfaction
after having done well. In other words, although their defensive expectations
may cushion the impact of failure, defensive pessimists may also water
down enjoyment of success (Showers, 1992).
Our initial theorizing about defensive pessimism focused on the negative thinking that we had observed among people using defensive pessimism.
We somewhat blithely assumed that, just as optimistic expectations seemed
the opposite of pessimistic expectations, thinking about positive outcomes
was the natural opposite of thinking about negative outcomes. For example,
one question stated: I often think about what it will be like if I do poorly
in an academic situation, and another stated I often think about what it
will be like if I do very well in an academic situation. Our initial scoring
of the questionnaire (which was used to categorize prescreened research
participants as defensive pessimists or optimists for subsequent experiments)
was theoretically, not empirically, driven: We considered thinking about
doing poorly to be a defensive pessimist item and thinking about doing well
to be an optimist item.
Subsequent research has shown that this initial conception of defensive
pessimism was mistaken in that the people who tend to reflect extensively
about possible negative outcomes also tend to reflect extensively about
possible positive outcomes. In contrast, people who report that they do not
tend to think about possible negative outcomes do not tend to think about
possible positive outcomes either. In other words, thinking about positive
outcomes correlates positively with thinking about negative outcomes, and
both are part of the thinking-through process by which defensive pessimists
increase their feelings of control and decrease their anxiety. The current
version of the Defensive Pessimism Questionnaire (DPQ) contains several
items designed to index this thinking-through process, as well as items
designed to measure pessimism (see Exhibit 4.1). Factor analyses of the scale
show that although all the items load satisfactorily on one major unrotated
factor, oblique rotation also produces two correlated factors that I have
labeled Refkctivity and Pessimism. Exhibit 4.1 indicates which items load
on the Reflectivity factor and which items load on the Pessimism factor (it
also includes filler and experimental items).
In current research, I continue to use a single defensive pessimism
score computed by summing both the pessimism and reflectivity items (with
appropriate reverse scoring). In addition, however, I also compute separate

82

JULIE K . NOREM

Exhibit 4.1.
The Revised Defensive Pessimism Questionnaire
When you answer the following questions, please think about how you
prepare for and think about (academic/social) situations.a Each of the
statements below describes how people sometimes think or feel about these
kinds of situations. In the blank space beside each statement, please indicate
how true it is of you, in (academic/social) situations.

1. I go into these situations expecting the worst, even though I know I will
probably do OK.(PESS)
2. I generally go into these situations with positive expectations about how
I will do. (PESS-R)
3. Ive generally done pretty well in these situations in the past.b
4. I carefully consider all possible outcomes before these situations.

(REFL)
5. When I do well in these situations, I often feel really happy. (Filler)
6. I often worry, in these situations, that I wont be able to carry through
my intentions. (PESS)
7. I often think about how I will feel if I do very poorly in these situations.
(REFL)
8. I often think about how I will feel if I do very well in these situations.

(REFL)
9. When I do well in these situations, it is usually because I didnt get too
worried about it beforehand. (Filler)
10. I often try to figure out how likely it is that I will do very poorly in these
situations. (REFL)
11. Im careful not to become overconfident in these situations.
(Experimental item)
12. I spend a lot of time planning when one of these situations is coming
up. (REFL)
13. When working with others in these situations, I often worry that they will
control things or interfere with my plans. (Experimental item)
14. I often try to figure out how likely it is that I will do very well in these
situations. (REFL)
15. In these situations, sometimes I worry more about looking like a fool
than doing really well. (PESS)
16. Prior to these situations, I avoid thinking about possible bad outcomes.
(REFL-R)
17. Considering what can go wrong in academic situations helps me to
prepare. (REFL)
Note. PESS = Item loads on Pessimism factor; REFL = Item loads on Reflectivity factor; R = revenescored item.
The domain of interest should be specified here.
This item is included to differentiate between those who are realistically pessimistic and those who are defensively pessimistic, on the assumption that those who report having done very badly in the past are realistic when they anticipate doing badly in the future. In college student samples, typically fewer than 20% of
respondents rate themselves below 5 on this item.

DEFENSIVE PESSIMISM

83

pessimism and reflectivity scores for further exploration of the roles of

those two processes. Recently, Martin ( 1999) found that expectations and
reflectivity show somewhat different relations to other variables over time,
and it will be important to continue to examine how these processes work
separately and in conjunction.
The revised version of the scale correlates at r = .65 with the original
OPPQ and has a higher reliability (Cronbach's alpha = .78).The Reflectivity
and Pessimism subscales have comparable reliabilities (average Cronbach's
alphas = .74). In a recent longitudinal study of women at Wellesley College,
the DPQ showed a 3-year test-retest reliability of r = .55 (N = 67) from
the first year to the senior year of college.
Much of the research using the DPQ has been experimental, and
it has shown good predictive validity when used to prescreen research
participants. For prescreening purposes, those who score in the upper tertile
or quartile (depending on sample size and particular distributions) of the
DPQ are classified as defensive pessimists, those in the lower tertile or quartile
are classified as strategic optimists, and those in the middle of the distribution
are considered achematic with respect to these two strategies. When using
the scale for prescreening, Item 3 , which asks about previous performance,
has been used to distinguish realistic pessimists from defensive pessimists.
The assumption has been that those people who have done badly in the
past are realistic when they predict that they will do badly in the future,
whereas those people who report having done well in the past are being
defensive if they predict that they will do poorly in the future. This is also
congruent with theorizing that defensive pessimists do not misinterpret
objective success, but that their strategy follows from their subjective experience, an experience that includes considerable anxiety and uncertainty and
that is likely to have a negative influence on construction of future situations.
This item also allows some discrimination of those who appear unrealistically
optimistic, that is, those who report having done badly in the past but
nevertheless expect to do well in the future. In most college samples fewer
than 20% of the participants responding rate themselves lower than a score
of 5 on this item. Both optimistic and pessimistic individuals who score low
on Item 3 tend to have lower grade point averages and tend to perform
more poorly on experimental tasks than do those who score higher.
Other data attest to the convergent and divergent validity of the DPQ
(see Table 4.1; Illingworth & Norem, 1991; Norem & Crandall, 1991).
Theoretically, defensive pessimism is motivated by the need to manage
anxiety. Thus, it is not surprising that it is positively correlated with trait
anxiety and neuroticism measures, as well as with other measures that include
a strong anxiety component, such as the Fear of Negative Evaluation scale
(Watson & Friend, 1969), the Impostor Phenomenon scale (Clance, 1985),
and the Self-Handicapping scale (Jones & Rhodewalt, 1982). The DPQ is

84

JULIE K. NOREM

Table 4.1.
Divergent and Convergent Correlates of the Defensive Pessimism
Questionnaire (DPQ)
DPQ
Academic version

DPQ
Social version

-.29
.22
.23
-.24
.05

-.36
.27
.I 1
-.20
.04

Need for Cognition (n Cog)

Need for Structure (n Struct)
Fear of Negative Evaluation (FNE)

.I3
.I5
.22

.09
.32
.36

Beck Depression Inventory (BDI)

Self-Handicappingscale (SHS)
Repression-Sensitization (R-S)
(high score = sensitizing)
Optimism (LOT)
DPQ Social version

.I 8
.27
.26

.22
.49
.29

-.23
-.38

.30

Coed Sample l a
NEO-FFI Extraversion
NEO-FFI Neuroticism
NEO-FFI Conscientiousness
NEO-FFI Agreeableness
NEO-FFI Openness

Coed Sample 2b
ASQ-Internal
ASQ-Stable
ASQ-Global

.23
.I2
.I7

Female samplec
BFI-Extraversion
BFI-Neuroticism
BFI-Conscientiousness
BFI-Agreeableness
BFI-Openness

-.33
.46
.10
-.18
-.27

-.46
-.49
-.02
-.15
-.19

Self-Attributes Questionnaire (SAQ)

Self-clarity (SC)
Impostor Phenomenon scale (IPS)
Optimism (LOT)
DPQ Social version

-.31
-.28
.48
-.35
.48

-.27
-.43
.64
-.30

.38
.02

.52
.17

RRQ-Rumination
RRQ-Reflection

Note. NEO-FFI, Costa & McCrae, 1992; n Cog, Cacioppo, Petty, & Kao, 1984; n Struct, Neuberg &
Newsorn, 1993; FNE, Watson & Friend, 1969; BDI, Beck, 1976; SHS, Jones & Rhodewalt, 1982; R-S, Epstein & Fenz, 1967 (this version of the R-S scale was developed to minimize its correlation with anxiety);
LOT, Scheier & Carver, 1985; ASQ, Peterson et al., 1982; BFI, John, Donahue, & Kentle, 1991; SAQ,
Pelharn & Swann, 1989; SC, Campbell et al., 1996; IPS, Clance, 1985; RRQ, Trapnell & Qrnpbell, 1999.
The Defensives Pessimism Questionnaire (DPQ & R-DPQ) copyright 1994 by Julie K. Norem. Reprinted
with permission.
'For Sample 1, N = 189 except for LOT where N = 400.
b N = 180.
' N = 87.

DEFENSIVE PESSIMISM

85

also clearly related to, but far from redundant with, measures of other
constructs that include aspects of negative thinking, such as RepressionSensitization (Epstein & Fenz, 1967), and the RRQ-Rumination (Trapnell
& Campbell, 1999).' The DPQ does not appear, however, to be correlated
with measures of more general motivations such as need for cognition or
need for structure, or, as a domain-specific strategy for pursuing personal
goals, should it correlate with those general motives.
Also not surprisingly, defensive pessimism correlates negatively with
self-esteem and with self-clarity (Campbell et al., 1996). Given a willingness
to think about negative aspects of both situations and themselves, defensive
pessimists include those negative aspects of self in their self-evaluations,
and they then score lower in self-esteem. Defensive pessimism, however,
may provide individuals experiencing low self-esteem with a way to work
effectively toward raising their self-esteem and increasing their sense of
clarity about themselves (Norem, 1996; chap. 16, this volume).
The DPQ is intended for use as a domain-specific measure of strategies,
and the specific wording of the items should reflect the domain under study.
As mentioned above, people may use different strategies in different domains,
and this hypothesis is supported by the small-to-moderate correlation between the social and academic versions of the scale, which range between
.30 and S O . Thus far, to my knowledge, defensive pessimism has been
researched in academic (see Cantor & Norem, 1989; Cantor, Norem, Niedenthal, Langston, & Brower, 1987; Norem & Cantor, 198613, 1990;Norem
& Illingworth, 1993), social (Showers, 1988, 1992), and recreational or
sports contexts (Spencer, 1993; Spencer & Norem, 1996) using appropriate
versions of the DPQ. Health defensive pessimism has also been researched
(see Kiehl, 1995; Norem & Crandall, 1991) using a somewhat different set
of specifically focused items. The correlations between the domain-specific
versions of the DPQ and other variables are likely to reflect the particular
domain under study.
It is worth noting that the social version of the DPQ correlates somewhat higher with several other measures-for example, the Self-Handicapping scale, the Self-clarity measure, the Impostor Phenomenon scale, and
the RRQ-Rumination measure-than does the academic version. This may
be because the Social DPQ, worded so that ratings are given for social
situations, taps into a broader domain than the academic version, which
presumably refers to a more restricted goal domain.

'The second scale of the RRQ is labeled refictiun, and thus, it may seem as if it should correlate
positively with the reflectivity items on the DPQ. The actual items of the RRQ-Reflection scale,
however, focus primarily on the enjoyment of self-reflection, whereas the reflectivity of the defensive
pessimist tends to be task-focused, motivated by anxiety, focused on negative possible outcomes, and
thus not particularly enjoyable. In other words, people using defensive pessimism before a task are

86

JULIE K. NOREM

EMPIRICAL RELATIONSHIP TO OTHER OPTIMISM AND

PESSIMISM CONSTRUCTS
Theoretically, as noted above, defensive pessimism as a strategy is
distinct from both trait optimism or pessimism, and from a pessimistic
attributional style, and the correlations in Table 4.1 support the distinction
among those constructs. Thus, although defensive pessimism is clearly related
to trait optimism as it is measured by the Life Orientation Test (Scheier
& Carver, 1985), the correlations are not high enough to suggest that they
are measuring the same construct. Moreover, the domain-specificity evident
in the correlations between the social and academic versions of the DPQ
further distinguish it from trait optimism, which should affect expectations
across different domains and situations.
Similarly, the DPQ does not correlate highly with the subscales of
the Attributional Style Questionnaire (Peterson et al., 1982): Its highest
correlation is with the Internality subscale (r = .23; Norem & Sellars, 1995),
which fits other data suggesting that defensive pessimists make internal
attributions. This contrasts with those people who have a depressogenic
attributional style, however, in that defensive pessimists tend to make internal attributions (specifically, to both effort and ability) for both positive
and negative outcomes relevant to the domain in which they use their
strategy. In addition their attributions are not consistently global or to
stable characteristics as is the case with people who have depressogenic
attributional style.

RESEARCH AND ONGOING THEORETICAL DEVELOPMENTS

One of the most interesting developments in research on defensive
pessimism has been elaboration of the role of reflectivity in the defensive
pessimism strategy and the corresponding exploration of the role of avoidance of reflectivity in strategic optimism. In the following I review research
that supports the contention that individuals using defensive pessimism set
defensively low expectations and extensively reflect on possible outcomes
prior to a performance to feel more in control and to keep their anxiety
from interfering with their performance. This research also suggests that
some individuals use strategic optimism: They do not feel particularly anxious
prior to a performance, and they set high expectations. They then actively
avoid reflecting on possible outcomes-even positive ones-because doing
so would make them anxious and jeopardize their performance.
not reflective because they enjoy being reflective, they are reflective because it serves a vital function
for them, even if it is somewhat aversive.

DEFENSIVE PESSIMISM

87

Most of the research on defensive pessimism has contrasted it with

strategic optimism. People using defensive pessimism construe their goals in
significantly different ways than do strategic optimists. Academic defensive
pessimists report that academic goals are more stressful and more difficult
than do strategic optimists, and they feel less in control and more anxious
with respect to those goals-even though the performance histories of these
two sets of participants are comparable. This is true whether one asks people
to appraise goals they generate themselves and categorize as achievement
related, to report on their daily affect in goal-relevant situations by using
experience-sampling methods, or to report their feelings of anxiety and
control prior to a specific performance such as an exam or an experimental
task (Cantor & Norem, 1989; Cantor et al., 1987; Norem, 1987, 1989;
Norem & Cantor, 1990; Norem & Illingworth, 1993).
Thus, defensive pessimists are, in an important psychological sense,
facing a different set of situations than strategic optimists. They must figure
out how to manage their anxiety so that they can approach their goal and
how to keep their anxiety from interfering with their performance. The
optimists task, in contrast, is to avoid arousing anxiety by maintaining their
positive outlook.
One of the ways to understand how processes unfold over time, as
well as to highlight the functions of these particular strategies, is to interfere
with them. Thus, in one of the first experiments on defensive pessimism,
Norem and Cantor ( 198613) interfered with the expectation-setting component of defensive pessimism by telling participants in an experiment that
they should expect to do well. Defensive pessimists who were encouraged
subsequently performed significantly more poorly on the experimental tasks
than did those who were not encouraged (see also Rich & Dalheimer, 1989).
Norem and Illingworth (1993) also tested directly the anxiety management functions of the defensive pessimists playing-through process, as well
as the strategic optimists avoidance of reflectivity prior to a task. They
randomly assigned strategic optimists and defensive pessimists to either a
thought-listing condition, in which they had to list their thoughts about
an upcoming task in response to a series of probes, or to a distraction
condition, in which they worked on a task that had nothing to do with
their pending performance. The thought-listing condition was designed to
mimic the typical reflectivity or playing-through process used by defensive
pessimists prior to a performance, and the distraction condition was intended
to simulate the typical preperformance behavior of strategic optimists, as
well as to make reflecting on the upcoming task difficult.
As predicted, defensive pessimists were significantly more anxious in
the condition in which they were prevented from reflecting prior to their
performance (Norem & Illingworth, 1993, Study l ) , and they performed
significantly more poorly in that condition. Just as important, the optimists

88

JULIE K. NOREM

were significantly more anxious when they had to reflect about the upcoming
task than when they could distract themselves as usual. Mediational analyses
show that managing anxiety really is the key to these performance results.
This account is further bolstered by physiological data from this study,
showing that skin conductance readings illustrate the same interaction
pattern as the self-report and performance data.
In a conceptual replication of this experiment, graduate nursing students participated in an experience-sampling study, in which half of them
reported on their progress toward self-generated goals when they were randomly beeped to fill out emotion and activity reports (Norem & Illingworth, 1993, Study 2). We expected that asking them specifically about
their progress would induce active reflection. Defensive pessimists who were
prompted to report on their progress reported more positive mood, felt more
in control, and found their life tasks easier over a 7-day period than did
defensive pessimists who were not prompted. On a follow-up questionnaire,
the prompted defensive pessimists also reported more progress toward their
goals. In contrast, optimists who were induced to be reflective by this method
reported more negative affect, found their tasks more difficult, and reported
less overall progress than did optimists who were not prompted to reflect.
I have used the metaphor harnessing anxiety to describe how the defensive pessimists strategy works to allow them to do what they want to do,
without being debilitated by their anxiety and negative affect. Defensive
pessimism appears to function as a do-it-yourself cognitive therapy for
anxiety; indeed, it resembles the worst-case analysis and mental rehearsal
techniques often used by cognitive therapists (Beck, 1976; Ingram, 1986;
Ingram & Hollon, 1986). In contrast, the optimists seem able to avoid
becoming anxious by focusing away from thoughts about the upcoming task.
If they are forced to think about possible outcomes, though, they do become
anxious and perform more poorly. Trying to impose a different strategy on
either group disrupts performance, but each group does quite well when
using its preferred strategy.
Further data extend the understanding of the role of reflectivity and
avoidance of reflectivity for defensive pessimists and strategic optimists,
respectively. For example, Spencer and Norem (1996) examined performance on a dart-throwing task as a function of strategy and different kinds
of imagery. Strategic optimists and defensive pessimists were randomly assigned to one of three imagery conditions that were modeled after the
guided-imagery techniques most commonly used by sports psychologists to
help athletes control anxiety and psych up for their performances. Coping
imagery involves imagining something going wrong during a performance
or competition and then imagining fixing or recovering from whatever has
happened. Mastery imagery involves imagining a perfect, flawless performance. Relaxation, in contrast to the other two techniques, involves focusing
DEFENSIVE PESSIMISM

89

away from thoughts about the performance and focusing on relaxing completely. Conveniently, these techniques represent useful ways to operationalize different important facets of reflectivity. Specifically, coping imagery
closely resembles defensive pessimists reflectivity prior to an upcoming task
or performance. Relaxation imagery, in contrast, is similar to what strategic
optimists do, in that it involves avoidance of thoughts about performance.
Mastery imagery is an especially interesting case, because it involves reflective thought, but only about positive outcomes, in contrast to the defensive
pessimists reflective thought about possible negative outcomes. Thus it is
similar to the defensive pessimists strategy because of the reflectivity involved, but also similar to the optimists strategy because it avoids negative thinking.
We predicted that as in previous research both defensive pessimists
and strategic optimists would perform best on the dart-throwing task when
they were in the conditions that best matched their preferred strategies,
that is, the coping imagery condition for the defensive pessimists and the
relaxation condition for the strategic optimists. The results supported our
hypothesis: Defensive pessimists performed significantly better in the coping
imagery condition than in the other two conditions, and strategic optimists
performed significantly better in the relaxation condition than in the other
two conditions. Both groups performed worst in the mastery conditionwhich did not fit either groups strategy-and, as in previous experimental
studies, there were no overall differences in performance between defensive
pessimists and strategic optimists. Thus, both groups were relatively debilitated in the conditions that did not fit their strategies, and both groups did
relatively well in the conditions that matched their strategies.
Sanna (1996,1998) also has shown that the content and consequences
of pre- and posttask cognition for defensive pessimists and strategic optimists
are very different. Congruent with earlier research, his results showed that
defensive pessimists preferred upward prefactuals as they prepared for a
performance. Prefactual thinking refers to mental simulation before the fact,
and upward prefactuals are thoughts about how things might turn out better
than expected (e.g., If I werent so anxious, my speech would go really
well). Strategic optimists, in contrast, preferred not to engage in prefactual
thinking (or reflectivity). They also generated more downward counterfactuals (Sanna, 1996). Counterfactuals are mental simulations after the fact of
outcomes that are different from actual outcomes. Downward counterfactuals
focus on outcomes that are worse than the actual ones (e.g., If I hadnt
crammed at the last minute, I would really have bombed the test). Sanna
further showed that negative mood facilitates the defensive pessimists preferred preparation and their subsequent performance, and positive mood
interferes with it (Sanna, 1998). Conversely, positive mood facilitates the
optimists preferred strategy, and negative mood interferes with it and impairs

90

JULIE K. NOREM

performance. Norem and Illingworth (Illingworth, 1993; Norem & Illingworth, 1999)also found that defensive pessimists performed best in a negative
mood induction condition and worst in a positive mood induction condition,
whereas strategic optimists performed best in a positive mood condition.

STRATEGY AND ADAPTATION

The research described earlier suggests that defensive pessimism and
strategic optimism have important differences in costs and benefits and
different potential risks. As Norem and Illingworth (1993) argued, results
from experimental studies in the laboratory support the contention that
both defensive pessimism and strategic optimism can work quite well: Across
all the laboratory experiments reviewed here, strategic optimists and defensive pessimists perform equally well when they are in conditions that permit
or facilitate use of their preferred strategy. Each strategy, however, leaves
its user potentially vulnerable to different kinds of disruption (Sanna, 1998).
Thus, defensive pessimists seem to be vulnerable to positive mood
because it interferes with their reflectivity prior to a task. It may at first
seem counterintuitive that making anxious people feel good would disrupt
their performance. Being in a good mood, however, may serve as a cue to
stop preparation (Martin, Ward, Achee, & Wyer, 1993). For defensive
pessimists, abandoning their anticipatory strategy may leave them significantly less prepared than they would otherwise be. Feeling good may then
mean that they are even more vulnerable to the disruptive effects of anxiety
that is re-aroused once they have to begin working on the task.
The performance of strategic optimists, in contrast, seems less vulnerable overall to mood inductions prior to a task, although their performance
does suffer somewhat in negative mood conditions (Norem & Illingworth,
1999; Sanna, 1998; Showers, 1992). Manipulations that require the strategic
optimists to reflect prior to a performance do disrupt their outcomes (Norem
& Illingworth, 1993; Spencer & Norem, 1996). It may be that preperformance mood has a less consistent effect than direct manipulation of thought
content because being in a negative mood prompts attempts at mood repair
that do not specifically involve reflecting about the upcoming task and do
not generate anxiety about that task.
When looking at discrete actual performances, defensive pessimism
and strategic optimism seem comparably successful for those who typically
use them. There are other outcomes to consider, however. For example,
defensive pessimists usually report being significantly less satisfied with their
performances than do strategic optimists in experimental studies (Norem
& Cantor, 1986a, 198613; Norem & Illingworth, 1993). Similarly, in field
studies, defensive pessimists typically report less satisfaction with both their
ilEFENSIVE PESSIMISM

91

past achievements and their current endeavors, and they also report more
negative affect than strategic optimists (Cantor et al., 1987; Norem, 1987;
Norem & Cantor, 1990; Norem & Illingworth, 1993). T o the extent that
one can consider affect an outcome of ones strategy, it appears that in this
respect defensive pessimism is less effective than strategic optimism.
There are also some data suggesting that long-term outcomes for defensive pessimists may be worse than those for strategic optimists. Cantor and
Norem (1989) reported that after 3 years in college, defensive pessimists
reported slightly lower grade point averages and more physical and psychological symptoms than did strategic optimists. It is of interest to note that
the worse outcomes were largely attributable to those students who used
defensive pessimism in both social and academic domains. The interpersonal
consequences of defensive pessimism are largely unstudied (Norem, 1991),
but there are data suggesting that people react negatively to others anxiety
(Leary, Kowalski, & Bergen, 1988), as they do to others depression (Coyne,
1976).T o the extent that their strategy is visible to others, defensive pessimists
may create negative impressions or annoy the people around them.2
Recall also that the correlation (Table 4.1) between social defensive
pessimism and indicators of self-esteem difficulties were stronger than those
for academic defensive pessimism. It may be that defensive pessimism is less
effective as a strategy for social situations, in which success is more subjective
and less clearly related to discrete performances for which one can prepare.
(See chapters 12 and 16 in this volume for discussions of the cultural
contexts influencing evaluation of pessimism and optimism.) Of course it
may also be that social anxiety is more generally debilitating than anxiety
about academic perf~rmance.~
Both the affect results and the long-term data suggest that there are
significant costs to defensive pessimism. This does not mean, however, that
there are no potential costs to strategic optimism. Given the experimental
data suggesting that strategic optimists performance may be disrupted by
negative mood and by performance-related thoughts, it is reasonable to ask
how they might fare in situations in which attention to negative information
*Female defensive pessimists list boring and annoying as characteristics of their feared possible
selves more often than do strategic optimists, sugggesting that they may have some insight into the
potential effects that their strategy may have on others.
jFrequently those hearing about defensive pessimism assume that its function is primarily impression
management, that is, that the defensive pessimists are not really pessimistic, but are only trying to
control the expectations of their audience or garner reassurance and praise. T h e available data,
however, do not support this interpretation, at least not in its strong form. For example, the patterns
of anxiety change as a function of strategy use or lack thereof that are observed in self-report data
are repeated in skin conductance measures (Norem & Illingworth, 1993, Study 1). In addition, in
Norems research, defensive pessimists report lower expectations and higher anxiety regardless of
whether those reports are made anonymously or publicly. Finally, defensive pessimism is correlated
with lower scores on both the self-deception and impression management subscales of the BIDR
(Paulhus, 1990).

92

JULIE K. NOREM

might be important to future outcomes or to situations that are somehow

unavoidable and anxiety arousing.
Kiehl (1993, for example, found that defensive pessimists saw AIDSrelated risk behavior as more risky than did strategic optimists. Norem and
Crandall (1991) found that defensive pessimists were more interested in
receiving information about a fictitious disease that they might have contracted than were strategic optimists. These results are suggestive: If the
strategic optimists' attempts to maintain their positive outlook include ignoring or discounting important information or the potential consequences of
risky behavior, then this needs to be counted as a significant potential cost
to the strategy (see also Weinstein, 1980, for work on the risk perceptions
of unrealistic optimists).
Just as important, one must be cautious in evaluating any strategy
solely by comparing it to an alternative strategy without consideration of
the broader life context in which it is used. The effectiveness of a strategy
should be assessed in terms of the goals or problems of those who use it.
Strategy use in the real world is nonrandomly determined: Defensive pessimism is used by those who are anxious. They must manage their anxiety
to perform well, in contrast to the strategic optimists, who are less troubled
initially by anxiety. Defensive pessimism seems an excellent strategy for
those who are anxious, precisely because it addresses their psychological
reality: the need to control anxiety, which rarely simply goes away by wishful
thinking. Although it is interesting to compare the outcomes of defensive
pessimists and strategic optimists, one must also consider how the anxious
defensive pessimists might fare without their strategy.
Sanna (1998) noted that in both the mental simulation and the coping
literature there is a distinction between the preparative or self-improvement
functions and the affective functions of simulation and coping strategies
(Folkman & Lazarus, 1991; Markman, Gavanski, Sherman, & McMullen,
1993; Wood & Taylor, 1991). He argued that defensive pessimists seem to
make more use of the preparative functions of mental simulation, whereas
strategic optimists use the affective functions to maintain their positive
outlook. I further suggest that defensive pessimism may be an especially
useful strategy for self-improvement. In a study of social performance, participants made a 3-minute videotape in which they talked about themselves,
ostensibly so that others could get to know them (similar to a dating service
tape; Norem, 1991). Observers watched the tape and rated the performance
on the same scales that the actors used to rate themselves. Overall, observers
did not rate social defensive pessimists and strategic optimists differentl~.~
'It is important to note that observers only saw the actual performance in this study, not the
preparation for the performance. If observers had watched or heard the anxious defensive pessimists
as they prepared, they might have rated them less positively.

DEFENSIVE PESSlMISM

93

However, defensive pessimists rated themselves higher than did observers

on a measure of how much they needed to improve their social self-presentation. In contrast, strategic optimists rated themselves lower on need for
improvement than observers rated them. Participants in this study were
given both negative and positive false feedback about their videotapes and
then given a surprise recall test about that feedback later in the study.
Overall, defensive pessimists were less accurate in their recall of the feedback,
primarily because they recalled positive feedback as significantly less positive
than it actually was; they were accurate in remembering the negative feedback. In contrast, optimists remembered the negative feedback as significantly less negative than it actually was, and they were more accurate in
their recall of the positive feedback.
The defensive pessimists did not feel good relative to the strategic
optimists in this study, nor were they particularly accurate overall; nevertheless, they clearly did pay attention to negative feedback, and they were
focused on improving their performance. In the real world, negative feedback
may be informative and helpful to ones attempts to improve oneself, even
if it is unpleasant data to process. Defensive pessimists are clearly willing
to hear that feedback, and thus, they would seem more likely than strategic
optimists to benefit from it.
Further data specifically suggest that defensive pessimism may be useful
for self-improvement and generally suggest that it is a useful strategy for
those who are anxious. In a longitudinal study of female college students,
I compared defensive pessimists, not only to strategic optimists, but also to
highly anxious students who did not report using defensive pessimism
(Norem, 1996). Those students using defensive pessimism were not significantly different from the others in the anxious group in levels of trait anxiety,
in any of the Big Five personality traits, in SAT scores, in high school
grade point average, or in self-esteem at the beginning of their first year of
college. Both of the anxious groups initially reported lower self-esteem than
did strategic optimists in the sample. Over their college years, however, the
defensive pessimists showed significant increases in self-esteem. The anxious
students who did not use defensive pessimism, in contrast, showed no significant increase in self-esteem through college. In addition, defensive pessimists
achieved both higher grade point averages and greater subjective satisfaction
than did other anxious students. Defensive pessimism, in this comparison,
appears quite adaptive.
It is important to note, however, that although defensive pessimism
may be helpful, those who use it do not necessarily become less anxious or
generally more positive over time. Indeed, there may be self-perpetuating
effects of the strategy (Sanna, 1998).Defensive pessimists are likely to recall
their anxiety prior to any given success, and that is likely to color their
anticipation of future endeavors. They are likely to recall their effort and

94

JULIE K. NOREM

to attribute their success to their effort. This internal attribution is positive

in the sense that it suggests success is under ones control and replicable,
but it also places emphasis on the necessity for hard work, and it has
affectively different consequences than believing that success is relatively
assured because of ones ability. Recalling the social performance study, if
defensive pessimists typically recall or interpret positive feedback as less
positive than it actually is, and if they focus particularly on negative feedback,
then they are likely to maintain their sense of anxiety, their low expectations,
and their focus on potential negative outcomes the next time they encounter
a similar situation. Correspondingly, if the optimists pay relatively less attention to negative feedback or are more likely to remember it as positive, or
if they simply remember positive feedback better than negative feedback,
the next time they enter a similar situation, they are not likely to feel
anxious or stressed.
Despite the potentially self-perpetuating affective cycle that might be
associated with each strategy, however, strategies are hypothesized to be
potentially changeable. Were there to be significant changes in construal
of self, beliefs about the world, or appraisal of goals, one might expect
corresponding strategy change. A spectacular failure, especially one that
could be traced to lack of preparation for the unexpected, could induce
more reflection among strategic optimists. Habituation to a particular situation might sufficiently decrease anxiety for the defensive pessimists so that
they could anticipate without worry. Indeed, Martin (1999) found significant
changes in strategy during the first 2 years of being at a university among
Australian students. He showed that there was significantly more change
in strategies than in other personality variables over this time, which fits
with the theoretical conception of strategies as more mutable than traits
or motives. In addition, he was systematically able to relate strategy change
to significant changes in life circumstances. Specifically, those students who
became more anxious about their academic work because of their experiences
were more likely to change from using strategic optimism to using defensive
pessimism. In contrast, those students who began to feel as if they were in
control of their academic work were more likely to adopt a more optimistic strategy.

CONCLUSION AND FUTURE DIRECTIONS

Research on defensive pessimism and strategic optimism is relevant
to other research on optimism and pessimism for several reasons. It is useful
to note that pessimistic expectations can be merely defensive, that is, that
they do not necessarily lead to self-fulfillingprophecies. Defensive pessimism,
although perhaps in some sense a special case, suggests that personality
DEFENSIVE PESSIMISM

95

places limits on the extent to which one can assume that positive thinking
always leads to positive outcomes. More broadly, the research reviewed
above points to the importance of considering expectations, or virtually any
single trait or tendency, in the broader context of individuals personalities
and of their life contexts. This research shows that there are multiple paths
that people may navigate toward their goals and that sometimes their goals
reflect differences in their starting points. This consideration is important
both to efforts to capture the texture of real life in psychological theories,
as well as to attempts to apply the results of psychological research in
therapeutic or other contexts. World outlooks are not one size fits all
because people do not all live in the same world. Psychologists and educators
do a disservice to the people they are trying to help if they ignore individual
differences in temperament, history, and circumstance that make particular
strategies more or less appropriate for specific individuals (see chap. 12 in
this volume).
A number of important questions about defensive pessimism and strategic optimism await further research. As already discussed, there is relatively
little work on the interpersonal consequences of defensive pessimism. It
will be interesting to explore both short- and long-term reactions to individuals who use the strategy. One obvious question is whether ones own preferred
strategies influence ones reactions to others strategies, for example, do
defensive pessimists get along better with other defensive pessimists than
with strategic optimists?
The influence of both strategic optimism and defensive pessimism on
decision making is also a potentially fruitful arena for further exploration.
One might reasonably predict that defensive pessimists would be relatively
averse to risk, whereas strategic optimists might take bigger risks. Altematively, the defensive pessimists extensive preparation might make them
more willing to take some kinds of risks, for example, in cases in which
extensive information is available. In a domain such as the stock market,
the two strategies could lead to markedly different outcomes over time: The
defensive pessimists might be expected both to have fewer large losses and
fewer large gains than do the strategic optimists. More generally, relatively
little research on these strategies outside of the academic domain, among
people other than college students has been conducted.
One of the most important remaining questions about these strategies
concerns their developmental origins. Although there are some data about
the family environments of defensive pessimists and strategic optimists
(Norem, 1987), there has been little systematic exploration of the factors
that might influence who develops each strategy. Similarly, virtually nothing
is known about when the strategies might emerge. Also little is known
about how flexible individuals might be in their use of various strategies,
which may prove to be very important to their adaptive value.

96

JULIE K . NOREM

A pessimist might see the number and range of these questions as

daunting, but let us reflect that although the pitfalls encountered as we
continue this research may be considerable, the potential rewards in understanding are also great, and the journey itself should prove fascinating. A
good defensive pessimist, then, would confront his or her anxiety and persevere.

REFERENCES
Adler, A. (1935/1979). Typology of meeting life problems. In H. L. Ansbacher &
R. R. Ansbacher (Eds.), Alfred Adkr: Superiority and social interest (pp. 66-71).
New York: W. W. Norton.
Beck, A. T. (1976). Cognitive therapy and the emotional disorders. New York: Intemational Universities Press.
Berglas, S. ( 1985). Self-handicapping and self-handicappers: A cognitive-attributional model of interpersonal self-protective behavior. In R. Hogan (Ed.),
Perspectives in personality (Vol. 1, pp. 235-270). Greenwich, CT:JAI Press.
Caccioppo, J. T., Petty, R. E., & Kao, C. F. (1984). The efficient assessment of
need for cognition. Journal of Personality Assessment, 48, 306-307.
Campbell, J. D., Trapnell, P. D., Heine, S. J., Lavalee, L. F., Katz, I. M., & Lehman,
D. R. ( 1996). Self-concept clarity: Measurement, personality correlates, and
cultural boundaries. Journal of Personality and Social Psychology, 70, 141-156.
Cantor, N., & Kihlstrom, J. F. (1987). Personality and social intelligence. Englewood
Cliffs, NJ: Erlbaum.
Cantor, N., & Norem, J. K. (1989). Defensive pessimism and stress and coping.
Social Cognition, 7, 92-1 12.
Cantor, N., Norem, J. K., Niedenthal, P. M., Langston, C. A., & Brower, A. M.
( 1987). Life tasks, self-concept ideals, and cognitive strategies in a life transition. Journal of Personality and Social Psychology, 53, 1178-1191.
Clance, P. R. (1985). The impostor phenomenon: Overcoming the fear that haunts your
success. Atlanta, GA: Peachtree.
Costa, P. T., & McCrae, R. R. (1992). Revised NEO Personality Inventory (NEOPI-R) and NEO Five-Factor Inventory (NEO-FFI) professional manual. Odessa,
FL: Psychological Assessment Resources.
Coyne, J. (1976). Depression and the response of others. Journal of Abnormal
Psychology, 85, 29-45.
Cuddihy, N., & Ronan, G. F. (1990, August). Relationshipbetween optimism, defensive
pessimism and personal problem-solving. Paper presented at the 98th Annual
Convention of the American Psychological Association, Boston, MA.
Epstein, S., & Fenz, W. B. (1967). The detection of areas of emotional stress
through variations in perceptual threshold and physiological arousal. Journal
of Experimental Research in Personality, 2, 191-199.
DEFENSIVE PESSIMISM

97

Folkman, S., & Lazarus, R. S. (1991). Coping and emotion. In A. Monat & R. S.
Lazarus (Eds.), Stress and coping: An anthology (3rd ed., pp. 207-227). New
York: Guilford.
Hammontree, S. R., & Ronan, G. F. (1992, August). Optimism, pessimism and
defensive pessimism. Paper presented at the 100th Annual Convention of the
American Psychological Society, Washington, DC.
Higgins, R. L., & Harris, R. N. (1988). Strategic alcohol use: Drinking to selfhandicap. Journal of Social and Clinical Psychology, 6, 191-202.
Illingworth, K. S. S. (1993). Cognitive strategies and mood: The role of affect in the
strategic use of optimism and defensive pessimism. Unpublished doctoral dissertation, Northeastern University, Boston, MA.
Illingworth, K. S. S., & Norem, J. K. (1991). Convergent and divergent correlates of
optimism and defensive pessimism. Paper presented at the Midwestern Psychological Association, Chicago, IL.
Ingram, R. E. (Ed.). (1986). Information processing approaches to clinical psychology.
Orlando, FL: Academic Press.
Ingram, R. E., & Hollon, S. D. (1986). Cognitive therapy of depression from an
information processing perspective. In R. E. Ingram (Ed.), Informationprocessing
approaches to clinical psychology (pp. 22-43). San Diego, CA: Academic Press.
Isaacowitz, D. M. & Seligman, M. E. P. (1998 August). Prevention of depression
in older adults: Theory, methodology, and pitfalls. Invited paper presented in
Opportunities and Pitfalls in Adult and Older Adult Prevention Research.
Symposium at the annual meeting of the American Psychological Association,
San Francisco, CA.
John, 0.P., Donahue, E. M., & Kentle, R. L. (1991, July). The BigFive Inventoryversions 4a and 54 (Tech. Rep. No. 10). Berkeley, CA: University of California
at Berkeley, Institute of Personality and Social Research.
Jones, E. E., & Rhodewalt, F. (1982). Self-Handicapping Scale. Unpublished scale.
Princeton University, NJ.
Kiehl, E. (1995). Attitudes towards AIDS among unrealistic optimists, health optimists,
and health defensive pessimists. Unpublished honors thesis. Psychology Department, Wellesley College, Wellesley, MA.
Leary, M. R., Kowalski, R. M., & Bergen, D. J. (1988). Interpersonal information
acquisition and confidence in first encounters. Personality and Social Psychology
Bulletin, 14, 68-77.
Markman, K. D., Gavanski, I., Sherman, S. J., & McMullen, M. N. (1993). The
mental simulation of better and worse possible worlds. Journal of Experimental
Social Psychology, 29, 87-109.
Martin, A. (1999). Self-hndicapping and defensive pessimism: Predictors and consequences from a self-worth motivation perspective. Unpublished doctoral dissertation, University of Western Sydney, Sydney, Australia.
Martin, L. L., Ward, D. W., Achee, J. W., & Wyer, R. S. (1993). Mood as input:
People have to interpret the motivational implications of their moods. Journal
of Personality and Social Psychology, 64, 317-326.

98

JULIE K. NOREM

Neuberg, S. L., & Newsom, J. T. (1993). Personal need for structure: Individual
differences in the desire for simple structure. Journal of Personality and Social
Psychology, 65, 113-131.
Norem, J. K. (1987). Strategic realities: Optimism anddefensive pessimism. Unpublished
doctoral dissertation, University of Michigan, Ann Arbor.
Norem, J. K. ( 1989). Cognitive strategies as personality: Effectiveness, specificity,
flexibility, and change. In D. M. Buss & N. Cantor (Eds.), Personality psychology:
Recent trends and emerging issues (pp. 45-60). New York: Springer-Verlag.
Norem, J. K. (1991, May). Self-enhancementand self-deception: Some costs of optimism
and defensive pessimism. Paper presented at the Midwestern Psychological Association, Chicago, IL.
Norem, J. K. (1996, August). Cognitive strategies and the rest of personality. Paper
presented at the Annual Meeting of the American Psychological Association,
Toronto, Canada.
Norem, J. K. (1998). Should we lower our defenses about defense mechanisms?
Journal of Personality, 66, 895-917.
Norem, J. K., & Cantor, N. (1986a). Anticipatory and post hoc cushioning strategies: Optimism and defensive pessimism in riskysituations. Cognitive Therapy
and Research, 10, 347-362.
Norem, J. K., & Cantor, N. (198613). Defensive pessimism: Harnessing anxiety as
motivation. Journal of Personality and Social Psychology, 51, 1208-1217.
Norem, J. K., & Cantor, N. (1990). Cognitive strategies, coping and perceptions
of competence. In R. J. Sternberg & J. J. Kolligian (Eds.), Competence considered
(pp. 190-204). New Haven, CT:Yale University Press.
Norem, J. K., & Crandall, C. S. (1991, June). Defensive pessimism and repressionsensitization show discriminant validity. Paper presented at the 3rd Annual Convention of the American Psychological Society, Washington, DC.
Norem, J. K., & Illingworth, K. S. S. (1993). Strategy-dependent effects of reflecting
on self and tasks: Some implications for optimism and defensive pessimism.
Journal of Personality and Social Psychology, 65, 822-835.
Norem, J. K., & Illingworth, K. S. S. (1999). Mood and performance among strategic
optimists and defensive pessimists. Unpublished manuscript. Department of Psychology, Wellesley College. Wellesley, MA.
Norem, J. K., & Sellars, R. (1995). Unpublished correlation matrix. University of
Virginia, Charlottesville.
Paulhus, D. L. (1990). Measurement and control of response bias. In J. P. Robinson,
P. Shaver, & L. Wrightsman (Eds.), Measures of personality and social psychological attitudes (pp. 17-60). San Diego, CA: Academic Press.
Pelham, B. W., & Swann, W. B., Jr. (1989). From self-conceptions to self-worth:
O n the sources and structure of global self-esteem. Journal of Personality and
Social Psychology, 57, 672-680.
DEFENSIVE PESSIMISM

99

Peterson, C., Semmel, A., von Baeyer, C., Abramson, L. Y., Metalsky, G. I., &
Seligman, M. E. P. (1982). The Attributional Style Questionnaire. Cognitive
Therapy and Research, 6, 287-299.
Rich, A. R., & Dalheimer, D. (1989). The power of negative thinking: Irrational
cognitions. Journal of Cognitive Psychotherapy: An lnternational Quarterly, 3 ,
15-30.
Sanna, L. J. ( 1996). Defensive pessimism, optimism, and simulating alternatives:
Some ups and downs of prefactual and counterfactual thinking. Journal of
Personality and Social Psychology, 71, 1029-1036.
Sanna, L. J. (1998). Defensive pessimism, and optimism: The bitter-sweet influence
of mood on performance and prefactual and counterfactual thinking. Cognition
and Emotion, 12, 635-665.
Scheier, M. F., & Carver, C. S. (1985). Optimism, coping and health: Assessment
and implications of generalized outcome expectancies. Health Psychology, 4,
2 19-247.
Showers, C. (1988). The effects of how and why thinking on perceptions of future
negative events. Cognitive Therapy and Research, 12, 225-240.
Showers, C. ( 1992). The motivational and emotional consequences of considering
positive or negative possibilities for an upcoming event. Journal of Personality
and Social Psychology, 63, 474-483.
Showers, C., & Ruben, C. (1990). Distinguishing defensive pessimism from depression: Negative expectations and positive coping mechanisms. Cognitive Therapy
and Research, 14, 385-399.
Spencer, S. M. (1993). Defensive pessimism andstrategic optimism in the athletic domain:
An evaluation of performance outcomes and health-related aspects. Unpublished
doctoral dissertation, Northeastern University, Boston, MA.
Spencer, S. M., & Norem, J. K. (1996). Reflection and distraction: Defensive
pessimism, strategic optimism, and performance. Personality and Social Psychology Bulletin, 22, 354-365.
Taylor, S. E., & Brown, J. D. (1988). Illusion and well-being: A social psychological
perspective on mental health. Psychological Bulletin, 103, 193-210.
Trapnell, P. D., & Campbell, J. D. (1999). Private self-consciousness and the fivefactor model of personality: Distinguishing rumination from reflection. Journal
of Personality and Social Psychology, 76, 305-319.
Watson, D., & Friend, R. (1969). Measurement of social-evaluative anxiety. Journal
of Consulting and Clinical Psychology, 3 3 , 448-457.
Weinstein, N. D. (1980). Unrealistic optimism about future life events. Journal of
Personality and Social Psychology, 39, 806-820.
Wood, J. V., & Taylor, S. E. (1991). Serving self-relevant goals through social
comparison. In J. Suls & T. A. Wills (Eds.), Social comparison: Contemporary
theory and research (pp. 23-49). Hillsdale, NJ: Erlbaum.

f 00

JULIE K. NOREM

OPTIMISM AND HOPE CONSTRUCTS:

VARIANTS ON A POSITIVE
EXPECTANCY THEME
C. R. SNYDER, SUSIE C. SYMPSON, SCOTT T. MICHAEL,
AND JEN CHEAVENS

In recent years, the effects of positive thinking have received growing

attention by psychologists and health professionals (Snyder & McCullough,
2000). Since the late 1950s, increasing numbers of physicians have acknowledged the benefits of thoughts and feelings characterized by hope (Frankl,
1963; Pelletier, 1977; Schmale, 1972; Siegel, 1986; Simonton, MatthewSimonton, & Creighton, 1978). As early as 1959, for example, Karl Menninger called on his colleagues to rediscover the power of hope. His sentiments were echoed in a 1975 commencement address in which Jerome
Frank acknowledged that Cartesian dualism, embraced by the world of
medicine, resulted in practitioners ignoring the hope and faith of their
clients. He went on to suggest that hope was crucial for the course of
treatment (see Snyder, Michael, & Cheavens, 1999). These pioneers in the
general area of positive thinking would be pleased to see that the advantages

long thought to be associated with a positive mental attitude are being

confirmed by recent psychological researchers (Scheier & Carver, 1985,
1992; Snyder, 1994, 2000a, 2000b; Snyder, Irving, & Anderson, 1991). As
psychology has embraced the importance of a positive mental attitude, a
number of theoretical approaches have been proposed. In this chapter, we
examine two of these approaches: optimism and hope.
In the seminal introductory article on their concept, Scheier and
Carver (1985) defined optimism as the stable tendency to believe that good
rather than bad things will happen (p. 219). Their Life Orientation Test
(LOT) was revised nearly a decade later (LOT-R; Scheier, Carver, &
Bridges, 1994).For a more thorough description of the LOT, see the methodological introduction (chap. 9, this volume).
A cognitive model of hope has been proposed as another explanation
for the positive thinking process (Snyder, 1989,1994,2000a; Snyder, Irving,
et al., 1991). Although most people may think of an optimist as being
hopeful, the two concepts as described in this chapter have similarities, as
well as some differences. In the following sections, we present Snyders
model of hope and the measures developed to assess individual differences
in levels of hope. Then the similarities and differences between hope and
optimism are discussed. Research findings relevant to hope also are addressed.
Finally, brief statements are made about the major distinction and fundamental shared characteristic between the concepts of optimism and hope.

HOPE THEORY
Snyder and his colleagues (Snyder, Harris, et al., 1991; Snyder, Irving,
et al., 1991) conceptualized hope as a cognitive set that is directed at goal
attainment. Hope is seen as having two interrelated components that are
reciprocal. First, one must have a perception of successful agency regarding
ones goals. Agency involves a sense of successful use of energy in the
pursuit of goals in ones past, present, and future. Thus, agency is the mental
motivation that one uses to initiate and sustain movement toward a goal
(Snyder, 1994,2000a). The second component necessary to this conception
of hope is the perceived ability to generate successful routes or pathways
to attain ones goals. In other words, puthwuys thinking is the perceived
capability of imagining ways to reach a given goal, including the formation
of subgoals along the way (Snyder, 1994, 2000a). Thus, hope is formally
defined as a positive motivational state that is based on an interactively
derived sense of successful (a) agency (goal-directed energy), and (b) pathways (planning to meet goals) (Snyder, Irving, et al., 1991, p. 287).
According to hope theory, successful movement toward ones goals
requires both agency and pathways thinking. In this regard, one may have

102

SNYDER ET AL.

known people who profess to have the requisite agency thinking to propel
them toward a future destination, yet they lack the perceived ability to plan
effective strategies (pathways) to get there. On the other hand, one also
may be familiar with those people who have thought of numerous ways to
achieve their goals, but they seem to lack the agency to implement these
plans. These people may appear to be idle dreamers, or they even may be
labeled as just plain lazy. In our model, we suggest that the person lacking
either agency or pathways thinking also lacks hope. In this regard, Snyder
and his colleagues (Snyder, Harris, et al., 1991) argued that to sustain
movement toward the goals in ones life . . . both the sense of agency and
the sense of pathways must become operative. That is, both agency and
pathways are necessary, but neither is sufficient to define hope (p. 571).
In Figure 5.1 we see that ones dispositional agency and pathways (at
the far left, learning history) provide the starting point for any potential
goal-pursuit activity. For detailed discussions of the developmental processes
that produce agency and pathways thought, see Snyder (1994, chap. 3),
Snyder (2000a), and Snyder, McDermott, Cook, and Rapoff (1997).
The next step in the hope model is an appraisal of a goal outcome
value, such that a goal must be of sufficient magnitude to command sustained

Figure 5.1. Schematic of feed-fonnrard and feed-back functions involving agentic and
pathways goal-directed thoughts in hope theory.

OPTIMISM AND HOPE CONSTRUCTS

103

attention (see the pre-event stage in Figure 5.1). Assuming that the goal
is of sufficient value, the event sequence begins with the agency and
pathways components iterating so as to enhance each other. That is to say,
the agency and pathways components are continually affecting and being
affected by each other. Thus, there is a reciprocal action between the belief
that one can attain ones goals and the perception that there are effective
strategies to attain those goals.
Based on the iterative agency and pathways cognitive analysis, the
person moves toward the goal and either does or does not attain it (see
goal behavior in Figure 5.1). As can be seen in the boldface arrows that
run left to right, a persons goal-directed thinking advances, but with the
possibility of feedback within the major components (observe arrows going
both ways). After the goal attainment or nonattainment, there are feedback
processes that reflect the impact of the resulting goal engagement on the
other components of the hope model (see regular weight arrows moving
from right to left). Riding on these feedback lines, the persons emotional
reaction is being fed back to the components in the system. In hope theory,
positive emotions are hypothesized to result from perceived successful goal
attainment; conversely, negative emotions reflect a perceived impeded or
unattained goal outcome. Thus, in hope theory, perceptions about ones
success in goal pursuits drive emotion (see Snyder et a1.,1996, for empirical
support). Overall we have posited an interrelated system of teleological
thinking that permits modifying feedback at various points in the temporal
sequence of our hope model. Accordingly, through the feedback process,
an individuals level of hope is sustained or changed over time.
Optimism, as defined by Scheier and Carver (1985) and hope, as
defined by Snyder and his colleagues (1991) are constructed within an
expectancy-value framework to motivation; moreover, both are conceptualized as relatively stable characteristics that reflect general expectations about
the future. These concepts differ, however, in how the expectancies are
conceptualized to influence behaviors. Although they propose that outcome
expectancies are the most important element that is related to predicting
goal-directed behaviors, Scheier and Carver ( 1985) placed relatively less
emphasis on the bases of these expectancies. Scheier and Carver intentionally de-emphasized the role of personal efficacy in optimism. On this point
they wrote,
Our own theoretical approach emphasizes a persons expectancies of
good or bad outcomes. Most of the current measures that might otherwise be seen as adequate measures of optimism have confounded these
outcome expectancies with a host of related variables such as morale,
meaningfulness, well-being, and most notably, attributions of causes for
the expectancies. It is our position that outcome expectancies per se
are the best predictors of behavior rather than the bases from which

104

SNYDER ET AL.

those expectancieswere derived. A person may hold favorable expectancies for a number of reasons-personal ability, because the person is
lucky, or because he is favored by others. The result should be an
optimistic outlook-expectations that good things will happen.
(Scheier & Carver, 1985, p. 223)
The hope model involves two types of expectancies: agency and pathways. The agency component is analogous to an efficacy expectancy, a belief
in ones ability to successfully begin and maintain movement toward ones
goals. As we have described in the hope model, ones perceived agency is
based on previous goal-pursuit experience and feedback in a given goal
pursuit (Snyder, Irving, et al., 1991). Agency interacts with the perceived
ability to generate strategies that can be used to achieve those goals (i.e.,
the pathways component of hope); further, it is the continuous interaction
of these two cognitive components that results in an outcome expectancy
related to goal pursuits. Thus, in the Scheier and Carver model of optimism,
it appears to be implicitly assumed that the pathways-related expectations
in relation to goals are being tapped, whereas in hope theory, there is an
explicit emphasis on the pathways component. T o highlight this distinction,
the optimism model rests on agency-like expectancies regarding goal attainment, whereas in the hope model, the pathways component of goal-directed
thought is added.
Although optimism as measured by the LOT does not explicitly tap
the pathways component, it should be acknowledged that from the earliest
(e.g., Scheier & Carver, 1985) to most recent papers (e.g., Carver & Scheier,
1999), there are indications that optimists implicitly may use such planful
thought. That is to say, the optimists, as measured by the LOT, exhibit
elevated problem-focused coping (Scheier, Weintraub, & Carver, 1986;
Strutton & Lumpkin, 1992) and planfulness (Fontaine, Manstead, &
Wagner, 1993; Friedman et al., 1992). Therefore, the positive goal-directed
expectancies as tapped by responses to the LOT and LOT-R may capture
pathways-related thinking. In this regard, hope theory and related measurement devices for children (Snyder, Hoza, et al., 1997) and adults (Snyder,
Harris, et al., 1991; Snyder, Sympson, et al., 1996) explicitly build in the
pathways component. O n this point, because items directly tapping pathways
are not on the LOT or the LOT-R, it is not surprising that the pathways
component of the Hope Scale has been found to be orthogonal to items
on the LOT in a factor analysis (Magaletta & Oliver, 1999). How the Hope
Scale with its explicit pathways component can provide unique predictive
variance in relation to the LOT will be discussed subsequently.
In keeping with the two-component definition of hope, a scale was
developed to assess individual differences in levels of dispositional hope.
The Hope Scale (Snyder, Harris, et al., 1991) contains both the agency
and pathways components of hope, and it is the focus of the next section.
OPTIMISM AND HOPE CONSTRUCTS

105

As shown in Appendix 5.1, the Hope Scale contains four items tapping
agentic thought, four items tapping pathways thought, and four filler items.
The respondent uses an 8-point response continuum to rate the personal
applicability of each item, with total Hope Scale scores ranging from a low
of 8 to a high of 64. Additionally, for those people who may be interested
in an ongoing, momentary index of hope, we have developed the State
Hope Scale (Snyder, Sympson, et al., 1996),which can be seen in Appendix
5.2. This brief index has three pathways and three agency items, and people
use the same response continuum as for the dispositional Hope Scale. Unlike
the dispositional Hope Scale, the items on the State Hope Scale reflect
ongoing goal-related thoughts, and the person is asked to respond while
thinking about the here and now. Total State Hope Scale scores can
range from a low of 6 to a high of 48.

COMPARING THE DISPOSITIONAL HOPE SCALE

AND THE LOT
The Hope Scale (Snyder, Harris, et al., 1991), and the LOTLOT-R
(Scheier &Carver, 1985; Scheier et al., 1994)have many things in common.
Fundamentally, both stem from an expectancy-value approach to motivation, both were designed to assess individual differences in what are thought
to be stable characteristics reflecting general expectations about the future,
and both have been submitted to considerable validational procedures and
used in a number of studies.
Factor Structure

Although both the LOT and the Hope Scale have been found to have
two factors, Scheier and Carver (1985) described the factors of the LOT
as positive and negative views of the same concept (Im always optimistic
about my future and I rarely count on good things happening to me).
In the validation studies of the subsequently developed LOT-R, Scheier,
Carver, and Bridges ( 1994)found one factor, which they labeled optimism.
In another analysis, Affleck and Tennen (1996) suggested that the LOT
contains optimism and pessimism as two orthogonal (independent) constructs; moreover, confirmatory factor analyses have revealed that the LOT
consists of separate optimism and pessimism factors (Robinson-Whelen,
Kim, MacCallum, & Kiecolt-Glaser, 1997).
Factor analytic procedures have shown that the Hope Scale has two
distinct factors that correspond with the theoretical model. Four agency
items tap individuals sense of successful determination in relation to their
goals. Of these, one item refers to the past (Ive been pretty successful in
106

SNYDER ET AL.

life), two items refer to the present (I energetically pursue my goals and
I meet the goals that I set for myself), and one item refers to the future
(My past experiences have prepared me well for my future; Snyder, Harris,
et al., 1991). The second factor consists of the four items assessing an
individuals belief in his or her ability to generate the pathways to goals
and overcome any obstacles that may be encountered (e.g., I can think of
many ways to get out of a jam. Even when others get discouraged, I know
I can find a way to solve the problem; Snyder, Harris, et al. 1991).Although
factor analyses of the Hope Scale support the existence of the two separate
components, the agency and pathways components are positively correlated
with each other (typical magnitudes of rs = .40). Thus, although agency
and pathways are related, they are not measuring the same thing (Snyder,
Harris, et al., 1991). Furthermore, confirmatory factor analytic procedures
have shown that the Hope Scale indeed reflects an overarching hope construct that is undergirded by the agency and pathways components (Babyak,
Snyder, & Yoshinoba, 1993).
In light of the conflicting findings regarding the factor structure of the
LOT/LOT-R, additional studies would be enlightening. One such possible
study would involve giving a large sample of research participants both the
LOT-R and Hope Scale. We hypothesize that factor analysis on all the
items would show that the optimism items of the LOT-R and the agency
items of the Hope Scale load on one factor. Another means of investigating
a similar question would be to examine the subscale patterns of correlations.
In this regard, the LOT-R optimism items should relate more strongly to
the agency than the pathways subscales of the Hope Scale. These predicted
findings would support our speculation that hope may add predictive power
beyond optimism in that both the agentic and pathways components are
explicitly tapped by the Hope Scale. The only reported study related to the
aforementioned point was a factor analysis conducted by Magaletta and
Oliver (1999), who found that the pathways component in the Hope Scale
was orthogonal to LOT items.
Internal Consistency

Any scale should demonstrate that the items are internally consistent
(i.e., items relate together so as to measure the same concept, such as
optimism or hope). Numerous studies have yielded Cronbachs alpha coefficients (a statistical procedure for corroborating the internal cohesiveness
of scale items with each other) within acceptable ranges for the overall
Hope Scale, as well as the two subscales. In the initial scale development
article including eight samples (Snyder, Harris, et al., 1991), the Hope
Scale exhibited alpha coefficients ranging from .74 to .84 (with statistically
significant item-remainder coefficients of .23 to .63). The four-item agency
OPTIMISM AND HOPE CONSTRUCTS

107

subscale had alpha coefficients ranging from .71 to .76 (significant itemremainder coefficients [the extent to which individual items correlate with
the total of all other items when those individual item scores are subtracted]
of .40 to .72), and the four-item pathways subscale had alpha coefficients
ranging from .63 to .80 (significant item-remainder coefficients of .36 to
-63). Subsequent studies routinely have produced alpha coefficients ranging
in the high .70s to the high .80s (Snyder, Sympson, et al., 1996).
Scheier and Carver (1985) reported a Cronbach alpha coefficient of
.76 for the LOT; similarly, Scheier, Carver, and Bridges (1994) reported a
Cronbach alpha coefficient of .78 for the LOT-R. Thus, both the LOT/
LOT-R and the Hope Scale have acceptable internal reliabilities.
Temporal Stability
Because the Hope Scale was developed as a dispositional index, one
would expect an individuals scores to remain stable across time. Different
samples of college undergraduates have produced significant test-retest correlations: r = .85 over a 3-week interval (Anderson, 1988); r = .78 over a
4-week interval (Sympson, 1993);r = .73 over an 8-week interval (Harney,
1989); and rs = .76 and .82, respectively, over a 10-week interval in two
studies (Gibb, 1990; Yoshinobu, 1989).
Scheier and Carver (1985) found a test-retest correlation of .79 over
a 4-week interval for the LOT. Additionally, for the LOT-R, Scheier,
Carver, and Bridges ( 1994) reported acceptable test-retest correlations for
4 months ( r = .68), 12 months ( r = .60), 24 months ( r = .56), and 28
months ( r = .79). These correlations, taken together, indicate that the Hope
Scale, the LOT, and the LOT-R all exhibit both short- and long-term
temporal stability. That is to say, the score on any one of the scales derived
at one point in time is similar to the score when the same test is taken
again at a later time.
Convergent and Discriminant Validity

To demonstrate the concurrent validity of a new instrument, researchers need to examine the correlation of responses on the new scale with
responses to selected existing scales. Measures that are designed to gauge
similar constructs should yield positive relationships by way of correlations,
and scales evaluating conceptually opposed processes are expected to yield
negative correlations. This is the essence of convergent validity. Likewise,
no significant correlations should be found between scores on a new scale
and scores on those measures supposedly tapping unrelated concepts, which
is the core idea of discriminant validity. In this regard, both the Hope Scale
108

SNYDER ET AL.

and the LOT/LOT-R have been evaluated in relation to a number of other

measures. The results of identical comparisons are reviewed next.
As would be expected, both the LOT and Hope Scale scores were
positively correlated with Rosenbergs ( 1965) Self-Esteem Scale. The LOT
exhibited a correlation of .48 (Scheier & Carver, 1985), and the Hope
Scale correlation was 5 8 (Snyder, Harris, et al., 1991). In reflecting on the
theoretical underpinnings of these constructs, we think that these results
are understandable. Although the authors of the LOT acknowledge that
most optimists probably derive their sense of optimism from a history of
successes, in which they have demonstrated their personal mastery, they
also believed that there exists a substantial minority of people for whom
a sense of optimism derives from external causes (Scheier & Carver, 1985,
p. 221). In comparison, hope theory focuses on the experiences that people
have had that contribute to their belief in future events. Furthermore,
individuals with higher levels of hope would be expected to have an enhanced sense of self-esteem both because of past successes and because of
their beliefs that workable routes to future goal pursuits are likely.
Both Hope Scale and LOT scores have correlated positively with scores
on the Generalized Expectancy for Success Scale (GESS; Fibel & Hale,
1978). The GESS assesses expectancies for attaining goals across a variety
of situations, and it has exhibited positive correlations of .54 and .55 with
the Hope Scale in two separate studies (Gibb, 1990; Holleran & Snyder,
1990). The correlation of the GESS with the LOT was .56 in a study
reported by Smith, Pope, Rhodewalt, and Poulton (1989).
Those people who have higher levels of hope and optimism also should
score lower on measures of hopelessness and depression. When LOT and
Hope Scale scores were correlated with scores on the Hopelessness Scale
(Beck, Weissman, Lester, & Trexler, 1974) and the Beck Depression Inventory (BDI; Beck, Ward, Mendelsohn, Mock, & Erbaugh, 1961),the expected
inverse relationships were observed. Hope correlated -.5 1with the Hopelessness Scale, and -.42 with depression as measured by the BDI (Gibb, 1990),
whereas the LOT correlated -.47 and -.49, respectively, with the hopelessness and depression indices (Scheier & Carver, 1985).
In assessing the discriminant validity of each scale, their respective
authors used the same three measures and found similar results. The
Marlowe-Crowne Social Desirability Scale (Crowne & Marlowe, 1960)
correlated .30 with the Hope Scale and .26 with the LOT. This correlation
was explained by Snyder and colleagues as reflecting a slight positive selfpresentational style that is part of adaptive coping (Snyder, Harris, et al.,
1991; Snyder et al., 2000b). In that light, that a modest relationship exists
between social desirability and other positive constructs such as hope and
optimism is not surprising. On the other hand, the public and private selfconsciousness subscale scores from the Self-Consciousness Scale (Fenigstein,
OPTIMISM AND HOPE CONSTRUCTS

109

Scheier, & Buss, 1975) yielded nonsignificant correlations of .06 and -.03,
respectively, with the Hope Scale (Snyder, Harris, et al., 1991),and correlations of -.04 and -.05 with the LOT (Scheier & Carver, 1985). These
latter findings, lending discriminant validity support to each scale, were
predicted on the grounds that there was no theoretical reason to believe
that the optimism and hope constructs should be related to consciousness
about oneself.
The aforementioned results demonstrate the convergent and discriminant validity of both the Hope Scale and the LOT, yet they tell little about
the differences between these two indices. In the following section we
attempt to illustrate these differences more clearly.

Discriminant Utility
One reason for examining the correlations between a new and an
existing scale is to determine whether there is a need for a new measure or
whether it is merely replicating already existent indices. Although positive
correlations between scales measuring similar constructs are important, ideally the new scale should account for additional predictive variance in
relation to scores produced by these related constructs. In other words, there
should be significant and unique variance explained by the new scale beyond
that variance accounted for by the existent measures. Previously, we described this as discriminant utility (Snyder, Harris, et al., 1991).
There has been some controversy in this field of research about the
possibility that many of the relatively new concepts proposed to explain
and predict behaviors are, in reality, tapping neuroticism (Smith et al.,
1989; Smith & Rhodewalt, 1991; Watson & Pennebaker, 1989). In this
regard, Smith and his colleagues (Smith et al., 1989; Smith & Rhodewalt,
1991) have suggested that optimism, as measured by the LOT, is one construct that has a problem of incomplete construct validation (Smith &
Rhodewalt, 1991,p. 743). They argued that traditional procedures for evaluating convergent and discriminant validity require more stringent application than those used by Scheier and Carver (1985, 1987). Citing Campbell
and Fiske (1959), Smith and Rhodewalt (1991) maintained that
convergent validity is demonstrated by a large correlation between two
measures of a given trait, while discriminant validity is demonstrated
by correlations between two measures of separate traits that are smaller
than the correlations between the measures of a single trait. (p. 743)
In three samples, Smith and his colleagues (1989) found the LOT to
be significantly correlated at .56 with the GESS (Fibel & Hale, 1978) and
correlated at -.62 with measures of neuroticism (Eysenck & Eysenck, 1964)
and negative affect (Watson & Clark, 1984). Citing these results as clearly

110

SNYDER ET AL.

indicating a lack of discriminant validity, they contended that it is possible

that the LOT is simply an inversely scored measure of neuroticism (Smith
& Rhodewalt, 1991, p. 744). They went on to report that significant correlations between coping and the LOT were largely eliminated when neuroticism
was controlled through partial correlation (a statistical procedure for removing the common shared variance that a third variable displays in regard to
the relationship of two other variables); conversely, controlling for optimism
did not eliminate the significant relationships between neuroticism and
coping.
Scheier and Carver (1992) responded to this criticism with two counterarguments. First, they pointed out that the domain-specific focus of the
GESS (Fibel & Hale, 1978) makes it less than optimal as an alternative
measure to the LOT, which focuses on generalized optimism. Their second
response follows:
It is important to keep in mind that neuroticism is a multifaceted
construct which consists partly (though not entirely) of pessimism. Its
broad scope means that it confounds pessimism with other qualities,
such as emotional ability and worry. Thus, to ask whether an effect of
pessimism is really an effect of neuroticism begs the question of whether
all facets of neuroticism are important in producing the effect, or only
that part of neuroticism which is pessimism. (Scheier & Carver, 1992,
p. 216)

They then make the same argument regarding trait anxiety. In an

attempt to discount the notion that the effects of optimism or pessimism
can be better understood in terms of neuroticism, Scheier and his colleagues
reported that they are attempting factor analyses of the items of the LOT
and traditional measures of neuroticism (see Scheier & Carver, 1992, for
more details; see also Lucas, Diener, & Suh, 1996). In later research, Scheier
et al. (1994) have provided substantiating evidence that optimism is a viable
construct that differs from neuroticism. In answer to the previous criticisms,
Scheier, Carver, and Bridges ( 1994) reestablished the convergent and discriminant validity of the LOT, demonstrating its predictive validity beyond
alternative explanations such as neuroticism, trait anxiety, self-mastery, and
self-esteem; furthermore, in the same study, they validated a revised, shorter
version of the LOT, known as the LOT-R, that exhibited psychometric
standards equaling those found for the original scale.
We share some of Scheier and Carvers concerns regarding the role
of negative affectivity as a counterexplanation. Because those people with
higher levels of hope should be less likely to have high levels of negative
affect, Hope Scale scores should correlate negatively with measures of negative affect. Indeed, in a correlational study (Holleran & Snyder, 1990),
Hope Scale scores were significantly and negatively correlated with the
OPTIMISM AND HOPE CONSTRUCTS

111

Taylor Manifest Anxiety scale (TMAS; Taylor, 1953), T = -.47, p < .001,
and the State-Trait Anxiety Inventory, trait form (STAI-T, Spielberger,
Gorsuch, & Luchene, 1970), T = -.58, p < .001. T o examine whether the
effects of hope were explained by negative affect, hierarchical regression
analyses were performed. (These hierarchical regression analyses enable one
to ascertain any unique or special predictive power that a given scale may
have over other predictor scales in regard to a target or criterion variable
of interest.) Using the Problem-Focused Coping subscale from the revised
Ways of Coping Checklist (Folkman & Lazarus, 1985) as the criterion
variable and forcing the TMAS and the STAI-T into the equation at Step
1 (R2 = .06, p < .Ol), the prediction was significantly augmented when
Hope was entered at Step 2 (R2 = .03, p < .05). When Hope Scale scores
were entered first (R2= .09, p < .001), however, the TMAS and STAI-T
entered at Step 2 yielded no additional variance. Thus, these analyses reveal
that negative affect cannot account for the relationship of hope to problemfocused coping (Snyder, Harris, et al., 1991).
Two additional studies (Sigmon & Snyder, 1990a, 1990b) tested the
discriminant utility of the Hope Scale in relation to both positive and
negative affect. In both studies, the Hope Scale correlated negatively with
the negative affect items of the Positive and Negative Affect Schedule
(PANAS; Watson, Clark, & Tellegen, 1988) as expected (T = -.18). In
one study (Sigmon &Snyder, 1990b),Hope Scale scores correlated positively
with the PANAS positive affect items (T = .30). The Active Coping and
Planning subscales of the COPE (Carver, Scheier, & Weintraub, 1989)
were administered along with the PANAS and the Hope Scale. When
active coping was used as the criterion variable in hierarchical multiple
regressions, negative affect entered at Step 1 did not yield significant results
(R2 = .Ol), whereas positive affect entered at Step 2 (R2 = .075, p < .01)
and Hope Scale scores entered at Step 3 ( R 2= .09, p < .OOl) both significantly
augmented the predictions. When the steps were reversed, Hope Scale scores
predicted active coping (R2 = .15, p < .001) at Step 1, and positive affect
augmented the prediction at Step 2 (R2 = .03, p < .05), but negative affect
entered at Step 3 did not augment the prediction. When planning was
utilized as the criterion variable, neither positive affect nor negative affect
significantly predicted planning; however, Hope Scale scores entered at Step
3 significantly augmented the prediction (R2= .032, p < .05). Again, when
the order in which the variables were entered was reversed, neither positive
nor negative affect augmented the significant prediction by hope. Thus, there
is support for the predictive power of Hope Scale scores over affect measures.
Sigmon and Snyder (1990a) examined the relationship between hope
and overall well-being. Study participants completed the Hope Scale, the
PANAS, the STAI-T, and the Mental Health Inventory (Veit & Ware,
1983).Using hierarchical regression analyses with overall well-being as the

112

SNYDER ET AL.

criterion variable, Hope Scale scores significantly augmented the prediction

even when it was entered at Step 4 (R2 = .01, p < .05) after entering anxiety,
negative affect, and positive affect at Steps 1, 2, and 3, respectively. These
results indicate that hope accounts for unique variance (albeit very small)
over and above that of negative affect and positive affect in overall wellbeing.
In yet another study, we were interested in the predictive capabilities
of hope in relation to depression (see Snyder, Lapointe, Crowson, & Early,
1998). In two experiments, research participants first completed the Hope
Scale and the Beck Depression Inventory (BDI; Beck et al., 1961). We
then examined the preferences of these research participants for listening
to short, recorded statements that were either affirming (I can do this)
or disaffirming (I cant do this) in content. This preference choice was
accomplished by having the research participant move a toggle switch that
activated either the affirming or disaffirming statements (received auditorally
via headphones); accordingly, we could measure the percentage of time that
people spent listening to each tape. Results revealed that the Hope Scale
scores significantly predicted the choice of listening to more affirming than
disaffirming statements, and this effect remained when the shared variance
related to depression was removed by way of statistical procedures; conversely, however, depression scores did not predict the choice of these tapes
when the variance related to Hope Scale scores was removed. Thus, hopeful
thoughts proved more powerful than depressive feelings in guiding preferences for affirming information. In the degree to which the ongoing selftalk of people may guide the actual conduct of their lives, it appears that
hope may be related uniquely to the choice of more positive content for
such internal dialogues.
The studies described previously provide evidence that negative and
positive affectivity do not provide viable alternative explanations for the
relationship between the Hope Scale and well-being, active coping, or
planning. One additional research finding sheds light on the discriminant
utility of the Hope Scale in relation to the LOT. As noted earlier, the LOT
correlated positively with the Hope Scale in a study conducted by Holleran
and Snyder (1990). In that study, participants also completed the ProblemFocused Coping index of the Revised Ways of Coping Checklist (Folkman
& Lazarus, 1985) along with the LOT, the GESS, and the Hope Scale.
When using a hierarchical regression with problem-focused coping as the
criterion variable, our results suggested that Hope Scale scores were a somewhat better predictor than were either the GESS or the LOT scores. When
the LOT was forced into the equation at Step 1, R2 = .04,
p <.05; when
the GESS was forced into the equation at Step 2, it augmented the prediction
(R2 = .04, p < .05); finally, when Hope Scale scores were forced into the
equation at Step 3, they augmented the prediction further (R2 = .03,
OPTIMISM AND HOPE CONSTRUCTS

if3

p < .05). Conversely, in the same study when Hope Scale scores were forced
into the equation at Step 1, R2 = .085, p < .001; subsequently, neither the
LOT entered at Step 2 (R2 = .005) nor the GESS entered at Step 3 (R2 =
.015) augmented the prediction (Holleran & Snyder, 1990).Overall, therefore, these results indicate that the Hope Scale was able to predict uniquely
the problem-focused coping beyond the positive expectancies as tapped by
the LOT and the GESS.
These results, taken together, suggest that the two-component aspect
of hope theory, as well as the combining of the agency and pathways
components in the measurement instrument provide an increase in the
ability to understand and predict coping markers beyond that which can
be obtained through the LOT scores. Obviously, other studies are needed
to explore the relative predictive utility of the Hope Scale and the LOT in
relation to a variety of outcome measures. Given the theoretical relationship
between the LOT and the agency subscale component of the Hope Scale,
we would expect relatively strong relationships between these indices in
predicting outcomes. When the pathways component scores are added to
the agency component to yield the full Hope Scale, however, we expect
that the Hope Scale may provide enhanced variance (i.e., discriminant
utility) in relation to LOT scores in predicting a variety of outcome markers.

POSITIVE SEQUELAE OF HOPE

Since Snyder (1989) first proposed a new model of hope roughly a
decade ago, studies have found a positive relationship among hope and
benefits related to psychological adjustment, achievement, problem solving,
and health-related outcomes. In the following section we discuss each of
these issues.

Psychological Adjustment
Hope is related to psychological adjustment in several ways. Research
has shown that there is a positive correlation between hope and belief in
ones capabilities, as well as between hope and feelings of self-worth. In one
study, hope was positively correlated with childrens perceptions of scholastic
competence (rs = .35-S9), social acceptance (rs = .23-.43), athletic ability
(rs = .26-.35), and physical appearance (rs = .OO-.46; Snyder et al., 1997).
Therefore, hopeful thinking in children is related to their beliefs about
having the necessary abilities to accomplish specified goals. Hope also relates
to the presence of positive feelings toward oneself. Hopeful thinking is
associated with an increased feeling of self-worth in children (rs = .23-55;
Snyder et al., 1997), as well as adults (rs = .35-.75; Curry, Snyder, Cook,

114

SNYDER ET AL..

Ruby, & Rehm, 1997; Snyder et al., 1996). In adults, hope has a significant
positive correlation with self-esteem (Gibb, 1990). Further, higher hope is
associated with lower levels of depression in children (Snyder, Hoza, et al.,
1997). This latter relationship may be due, in part, to the different attributional styles of individuals with high and low hope scores (see chap. 3 in
this volume on attributional styles). The pessimistic attributional style is
one in which a person makes more internal, stable, and global explanations
for bad events, and external, stable, and global explanations for good events
(Kaslow, Tanenbaum, & Seligman, 1978). Children who score high on
hope scales tend to have internal, stable, and global attributions for positive
events, and external, stable, and global attributions for negative events
(Snyder, Hoza, et al., 1997).
Thus, hope appears to have an inverse relationship with the typical
pessimistic attributional style. That is to say, children with higher hope
scores are less likely to make characteristically pessimistic explanations for
their bad outcomes (e.g., Its my fault, its always my fault, and its this
way in all parts of my life). Other research supports this finding, showing
a significant inverse relationship between the BDI scores (Beck et al., 1961)
and hope ( r = -.42; Gibb, 1990).
Positive thinking and affect also have been found to relate to hope.
Research has shown that hope, as measured by the State Hope Scale, is
positively related to state measures of positive affect (rs = 55-.65) and
negatively related to state measures of negative affect (rs = -.47- - S O ;
Snyder, 1996). In addition, higher hope scores are related to ongoing states
involving more positive thinking. The State Hope Scale correlated positively
with the ratings of positive thoughts and negatively with the ratings of
negative thoughts. Together, this evidence suggests that individuals whose
scoring indicates higher hope both feel and think more positively on a dayto-day basis than do other individuals. This finding is further corroborated
by evidence indicating that college students scoring high on hope measures
as compared to those scoring low report feeling more inspired, confident,
energized, eager, and challenged by their goals (Snyder, Harris, et al., 1991).
Last, scores of higher hope have been described as becoming especially
important to people experiencing loss of a loved one, job, and so forth (see
Snyder, 1996; Snyder, 1998; Snyder, in press). Indeed, hope appears to be
critical for goals involving successful interpersonal relationships (see Snyder,
Cheavens, et al., 1997).
Achievement
Individuals possessing high hope also have increased performances in
the achievement arenas. Hope has related positively to achievement in
children even after variance associated with self-worth has been controlled
OPTIMISM AND HOPE CONSTRUCTS

115

statistically (Snyder, Hoza, et al., 1997). Additionally, college students with

high-as compared to low-hope indicators demonstrate increased success
in the academic realm (Snyder, Wiklund, & Cheavens, 1999). In a longitudinal study spanning 6 years, Hope Scale scores taken at the beginning of
college predicted both subsequent cumulative grade point average and graduation status, as well as whether a student was dismissed from school in poor
academic standing. The relationships between hope and both grade point
average and graduation status remained after partialling out the variance
due to ACT scores, suggesting that high hope is related to collegiate
achievement beyond ones previous high school knowledge base. Previous
studies with children (Snyder, Hoza, et al., 1997) and adults (Snyder, Harris,
et al., 1991) have shown that hope exhibits very small and statistically
nonsignificant (TS < .lo) relationships to scores on standardized tests of
intelligence.
Achievement in athletics also appears to be related to levels of hope.
Curry et al. (1997) found that the combination of dispositional and state
Hope Scale scores accounted for 56% of the variance related to track
meet performance for female collegiate athletes. Furthermore, measures of
confidence, self-esteem, and mood did not significantly augment the prediction of performance. Results also showed that Hope Scale scores provided
additional predictive information about performance beyond natural ability,
as rated by coaches. In summary, hope appears to be an independent factor
with predictive utility in the achievement realms involving academics
and sports.
Problem Solving
Research also supports the contention that hope is positively related
to problem solving. These findings should come as no surprise given that
the pathways component is a major part of hope theory. In studies designed
to validate the dispositional Hope Scale, a number of findings emerged that
support this contention (see Snyder, Harris, et al., 1991, for a review). As
previously noted, Holleran and Snyder (1990) found that hope is a significant, unique predictor of problem-focused coping (i.e., the Problem-Focused
Coping subscale of the revised Ways of Coping Checklist; Folkman &
Lazarus, 1985) when controlling for optimism (as measured by the LOT)
and negative affectivity (PANAS; Watson et al., 1988), and that neither
optimism nor negative affectivity added any significant unique predictive
value above that of hope. Following from hope theory, this finding is taken
as an indication that the pathways component of hope reflects problemsolving thinking in regard to goal achievement.
In another study directly related to hope and coping (Sigmon & Snyder,
1990b), hope predicted both active coping (i.e., taking steps to overcome

116

SNYDER ET AL

effects of stressors in a direct and effortful manner) and planning (i.e.,

producing strategies to cope with stressors), as measured by the Active
Coping and Planning subscales of the COPE (Carver et al., 1989) when
controlling for both positive and negative affectivity (PANAS). Thus, the
effects of hope on coping and problem solving cannot be explained by the
mediating effects of affectivity or optimism. As suggested earlier, we propose
that it is the explicit pathways component of hope theory that helps to
contribute the additional predictive power in explaining variance related
to problem solving.
Health-Related Concerns
Research suggests that hope confers advantages to people in coping
with illness-related problems. There are three reported studies relating hope
to health, all of which suggest that there are benefits accruing for those
with higher levels of hope. First, in the recovery process following spinal
cord injury, higher levels of hope were related to better coping and to less
depression (Elliott, Witty, Herrick, & Hoffman, 1991). Second, in a study
of nurses working in a high-pressure hospital unit, those with higher levels
of hope were less likely to report burnout (Sherwin et al., 1992). Third,
higher hope levels in adolescent burn survivors correlated with engaging
in fewer behaviors that would undermine recovery (Barnum, Snyder, Rapoff,
Mani, & Thompson, 1998).These bum survivors also had a greater number
of successful interactions with others, and a higher level of global self-worth.
Hope also appears to play an important role in coping with potentially
life-threatening diseases. In a recent research project, college women scoring
high on hope levels demonstrated greater knowledge about cancer and
increased hope-specific coping responses to an imagined task of coping with
cancer than did college women scoring low on hope levels, even when
controlling for academic achievement, experience with cancer (in family
or friends), and negative affectivity (Irving, Snyder, & Crowson, 1998).
This relationship seemed to prevail as the research participants projected
their reactions throughout the imagined course of the disease. Thus, hope
may facilitate adjustment across the various stages of coping with illnesses.
A growing body of research is illuminating the role of finding benefits
in adversity as a means of coping with health-related problems (see Affleck
& Tennen, 1996; Tennen & Affleck, 1999, for reviews). Hope appears to
play an important role in such benefit finding. In their review article, Affleck
and Tennen (1996) discussed findings from an ongoing study concerning
the psychological adjustment of individuals with fibromyalgia, a chronic
disease that involves high levels of pain. Participants with high hope
levels reported finding more benefits from their experience with the disease
than did those with low hope levels, even when researchers controlled for
OPTIMJSM AND HOPE CONSTRUCTS

117

optimism. The authors suggested that hope is an important factor in making

positive appraisals of adverse situations. Affleck and Tennen proposed that
reminding oneself about the benefits one accrues through the experience
of adversity serves as an active coping strategy. In this study involving
patients with fibromyalgia, Affleck and Tennen found that the correlation
between benefit finding and benefit reminding was not significant when
the pathways component of hope was partialled out. Thus, the hope component of planning routes to goals based on finding benefits in adversity plays
a crucial role in the use of benefit reminding as a coping strategy. Again,
this latter result highlights the importance of adding the pathways thinking
component to that of agency in order fully to tap hope. Overall, hopeful
thought appears to play a useful role in coping with health-related matters.
Summary of Positive Sequelae
The preceding research demonstrates the benefits of higher levels of
hope for children and adults and how hope relates to coping in a variety
of arenas, including academics, sports, problem-solving ability, and health.
Moreover, in comparison to optimism, as well as negative affectivity, positive
affectivity, locus of control, mood, depression, athletic aptitude, cognitive
or intellectual aptitude, and helplessness, the hope scales have provided
unique and augmenting variance in predicting outcomes. Thus, hope appears
to be a good predictor of positive adjustment and coping, and this predictive
utility cannot be explained by other individual differences indices that would
reflect counterexplanations (see Snyder, Cheavens, & Michael, 1999, for
a review).

CONCLUSION
The major difference in the models of optimism and hope is that
the former emphasizes agentic goal-related thinking, whereas the latter
emphasizes the mutual contribution of agentic and pathways goal-directed
thoughts. Although we point out some differences in the optimism and
hope theories, there are many important shared aspects of these two theories,
thus the subtitle, Variants on a Positive Expectancy Theme.

REFERENCES
Affleck, G., & Tennen, H. (1996). Construing benefits from adversity: Adaptational
significance and dispositional underpinnings. Journal of Personality, 64,899-

922.

118

SNYDER ET AL.

Anderson, J. R. (1988). The rok of hope in goal-setting, expectancy about success, and
coping. Unpublished doctoral dissertation, University of Kansas, Lawrence.
Babyak, M. A., Snyder, C. R., & Yoshinoba, L. (1993). Psychometric properties
of the Hope Scale: A confirmatory factor analysis.Journalof Research in Personality, 27, 154-159.
Barnum, D. D., Snyder, C. R., Rapoff, M. A., Mani, M. M., & Thompson, R.
(1998). Hope and social support in the psychological adjustment of children
who have survived burn injuries and matched controls. Childrens Health Care,
27, 15-30.
Beck, A. T., Ward, C. H., Mendelson, M., Mock, J., & Erbaugh, J. (1961). An
inventory for measuring depression. Archives of General Psychology, 4, 53-63.
Beck, A. T., Weissman, A., Lester, D., & Trexler, L. (1974). The measurement of
pessimism:The HopelessnessScale.Journal of Consultingand Clinical Psychology,
42, 861-865.
Campbell, D. T., & Fiske, D. W. (1959). Convergent and discriminant validation
by the multitrait-multimethod matrix. Psychological Bulktin, 56, 81-105.
Carver, C. S., & Scheier, M. F. (1999). Optimism. In C. R. Snyder (Ed.), Coping:
The psychology of what works (pp. 182-204). New York: Oxford.
Carver, C. S., Scheier, M. F., & Weintraub, J. K. (1989). Assessing coping strategies:
A theoretically based approach. Journal of Personality and Social Psychology,
56, 267-283.
Crowne, D. P., & Marlowe, D. (1960). A new scale of social desirability independent
of psychopathology.Journal of Consulting and Clinical Psychology, 24,349-354.
Curry, L. A., Snyder, C. R., Cook, D. L., Ruby, B. C., & Rehm, M. (1997). The
role of hope in academic and sport achievement. Journal of Personality and
Social Psychology, 73, 1257-1267.
Elliott, T. R., Witty, T. E., Herrick, S., & Hoffman, J. T. (1991). Negotiating
reality after physical loss: Hope, depression, and disability.Journal of Personality
and Social Psychology, 61, 608-613.
Eysenck, H. J., & Eysenck, S. B. G. (1964). Manual for the Eysenck Personality
Inventory. London: University Press.
Fenigstein, A., Scheier, M. F., & Buss, A. H. (1975). Public and private selfconsciousness:Assessment and theory. Journal of Consultingand Clinical Psychol0 0 , 43, 522-527.
Fibel, B., & Hale, W. D. (1978). The Generalized Expectancy for Success ScaleA new measure. Journal of Consulting and Clinical Psychology, 46, 924-93 1.
Folkman, S., & Lazarus, R. S. (1985). If it changes it must be a process: Study of
emotion and coping during three stages of college examination. Journal of
Personality and Social Psychology, 48, 150-170.
Fontaine, K. R., Manstead, A. S. R., & Wagner, H. (1993). Optimism, perceived
control over stress, and coping. European Journal of Personality, 7, 267-281.
Frank, J. D. (1975). The faith that heals. The Johns Hopkins Medical Journal, 137,
127-131.
OPTIMISM AND HOPE CONSTRUCTS

119

Frankl, V. (1963). Mans search for meaning. New York: Washington Square Press.
Friedman, L. C., Nelson, D. V., Baer, P. E., Lane, M., Smith, F. E., & Dworkin,
R. J. (1992). The relationship of dispositional optimism, daily life stress, and
domestic environment to coping methods used by cancer patients. Journal of
Behavioral Medicine, I 5, 127- 141.
Gibb, J. (1990). The Hope Scale revisited: Further validation of a measure of individual
differences in the hope motive. Unpublished masters thesis, University of Illinois
at UrbanaChampaign.
Harney, P. (1989). The Hope Scak: Exploration of construct validity and its influence
on health. Unpublished masters thesis, University of Kansas, Lawrence.
Holleran, S., & Snyder, C. R. (1990). Discriminant and convergent vdidation of the
Hope Scak. Unpublished manuscript, University of Kansas, Lawrence.
Irving, L. M., Snyder, C. R., & Crowson, Jr., J. J. (1998). Hope and coping with
cancer by college women. Journal of Personality, 66, 195-214.
Kaslow, N. J., Tanenbaum, R. L., & Seligman, M. E. P. (1978). The KASTAN-R:
A childrens attributional styk questionnaire (KASTAN-R-CASQ). Unpublished
manuscript, University of Pennsylvania, Philadelphia.
Lucas, R. E., Diener, E., & Suh, E. (1996). Discriminant validity of well-meaning
measures. Journal of Personality and Social Psychology, 73, 616-628.
Magaletta, P. R., & Oliver, J. M. (1999). The hope construct, will and ways: Their
relative relations with self-efficacy, optimism, and general well-being. Journal
of Clinical Psychology, 55, 539-55 1.
Menninger, K. (1959). The academic lecture on hope. The American Journal of
P~ychiamy,1 09, 48 1-49 1.
Pelletier, K. R. (1977). Holistic medicine: From stress to optimum health. New York:
Delacorte Press/Seymore Lawrence.
Robinson-Whelen, S., Kim, C., MacCallum, R. C., & Kiecolt-Glaser, J. K. (1997).
Distinguishing optimism from pessimism in older adults: Is it more important
to be optimistic or not be pessimistic?Journal ofPersonality and Social Psychology,
73, 1345-1353.
Rosenberg, M. (1965). Society and adolescent self-image. Princeton, N]: Princeton
University Press.
Scheier, M. F., & Carver, C. S. (1985). Optimism, coping, and health: Assessment
and implications of generalized outcome expectancies. Health Psychology, 4,
219-247.
Scheier, M. F., &. Carver, C. S. (1987). Dispositional optimism and physical wellbeing: The influence of generalized outcome expectancies. Journal of Personality,
55, 169-210.
Scheier, M. F., & Carver, C. S. (1992). Effects of optimism on psychological and
physical well-being: Theoretical and empirical update. Cognitive Therapy and
Research, 16, 201-228.
Scheier, M. F., Carver, C. S., & Bridges, M. W. (1994). Distinguishing optimism
from neuroticism (and trait anxiety, self mastery, and self-esteem): A reevalua-

120

SNYDER ET AL

tion of the Life Orientation Test. journal of Personality and Social Psychology,
67, 1063-1078.
Scheier, M. F., Weintraub, J. K., & Carver, C. S. (1986). Coping with stress:
Divergent strategies of optimists and pessimists. Journalof Personality and Social
Psychology, 51, 1257-1264.
Schmale, A. H. (1972). Giving up as a final common pathway to changes in health.
Advances in Psychosomatic Medicine, 28, 7 14-721.
Sherwin, E. D., Elliott, T. R., Rybarczyk, B. D., Frank, R. G., Hanson, S., &
Hoffman, J. (1992). Negotiating the reality of care giving: Hope, burnout, and
nursing. Journal of Social and Clinical Psychology, 11, 129-139.
Siegel, B. D. (1986). Love, medicine, and miracles: Lessons learned about self-healing
from a surgeons experience with patients. New York: Harper & Row.
Sigmon, S. T., & Snyder, C. R. (1990a). The idpendent contributions ofpositive and
negative affect and hope in predictingpsychologicalhealth. Unpublished manuscript,
University of Kansas, Department of Psychology, Lawrence.
Sigmon, S. T., 6r Snyder, C. R. (1990b). Positive and negative affect as a counterexplanation for the relationship between hope and coping strategies. Unpublished
manuscript, University of Kansas, Department of Psychology, Lawrence.
Simonton, 0. C., Matthew-Simonton, S., & Creighton, J. L. (1978). Getting well
again. New York: Bantam Books.
Smith, T. W., Pope, M. K., Rhodewalt, F., & Poulton, J. L. (1989). Optimism,
neuroticism, coping, and symptom reports: An alternative interpretation of the
Life Orientation Test. Journal of Personality and Social Psychology, 56,640-648.
Smith, T. W., & Rhodewalt, F. (1991). Methodological challenges at the social/
clinical interface. In C. R. Snyder & D. R. Forsyth (Eds.), Handbook of social
and clinical psychology: The health perspective (pp. 739-756). Elmsford, NY:
Pergamon Press.
Snyder, C. R. (1989). Reality negotiation: From excuses to hope and beyond. journal
of Social and Clinical Psychology, 8, 130-157.
Snyder, C. R. (1994). The psychology of hope: You can get there from here. New York:
Free Press.
Snyder, C. R. (1996). To hope, to lose, and hope again. Journal of Personal and
Interpersonal Loss, 1, 1-16.
Snyder, C. R. (1998). A case for hope in pain, loss, and suffering. In J. H. Harvey,
J. Omarzu, & E. Miller (Eds.), Perspectives on loss: A sourcebook (pp. 63-79).
Washington, DC: Taylor & Francis.
Snyder, C. R. (2000a). Genesis: Birth and growth of hope. In C. R. Snyder (Ed.),
Handbook of hope: Theory, measures, and applications (pp. 25-54). San Diego,
CA: Academic Press.
Snyder, C. R. (2000b). Hypothesis: There is hope. In C. R. Snyder (Ed.), Handbook
of hope: Theory, measures, and applications (pp. 3-21). San Diego, CA: Academic Press.
OPTIMISM AND HOPE CONSTRUCTS

121

Snyder, C. R. (in press). The hope mandala: Coping with the loss of a loved one.
In J. Gillham (Ed.), Optimism and hope. Radnor, PA: Templeton Foundation.
Snyder, C. R., Cheavens, J., & Michael, S. T. (1999). Hoping. In C. R. Snyder
(Ed.), Coping: The psychology of what works (pp. 205-231). New York: Oxford
University Press.
Snyder, C. R., Cheavens, J., & Sympson, S. C. (1997). Hope: An individual
motive for social commerce. Group Dynamics: Theory, Research, and Practice,
1 , 107-118.
Snyder, C. R., Harris, C., Anderson, J. R., Holleran, S. A., Irving, L. M., Sigmon,
S. T., Yoshinobu, L., Gibb, J., Langelle, C., & Harney, P. (1991). The will
and the ways: Development and validation of an individual-differencesmeasure
of hope. Journal of Personality and Social Psychology, 60, 570-585.
Snyder, C. R., Hoza, B., Pelham, W. E., Rapoff, M., Ware, L., Danovsky, M.,
Highberger, L., Rubinstein, H., & Stahl, K. J. (1997). The development and
validation of the childrenshope scale.Journal of Pediatric Psychology, 22,399421.
Snyder, C. R., Irving, L., & Anderson, J. R. (1991). Hope and health: Measuring
the will and the ways. In C. R. Snyder & D. R. Forsyth (Eds.), Handbook of
social and clinical psychology: The health perspective (pp. 285-305). Elmsford,
NY: Pergamon Press.
Snyder, C. R., Lapointe, A. B., Crowson, Jr., J. J., & Early, S. (1998). Preferences
of high- and low-hope people for self-referential input. Cognition and Emotion,
12, 807-823.
Snyder, C. R., & McCullough, M. (2000). A positive psychology field of dreams:
If you build it, they will come. . . Journal of Social and Clinical Psychology,
19, 151-160.
Snyder, C. R., McDermott, D., Cook, W., & Rapoff, M. (1997). Hope for the
journey. Boulder, CO: Westview.
Snyder, C. R., Michael, S., & Cheavens, J. (1999). Hope as a psychotherapeutic
foundation for nonspecific factors, placebos, and expectancies. In M. A. Huble,
B. Duncan, &S. Miller (Eds.), Heartandsoulofchange (pp. 179-200). Washing
ton, DC: American Psychological Association.
Snyder, C. R., Sympson, S. C., Ybasco, F. C., Borders, T. F., Babyak, M. A., &
Higgins, R. L. (1996). Development and validation of the State Hope Scale.
Journal of Personality and Social Psychology, 70, 321-335.
Snyder, C. R., Wiklund, C., & Cheavens, J. (1999, August). Hope and success in
college. Paper presented at the American Psychological Association, Boston.
Spielberger, C. D., Gorsuch, R. L., & Luchene, R. E. (1970). The State-Trait Anxiety
Inventory. Palo Alto, CA: Consulting Psychologists Press.
Strutton, D., & Lumpkin, J. (1992). Relationship between optimism and coping
strategies in the work environment. Psychological Reports, 71, 1179-1 186.
Sympson, S. C. (1993). Construction and validation of a state hope measure: A month
in the lives of college students. Unpublished masters thesis, University of Kansas, Lawrence.

122

SNYDER ET AL.

Taylor, J. A. (1953). A personality scale of manifest anxiety. Journal of Abnormal

and Social P ~ y ~ h o l48,
~ g ~285-290.
,
Tennen, H., & Affleck, G. (1999). Finding benefits in adversity. In C. R. Snyder
(Ed.), Coping: The psychology of what works (pp. 279-304). New York: Oxford Press.
Veit, C. V., & Ware, J. E. (1983). The structure of psychological stress and wellbeing in general populations. Journal of Consulting and Clinical Psychology,
51, 730-742.
Watson, D., & Clark, L. A. (1984). Negative affectivity:The disposition to experience negative emotional states. Psychological Bulletin, 96, 465-490.
Watson, D., Clark, L. A., & Tellegen, A. (1988). Development and validation of
brief measures of positive and negative affect: The PANAS scales. Journal of
Personality and Social Psychology, 54, 1063-1070.
Watson, D., & Pennebaker, J. W. (1989). Health complaints, stress, and distress:
Exploring the central role of negative affectivity. Psychological Review, 96, 233253.
Yoshinobu, L. (1989). Construct validation of the Hope Scale: Agency and pathways
components. Unpublished masters thesis, University of Kansas, Lawrence.

OPTIMISM A N D HOPE CONSTRUCTS

123

APPENDIX 5.1: THE ADULT TRAIT HOPE SCALE

Directions: Read each item carefully. Using the scale shown below, please
select the number that best describes you and put that number in the
blank provided.

1 = Definitely False

2 = Mostly False
3 = Somewhat False

4 = Slightly False
5

= Slightly True
6 = Somewhat True
7 = Mostly True
8 = Definitely True

____
~

_.~

~- ~

1. I can think of many ways to get out of a jam.

2. I energetically pursue my goals.
3. I feel tired most of the time.

4. There are lots of ways around

7.

8.
9.

____

any problem.

5. I am easily downed in an argument.

6. I can think of many ways to get the things in life that are important

10.
11.
12.

to me.
I worry about my health.
Even when others get discouraged, I know I can find a way to
solve the problem.
My past experiences have prepared me well for my future.
Ive been pretty successful in life.
I usually find myself worrying about something.
I meet the goals that I set for myself.

Note. Administers of the scale should refer to it as The Future Scale. The
agency subscale score is derived by summing item nos. 2,9, 10, and 12; the
pathways subscale score is derived by adding item nos. 1,4, 6, and 8. Items
3, 5 , 7, and 11 are fillers and are ignored for scoring purposes. The total
Hope Scale score is derived by summing the four agency and the four
pathways items. From The Will and the Ways: Development and Validation
of an Individual Differences Measure of Hope, by C. R. Snyder, C. Harris,
et al., 1991, Journal of Personality and Social Psychology, 60, p. 585. Copyright
1991 by the American Psychological Association. Reprinted with permission.

124

SNYDER ET AL.

APPENDIX 5.2: THE ADULT STATE HOPE SCALE

Directions: Read each item carefully. Using the scale shown below, please
select the number that best describes how you think about yourself right now
and put that number in the blank before each sentence. Please take a few
moments to focus on yourself and what is going on in your life at this moment.
Once you have this here and now mindset, go ahead and answer each
item according to the following scale:

1 = Definitely False
2 = Mostly False
3 = Somewhat False

4 = Slightly False

5 = Slightly True
6 = Somewhat True
7 = Mostly True
8 = Definitely True
-

1. If I should find myself in a jam, I could think of many ways to get

out of it.

2. At the present time, I am energetically pursuing my goals.

3. There are lots of ways around any problem that I am facing now.
4. Right now, I see myself as being pretty successful.
5 . I can think of many ways to reach my current goals.
6. At this time, I am meeting the goals that I have set for myself.
Note. The agency subscale score is derived by summing the three evennumbered items; the pathways subscale score is derived by adding the three
odd-numbered items. The total State Hope Scale score is derived by summing
the three agency and the three pathways items. Scores can range from a
low of 6 to a high of 48. Administers of the State Hope Scale should refer
to it as the Goals scale for the present. From Development and Validation
of the State Hope Scale, by C. R. Snyder, S. C. Sympson, et al., 1996,
Journal of Personality and Social Psychology, 65, p. 335. Copyright 1996 by
the American Psychological Association. Reprinted with permission.

OPTIMISM AND HOPE CONSTRUCTS

125

OPTIMISM AND
PHYSICAL WELL*BEING
CHRISTOPHER PETERSON AND LISA M. BOSS10

In 1991, we published Health and Optimism, a book that described

what at the time was characterized as the new psychology of optimism. We
focused on the implications of emerging theory and research for understanding physical health. What was our conclusion at that time? Thinking good
is linked to feeling well, an assertion consistent with many peoples common sense, yet until recently lacking solid scientific support. Since then,
psychologists have continued to study the association between optimism
and health. The original findings have been repeated and extended, and
we now know more about this intriguing topic. Our purpose here is to take
another look at the psychology of optimism and what it tells us about
physical well-being.
As in our original discussion, we inform this chapter with what the
evidence actually shows. We explain why it can be difficult to arrive at
definitive conclusions about psychological influences on physical health,
and thus, why the relevant research literature deserves to be taken seriously.
We draw boundaries around these influences. For example, optimism is one

I27

of the determinants of physical health, but not the only one, and in some
cases, not even a major one. We note what still remains to be learned about
optimism and health. Finally, we stress that optimism per se does not have
an automatic effect on health. If it did, smiley buttons could replace antibiotics, seat belts, and latex condoms (Peterson & Stunkard, 1989). Positive
thinking is powerful only when it sets into motion a complex cascade of
processes-biological, psychological, and social-that themselves lead to
good health.

ESTABLISHING THE OPTIMISM-HEALTH LINK

Before we describe the research in support of the link between optimism
and physical well-being, we need to address definition, measurement, and
research design. None of these matters is cut and dried, and the validity of
our conclusions rests on the convergence of studies that resolved these issues
in somewhat different ways.

What Is Health?
Physical health and illness prove as difficult to define and thus to
measure as the most fuzzy psychological constructs. Our belief is that no
single and simple definition of physical well-being is adequate. Rather, a
number of factors count toward our judgment of whether an individual is
ill, but most are neither necessary nor sufficient:

..
..
..

General complaints about feeling ill

Specific symptoms such as shortness of breath
Identifiable damage to the body
Presence of germs
Impairment of daily activities
A short-as opposed to a long-life

These factors may disagree with one another. Someone might feel fine but
harbor all sorts of germs. Someone else might be free of germs but feel
poorly. Or yet a third individual might live a long but impaired life, or a
short but vigorous one.
In making sense of health psychology research, we must understand
just how physical health and illness are measured. There is a tendency,
perhaps, to regard hard measures of health and illness, that is, those based
on physical tests, as the most valid, but this bias is not warranted. Biological
measures of illness, such as the presence of germs, are no more basic than
psychological measures, such as a persons general sense of well-being or
the degree to which he or she can lead an active life.

128

PETERSON AND BOSS10

What Is Optimism?
In our review, we focus on explanatory style: how an individual habitually explains the causes of bad events (Peterson & Seligman, 1984). This
cognitive personality variable is sometimes described as pessimistic (when
bad events are explained with stable, global, and internal causes) versus
optimistic (when bad events are explained with unstable, specific, and external causes). Explanatory style is a distant influence on the specific expectation that ones behaviors are related (or not) to important outcomes (Abramson, Seligman, & Teasdale, 1978).A pessimistic explanatory style encourages
individuals to believe that they are helpless, that nothing they do matters,
whereas an optimistic explanatory style encourages individuals to believe
that their behaviors do affect outcomes. The expectation of helplessness
versus efficacy is the theoretically critical link between explanatory style
and outcomes such as physical well-being, but most research in this tradition
does not measure this intervening variable.
Explanatory style took off as its own line of research when measures
of this individual difference were developed. Several such measures now
exist (Reivich, 1995), but a caveat is in order. The relationships of the
different explanatory style measures to one another have not been extensively investigated. They presumably measure the same characteristic and
should converge impressively. Studies to date show a range of convergence,
and in some cases, alternative measures are almost independent of one
another (Peterson, Bettes, & Seligman, 1985; Peterson & Park, 1997). At
the same time, the different measures of explanatory style invariably are
correlated in theoretically predicted ways with such outcomes as physical
well-being.
Taken together, these results are paradoxical, implying that measures
of explanatory style have better validity than reliability. A possible resolution
is that each measure taps explanatory style at a somewhat different level of
abstraction and distance from the actual causal attributions people offer for
events that occur. Future researchers need to use all of the available measures
of explanatory style in the same investigation, to test this speculation.

Attributional Style Questionnaire

As explained elsewhere (see chap. 3, this volume), explanatory style
is usually measured with the Attributional Style Questionnaire (ASQ; Peterson et al., 1982). Because the earliest version of the ASQ had-at bestmodest reliability (Tennen & Herzberger, 1986), researchers fell into the
practice of using composites formed by averaging different attributional
dimensions. The effect was twofold. O n the one hand, reliabilities were
typically increased because that many more items were used to estimate
someones explanatory style. But on the other, researchers were not able to
OPTIMISM AND PHYSICAL WELL-BEING

129

investigate the roles assigned by the helplessness reformulation to the specific

attributional dimensions (Carver, 1989). When researchers use lengthier
(and thus more reliable) versions of the ASQ (Peterson & Villanova, 1988),
they usually find that stability and globality are more related to health
outcomes than is internality. This pattern makes theoretical sense in that
stability and globality, as opposed to internality, directly influence the expectation of helplessness, determining when it is projected over time and
generalized across situation (Abramson, Metalsky, & Alloy, 1989).
Content Analysis of Verbatim Explanations
As noted by Gillham and colleagues (see chap. 3, this volume) a
second popular way of measuring explanatory style is a flexible content
analysis method: the Content Analysis of Verbatim Explanations (CAVE;
Peterson, Schulman, Castellon, & Seligman, 1992). Potential research participants neither able nor willing to participate in typical research can be
studied with the CAVE technique, as long as they have left behind suitable
material containing causal explanations about themselves. T o date, the
CAVE technique has been used with psychotherapy transcripts, interviews,
open-ended questionnaires, political speeches, sports stories, religious texts,
and song lyrics (see Lee & Peterson, 1997). The greatest methodological
virtue of this approach to studying explanatory style is that it allows researchers to conduct longitudinal research retrospectively. If suitable verbal material can be located from early in the lives of individuals whose long-term
fate is known, then the CAVE technique allows researchers to complete
very quickly studies that nonetheless span decades. For obvious reasons,
studies of explanatory style and physical well-being have frequently relied
on the CAVE (Peterson, Seligman, & Vaillant, 1988).
Cognates of optimism
As previous chapters in this volume have made clear, explanatory style
is not the only conceptualization of optimism in the research literature.
Accordingly, we touch on studies that look at the related constructs of
dispositional optimism: the general expectation that good events
will be plentiful (or not) in the future (Scheier & Carver,
1985); and
self-efficacy:the belief that one can (or cannot) enact a specific
behavior leading to a desired outcome (Bandura, 1986).
Dispositional optimism is usually measured with a brief self-report questionnaire called the Life Orientation Test (LOT), described earlier (see Introduction and chap. 2, this volume). Self-efficacy is usually measured by asking

130

PETERSON AND Bossro

an individual to indicate his or her confidence, using a percentage estimate,

that the behavior of concern can be performed.
A number of investigations concerned with how people perceive their
relative risk for specific illnesses also exist, and we describe some of these
studies as well. Most measure these expectations with a single item, asking
respondents to provide a percentage estimate that they will someday contract
an illness or to provide a comparison to peers (less likely, equally as likely,
more likely) that they will do so. A robust finding in this literature is that
people underestimate their risks: The average individual sees him- or herself
as below average in risk for a variety of maladies, which of course cannot
be. This phenomenon is identified as an optimistic bias (Weinstein, 1989),
and it is decried because it may lead people to neglect the basics of health
promotion and maintenance. Later in this chapter, we will discuss the
optimistic bias further because it seems to contradict our thesis that optimism
is healthy.
What Is the Optimal Research Design?
The ideal investigation of the psychological precursors of illness, including, of course, optimism and pessimism, should satisfy procedural criteria
such as these:

1. The research design must be longitudinal. Research participants

must be followed over time. Merely showing a contemporaneous association between psychological states and physical wellbeing leaves unanswered the direction of the implied effect.
Perhaps illness or health determines psychological state, or
perhaps another factor gives rise to both the psychological
state and health status.
2. The longitudinal design must span enough time for changes in health
to take place. This may require several years or even decades.
Because we are exploring new territory, we are not sure what
should be the right time frame. We must, therefore, be leery
of researchers who report no results from their studies trying
to link psychological states with health and illness. Perhaps
there really is no link to be found, and the researchers were
indeed on a wild goose chase. But perhaps the optimal time
span was not chosen.
3 . The initial health status of the research participants must be known
when the investigation begins. There is no point in finding that
sick people stay sick, while healthy people stay healthy. Many
researchers start with peoples current health status and then
try to work backward to reconstruct what they were like in
OPTIMISM AND PHYSICAL WELL-BEING

131

the first place. This is a reasonable procedure if the distant

past can be reconstructed with fidelity, but that is not always
the case. People distort their past, innocently or otherwise,
and may retrospectively make better sense of what ensued
than the facts warrant. Sick people may tell a story that rationalizes their sickness, and healthy people may rationalize their
health. When we see a story about a centenarian, the person
always attributes his or her longevity to something: eating
yogurt, drinking cranberry wine, staying busy, whatever. These
are pleasing accounts but hardly scientific explanations.
4. An adequate number of research participants must be studied.
Links between psychological states and illness, if and when
they exist, are apt to be subtle in nature and modest in strength.
Only with a large number of research participants can patterns
be discerned and potential confounds ruled out. Case studies
of single individuals, no matter how striking, can be ambiguous
precisely on this score.
5. The psychologicalstates of interest must be well-defined and measurable in a reliable and walid fashion. Researchers should avoid in
particular measures that are contaminated by a persons health
status, which means that they end up using one aspect of his
or her health to predict another aspect. The result, of course,
is that these aspects will be related but that this convergence
proves nothing. This has been a problem in stress research,
for instance, where illness is considered a stressful life event
that ends up being related to subsequent illness.
Along these lines, some critics argue that optimism measures
such as the LOT can be confounded by neuroticism, which
also influences health measures that rely on respondents selfreports (Costa & McCrae, 1987). A general inclination to
complain may produce misleading correlations between pessimism and illness. A study that independently measures and
controls for neuroticism (depression, dysphoria, whatever) is
preferable to one that does not.
6. Objectiwe measures of health and illness must be made at the time
that an inwestigation begins, during the study, and at its end. Health
and illness may seem easy for a researcher to ascertain, but
remember our previous discussion of the fuzziness of health
status. Knowing that no criterion of well-being is foolproof,
researchers nonetheless must settle on a strategy for measuring
it. As already emphasized, the various criteria of illness do not
perfectly agree. One of the best established findings in 20thcentury epidemiology is that morbidity and mortality do not

I32

PETERSON AND BOSSIO

line up perfectly once we look separately at men and women:

Women have more illnesses than men, but they also live
longer (Verbrugge, 1989).

How Should We Think About This Research?

Several general points need to be made concerning how best to explain
the influence of psychological factors on health and illness. First, Western
science has for centuries made too strict a distinction between what is
categorized as mind and what is categorized as body. However, psychological factors and biological factors are mutually influential. This mutual
influence is particularly important when our focus is on health and illness.
Second, despite the possible relevance of psychological factors to illness, we must not conclude that these are the only influences or even the
most important ones. We all carry with us vulnerabilities and resiliencies
as part of our genetic inheritance. One of the best predictors of longevity,
for example, is how long our ancestors have lived.
Third, although most psychologists believe that psychological factors
influence physical health and illness, there is considerable disagreement
about the specificity of this influence. Do certain psychological states or
traits make given illnesses more or less likely, or do they simply exert a
general influence on well-being? Over the years theorists have proposed
that angry people are at risk for heart disease, hopeless people for cancer,
and so on. However, research suggests that such highly specific links usually
do not exist (H. S. Friedman & Booth-Kewley, 1987). Still, a case can be
made for studying how psychological factors influence a particular illness.
Such a strategy holds constant many factors that potentially confound
broader investigations of general well-being, such as sociodemographic and
environmental risk factors for a specific illness, treatments, and the like
(Schulz, Bookwala, Knapp, Scheier, & Williamson, 1996).

EMPIRICAL STUDIES OF OPTIMISM AND HEALTH

To date, there exist several dozen studies that have attempted to
establish a correlation between optimism, rendered as explanatory style
(measured with the ASQ or the CAVE) or dispositional optimism (measured
with the LOT), and physical well-being. Taken together, these studies
If space permitted, we would survey research into the relatives of optimism and their link to
physical health, constructs such as health locus of control (Wallston & Wallston, 1978), hardiness
(Kobasa, 1979), sense of coherence (Antonovsky, 1985), positive illusions (Taylor, 1989), and so on.
Suffice it to say that in each line of investigation, there is empirical support for the conclusion that
positive thinking is correlated with good health, and negative thinking with poor health (Peterson,
1999).

OZTZMZSM AND PHYSICAL WELL-BEING

133

converge impressively (see reviews by Carver, Spencer, & Scheier, 1998;

Michaels, Michaels, & Peterson, 1997; Peterson, 1995,1999,2000; Peterson
& Bossio, 1991, 1993; Peterson, Maier, & Seligman, 1993; Peterson &
Seligman, 1987; Scheier & Bridges, 1995; Scheier & Carver, 1985, 1987,
1992).

Confirming Studies
Optimism predicts good health measured in a number of ways, from
self-report, to physician ratings of general well-being, to doctor visits, to
survival time following a heart attack, to immunological efficiency, to successful completion of rehabilitation programs to longevity. How strong is
the association between optimism and good health? Most relevant studies
report correlation coefficients in the .20-.30 range, which are moderate in
size and typical of correlations in psychological research. Research participants included adults across the life span, some initially healthy and others
initially quite ill. Many of these studies were longitudinal, spanning mere
weeks to almost 5 decades. And at least some of these longitudinal studies
statistically controlled for initial levels of health and potential confounds
involving tendencies to complain.
Optimism apparently impacts health at a number of junctures. It can
make the initial onset of illness less likely; it can minimize the severity of
illness; it can speed recovery; and it can make relapse less likely. Most of
the relevant studies, by virtue of rather stark correlational designs, do not
allow conclusions about when health is impacted. Perhaps the cognates of
optimism affect different aspects of physical well-being. Future investigations
of a more fine-grained nature are indicated.
Studies of self-efficacyare usually more specific, focusing on how people
respond once health is compromised in a given way. The following is a
good summary of the relevant research:
The evidence taken as a whole is consistent in showing that peoples
perceptions of their own efficacy are related to different forms of health
behavior. In the realm of substance abuse, perceived self-regulatory
efficacy is a reliable predictor of who will relapse and the circumstances
of each persons first slip. Strong percepts of efficacy to manage pain
increase pain tolerance . . . [Perceived efficacy with regard to] . . . eating
and weight predicts who will succeed in overcoming eating disorders.
Recovery from the severe trauma of myocardial infarction is tremendously facilitated by the enhancement of the patients and their spouses
judgments of their physical and cardiac capabilities. And self-efficacy
to affect ones own health increases adherence to medical regimens.
. . . While specific procedures may differ for different domains, the

134

PETERSON AND BOSSIO

general strategy of assessing and enhancing self-percepts of efficacy

to affect health . . . has substantial general utility. (OLeary, 1985,
pp. 448-449)

So, the belief that one can cope appropriately with a given malady predicts
successful rehabilitation or recovery.
Disconfirming Studies
Some studies fail to confirm the optimism-health association. In some
cases, the procedural criteria discussed earlier for an optimal research design
were not met. For example, Dua and Plumer (1993) found no association
between explanatory style and the general health of nursing students, but
only 27 research participants were included. This investigation had insufficient power to detect the hypothesized correlation (Peterson, Villanova, &
Raps, 1985; Robins, 1988). For another example, Lin and Peterson (1990)
found no association between explanatory style and recovery time from
colds among mostly healthy college students, but their investigation spanned
only a few weeks. A more adequate study would have looked at patterns in
recovery from repeated episodes of minor illness over months or years.
Other studies deserve more attention because they imply that positive
thinking may at times preclude good health. For example, John Henryism
is a personality characteristic reflecting the degree to which someone believes
that important life outcomes can be controlled if only sufficient effort is
exerted. Among those African-American males in the lower class who lack
the resources to make their belief in control a reality, John Henryism is
linked to elevated blood pressure and presumably to all of the negative
cardiac outcomes that result from hypertension (James, Hartnett, &
Kalsbeek, 1983). Along these lines, the Type A coronary-prone behavior
pattern seems to involve exaggerated beliefs about what the individual can
make happen, and this style, again, has been linked to cardiac morbidity
and mortality (M. Friedman & Rosenman, 1974).
The resolution suggested here is that optimism leads to good health
when this belief is at least in principle veridical. What makes optimism an
intriguing psychological variable is precisely that although it may be at odds
with the initial facts (Taylor, 1989), it can be self-fulfilling. If circumstances
do not allow an optimistic belief to be translated into a good outcome, then
this belief is irrelevant at best and health-damaging at worst because it leads
the individual to pursue impossible goals.
When unrealistic optimism is more a trait than a (transient) state, a
persons life is compromised. Consider such problematic styles of behaving
as hostility, perfectionism, risk taking, and narcissism; these can be recast
in terms of unrealistic optimism (Peterson, 1999). Although not all these

OPTlMISM AND PHYSICAL WELL-BEING

135

styles have been examined with respect to physical well-being, our strong
sense is that they are indeed unhealthy.
The pervasive optimistic bias in risk perception is unrealistic and
thus potentially hazardous. People who evidence the optimistic bias are
Pollyannas-foolishly optimistic-and should not be confused with people
who have an optimistic explanatory style. The positive expectations these
latter individuals entertain are entwined with a sense of efficacy.
As Bandura ( 1986) emphasized, self-efficacy, although not identical
with reality, is nonetheless influenced by the facts of the matter. Efficacy
expectations (I can perform this behavior) are advantageous only when
coupled with veridical outcome expectations (And this behavior has these
consequences). The optimistic bias in risk perception is the result of incorrect outcome expectations. However, the bias is reduced or eliminated to
the degree that individuals have first-hand experience with or extensive
knowledge about a given illness or injury and its risk factors (e.g., DeJoy,
1989; Kulik & Mahler, 1987; OHare, 1990; van der Velde, van der Pligt,
& Hooykaas, 1994).

QUESTIONS ABOUT THE HEALTH-OPTIMISM LINK

Can a lifetime of good health be encouraged by the early cultivation
of optimism? Gillham, Reivich, Jaycox, and Seligman (1995) have begun
an intervention program that involves teaching grade school children to
be more optimistic. Results to date suggest that optimism training makes
subsequent episodes of depression less likely. Perhaps studies of physical wellbeing can be piggy-backed onto this ongoing investigation (see Buchanan,
Gardenswartz, & Seligman, 1999).
We can similarly ask if optimistic ways of thinking acquired once an
individual falls ill will boost good health. Again, studies of optimism and
depression suggest that this question is worth pursuing. Cognitive therapy
that targets negative ways of thinking alleviates depression and prevents its
recurrence (Seligman et al., 1988). By implication, cognitive therapy for
the seriously ill might pay health dividends.
Certainly, the most basic question that arises from research into optimism and health involves the mechanisms linking the two. In our earlier
writing, we suggested that the link is overdetermined, and we sketched a
number of plausible routes (Peterson & Bossio, 1991).
For example, there may be an immunological pathway. Kamen-Siegel,
Rodin, Seligman, and Dwyer (1991) showed that optimism is positively
correlated with the vigor with which the immune system responds to an
antigen challenge. Several other researchers have looked at how explanatory

136

PETERSON AND BOSS10

style is linked to the progression of AIDS. Results here are mixed, but there
are hints that optimism predicts survival time and that this effect is mediated
in part by immunological factors (e.g., Blomkvist et al., 1994; Caumartin,
Joseph, & Gillespie, 1993; Rabkin, Remien, Katoff, & Williams, 1993; but
see Bofinger, Marguth, Pankofer, Seidl, & Ermann, 1993; Chuang, Jason,
Pajurkova, & Gill, 1992; Tomakowsky, Lumley, Markowitz, & Frank, 1996).
There may also be an emotional pathway between optimism and health.
A n extensive research literature has shown optimism to be incompatible
with depression (Sweeney, Anderson, & Bailey, 1986), and other studies
have linked depression to poor health and early death. At least part of this
latter path may be immunological (Schleifer, Keller, Siris, Davis, & Stein,
1985), and we begin to see the complexity involved in explaining why
optimism and health are associated.
There are probably several cognitive pathways between optimism and
health as well. Optimism is not an isolated belief but rather part of a complex
knowledge system that can impact physical well-being in numerous ways.
Dykema, Bergbower, and Peterson (1995), for example, showed that individuals with an optimistic explanatory style see the world as less filled with
hassles than do their pessimistic counterparts; this tendency is in turn linked
to better health.
Peterson and de Avila (1995), for another example, found that an
optimistic explanatory style is associated with the belief that good health can
be controlled (i.e., maintained and promoted). Indeed, in their research,
Peterson and de Avila showed that an optimistic explanatory style is positively correlated with what we have been discussing as an optimistic bias
in risk perception. However, this correlation was accounted for by the belief
that one was able to do things to reduce risk.
Another explanation of why optimistic thinking is related to physical
well-being points to a social pathway. People with a pessimistic explanatory
style are often socially isolated (Anderson & Arnault, 1985), and social
isolation is a consistent predictor of poor health (Cobb, 1976).
In general terms, the individuals social context can set the stage for
the optimism-health link. Sagan (1987) put forth the intriguing argument
that the dramatic increase in life expectancy in the Western world over
the centuries was due not to breakthroughs in medical or public health
practices but instead to cultural diffusion of the originally radical notion
that an individual is a discrete self able to have an effect on the world. Once
the idea of individual agency was invented, legitimized, and disseminated, the
findings we have been discussing here became possible.
Although we believe that the link between optimism and health is
complexly determined, the most typical and most robust mechanism is a
mundane behavioral pathway. Self-efficacy research consistently attests to
the benefic.ia1behavior that efficacy expectations set into motion. Indeed,
OPTIMISM AND PHYSICAL WELL-BEING

137

Bandura (1977) developed the self-efficacy construct as part of a theory

of behavior change. Similarly, Scheier and Carvers (1985) research on
dispositional optimism emerged from a concern with self-regulation: how
people align their behaviors with their goals and values. People who are
dispositionally optimistic have better health because they act differently
than do people who are dispositionally pessimistic; they cope differently,
by persisting as opposed to giving up (Carver et al., 1993; Scheier et al.,
1989; Strack, Carver, & Blaney, 1987). Individuals who are pessimistic
about contracting AIDS sometimes engage in riskier sexual behavior, an
example of a self-fulfilling prophecy if ever there were one (Kok, Ho, Heng,
& Ong, 1990).2
Our own research on explanatory style similarly attests to the critical
role of behavior in producing the optimism-health link. Peterson (1988)
found that an optimistic explanatory style was associated with a variety of
healthy practices: exercising, drinking in moderation, avoiding fatty foods,
and the like. Peterson, Colvin, and Lin (1992) similarly found that people
with an optimistic explanatory style were more likely than those with a
pessimistic explanatory style to respond to colds with appropriate actions:
resting and consuming more of Moms chicken noodle soup.
Our most recent study of optimistic explanatory style and physical
well-being looked at more than 1,000 individuals over almost 50 years
(Peterson, Seligman, Yurko, Martin, & Friedman, 1998).Pessimistic individuals had an increased likelihood of an early death, and the large sample size
made it further possible for us to investigate associations between explanatory
style and death due to different causes. We expected that death by cancer
and perhaps death by cardiovascular disease would be especially linked to
a pessimistic way of thinking. However, we found that pessimistic individuals
were most likely to die accidental or violent deaths. This effect was particularly pronounced for men.
Deaths like these are not random. Being in the wrong place at the
wrong time may be the result of an incautious and fatalistic lifestyle entwined not only with pessimism but also with the male gender role. This
study did not tell us what our deceased research participants were doing
when they died accidentally or violently, but we strongly suspect that their
behavior was somehow implicated, if only by affecting the settings they

T h i s result may not be a typical finding. Several studies have found that optimism is linked to
behaviors that place one at risk for AIDS (e.g., Bahr et al., 1993; Perkins, Leserman, Murphy, &
Evans, 1993), and still other investigations have discovered no association at all (e.g., Fontaine,
1994; Taylor et al., 1992). Perhaps both extreme pessimism and extreme optimism may be risk
factors. The need to ground optimism in reality before deeming it healthy or unhealthy is
underscored.

138

PETERSON AND BOSSIO

habitually entered (Buss, 1987). How does that old joke go? I broke my
nose in two places. Oh, yeah? If I were you, Id stay out of those places.

CONCLUSION
Optimistic thinking is linked to good health, but two important conditions must be satisfied. First, a persons optimism must lead him or her to
act in a vigorous and sustained fashion: Behavior is a critical link in the
process of attaining and sustaining physical well-being. Second, the behavior
encouraged by optimism must have a realistic link to health. A casual
reading of the popular literature on health and optimism might lead one
to conclude that good health is all in a persons mind, but thoughts and
beliefs must be situated within the person, and the person must be situated
within the world (see Chopra, 1993; Cousins, 1981; Peale, 1952; Siegel,
1988).
It is no wonder that consumers are attracted to pop psychology. Americans are justifiably confused about health. The scientific establishment must
bear considerable responsibility for this state of affairs. Citizens are deluged
with information about health risks yet provided with no guidance about
how to think about this information. For example, absolute risks should be
stressed along with relative risks. It is ludicrous to debate whether mosquitoes or dentists can transmit AIDS if we overlook the role played by sexual practices.
Furthermore, modem medical practice has become so specialized and
so narrowly biological that patients feel dealt out of the process of maintaining their own health. No wonder alternative medicine is thriving in the
United States today (Eisenberg et al., 1993), just as it did about a century ago
when conventional medicine was in a similar crisis (Weil, 1988). Alternative
approaches address the whole person: thoughts, feelings, and behaviors.
They allow the person to exercise choice. They sustain hope. Conventional
medicine, in contrast, dismisses the role of psychological factors in health
as placebo (that is, not real).
Finally, many health professionals, including ourselves in our own
research, tend to focus on the quantity of life: how long people live and
the degree to which they are free from disease. The quality of life-how
people live for as long or as short as they do-is surely as important. Living
well is sometimes referred to as wellness, a concept that cannot be captured
by longevity or freedom from disease (Barsky, 1988). Wellness involves a
zest for ongoing life, a fulfilling career, and satisfactory relationships with
family members and friends. When we think about psychological influences

OPTlMlSM AND PHYSICAL WELL-BEING

139

on health, we should consider how psychology can add to the quality of

life (Seeman, 1989).

REFERENCES
Abramson, L. Y., Metalsky, G. I., & Alloy, L. B. (1989).Hopelessness depression:
A theory-based subtype of depression. Psychological Review, 96,358-372.

Abramson, L. Y.,Seligman, M. E. P., & Teasdale, J. D. (1978).Learned helplessness

in humans: Critique and reformulation. Journal of Abnormal Psychology, 87,

49-74.
Anderson, C. A., & Arnault, L. H. (1985).Attributional style and everyday problems in living: Depression, loneliness, and shyness. Social Cognition, 3, 16-35.
Antonovsky, A. (1985).The sense of coherence as a determinant of health. In
J. D. Matarazzo, S. M. Weiss, J. A. Herd, N. E. Miller, & S. M. Weiss (Eds.),
Behavioral health: A handbook of health enhancement and disease prevention (pp.
114-129). New York: Wiley.
Bahr, G. R., Sikkema, K. J., Kelly, J. A., Fernandez, M. I., Stevenson, L. Y., &
Koob, J. J. (1993).Attitudes and characteristics of gay men who remain at
continued risk for contracting HIV infection. InternationalConference on AIDS,
9,697.
Bandura, A. (1977).Self-efficacy:Toward a unifying theory of behavioral change.
Psychological Review, 84, 191-215.
Bandura, A. (1986).Social foundations of thought and action. Englewood Cliffs, NJ:
Prentice-Hall.

Barsky, A.J. (1988).Worried sick: Our troubkd quest for wellness. Boston: Little,
Brown.
Blomkvist, V., Theorell, T., Jonsson, H., Schulman, S., Berntorp, E., & Stiegendal,
L. (1994).Coping style in relation to the consumption of factor concentrate
in HIV-infected hemophiliacs during the years after their infection became
known. Psychotherapy and Psychosomatics, 61,205-210.

Bofinger, F., Marguth, U., Pankofer, R., Seidl, O., & Ermann, M. (1993).Psychosocia1aspects of longterm-survivingwith AIDS. InternationalConferenceon AIDS,

9,878.
Buchanan, G. M., Gardenswartz, C. A. R., & Seligman, M. E. P. (1999,December
21).Physical health following a cognitive-behavioralintervention. Prevention
and Treatment, 2, Article 10. Available on the World Wide Web: http://

journals.apa.org/prevention/volume2/pre002001Oa.html
Buss, D. M. (1987).Selection, evocation, and manipulation. Journal of Personality
and Social Psychology, 53, 1214-1221.
Carver, C. S. (1989).How should multi-faceted personality constructs be tested?
Issues illustrated by self-monitoring, attributional style, and hardiness. Journal
of Personality and Social Psychology, 56, 577-585.

140

PETERSON AND BOSS10

Carver, C. S., Pozo, C., Harris, S. D., Noriega, V., Scheier, M. F., Robinson, D. S.,
Ketcham, A. S., Moffat, F. L., & Clark, K. C. (1993). How coping mediates
the effect of optimism on distress: A study of women with early stage breast
cancer. Journal of Personality and Social Psychology, 65, 375-390.
Carver, C. S., Spencer, S. M., & Scheier, M. F. (1998). Optimism, motivation,
and mental health. In H. S. Friedman (Ed.), Encyclopedia of mental health (Vol.
3, pp. 41-52). San Diego, CA: Academic Press.
Caumartin, S. M., Joseph, J. G., & Gillespie, B. (1993). The relationship between
social participation and AIDS survival in the Chicago MACSlCCS cohort.
International Conference on AIDS, 9, 886.
Chopra, D. (1993). Agekss body, timeless mind. New York: Harmony Books.
Chuang, H. T., Jason, G. W., Pajurkova, E. M., & Gill, M. J. (1992). Psychiatric
morbidity in patients with HIV infection. Canadian Journal of Psychiatry,
37, 109-115.
Cobb, S. (1976). Social support as a moderator of life stress. Psychosomatic Medicine,
38, 300-314.
Costa, P. T., & McCrae, R. R. (1987). Neuroticism, somatic complaints, and
disease: Is the bark worse than the bite? Journal of Personality, 55, 299-316.
Cousins, N. (1981). The anatomy of an illness. New York: Norton.
DeJoy, D. M. (1989). The optimism bias and traffic accident perception. Accident
Analysis and Prewention, 21, 333-340.
Dua, J., & Plumer, G. (1993). Relationship between attributional style, individualized attributional style, and health. Psychological Reports, 72, 913-914.
Dykema, J., Bergbower, K., & Peterson, C. (1995). Pessimistic explanatory style,
stress, and illness. Journal of Social and Clinical Psychology, 14, 357-371.
Eisenberg, D. M., Kessler, R. C., Foster, C., Norlock, F. E., Calkins, D. R., &
Delblanco, T. L. (1993). Unconventional medicine in the United States:
Prevalence, costs, and patterns of use. New England Journal of Medicine,
328, 246-252.
Fontaine, K. R. (1994). Effects of dispositional optimism on comparative risk perceptions for developing AIDS. Psychological Reports, 74, 843-846.
Friedman, H. S., & Booth-Kewley, S. (1987). The disease-prone personality: A
meta-analytic view of the concept. American Psychologist, 42, 539-555.
Friedman, M., & Rosenman, R. (1974). Type A behawior and your heart. New
York: Knopf.
Gillham, J. E., Reivich, K. J., Jaycox, L. H., & Seligman, M. E. P. (1995). Prevention
of depressive symptoms in schoolchildren: Two-year follow-up. Psychological
Science, 6, 343-351.
James, S. A., Hartnett, S. A., & Kalsbeek, W. D. (1983). John Henryism and
blood pressure differences among black men. Journal of Behawioral Medicine,
6, 259-278.
Kamen-Siegel, L., Rodin, J., Seligman, M. E. P., & Dwyer, J. (1991). Explanatory
style and cell-mediated immunity. Health Psychology , 10, 229-235.
OPTIMISM AND PHYSICAL WELL-BEING

141

Kobasa, S. C. (1979). Stressful life events, personality, and health: An inquiry into
hardiness. Journal of Personality and Social Psychology, 37, 1-1 1.
Kok, L. P., Ho, M. L., Heng, B. H., & Ong, Y. W. (1990). A psychosocial study
of high risk subjects for AIDS. Singapore Medical Journal, 31, 573-582.
Kulik, J. A., & Mahler, H. I. (1987). Health status, perceptions of risk, and prevention interest for health and nonhealth problems. Health Psychology, 6, 15-27.
Lee, F., & Peterson, C. (1997). Content analysis of archival data.Journal of Consulting and Clinical Psychology, 65, 959-969.
Lin, E. H., & Peterson, C. (1990). Pessimistic explanatory style and response to
illness. Behaviour Research and Therapy, 28, 243-248.
Michaels, C. E., Michaels, A. J., & Peterson, C. (1997). Motivation and health.
In P. Pintrich & M. Maeher (Eds.), Adwances in motivation and achievement
(Vol. 10, pp. 339-374). Greenwich, CT:JAI Press.
OHare, D. (1990). Pilots perception of risks and hazards in general aviation.
Aviation Space and Environmental Medicine, 61, 599-603.
OLeary, A. (1985). Self-efficacy and health. Behaviour Research and Therapy, 23,
437-451.
Peale, N. V. (1952). The power of positive thinking. Englewood Cliffs, NJ: Prentice-Hall.
Perkins, D. O., Leserman, J., Murphy, C., & Evans, D. L. (1993). Psychosocial
predictors of high-risk sexual behavior among HIV-negative homosexual men.
AIDS Education and Prevention, 5, 141-152.
Peterson, C. (1988). Explanatory style as a risk factor for illness. Cognitive Therapy
and Research, 12, 117-130.
Peterson, C. (1995). Explanatory style and health. In G. M. Buchanan & M. E.
P. Seligman (Eds.), Explanatory style (pp. 233-246). Hillsdale, NJ: Erlbaum.
Peterson, C. (1999). Personal control and well-being. In D. Kahneman, E. Diener,
& N. Schwarz (Eds.), Well-being: The foundations of hedonic psychology (pp.
288-301). New York: Russell Sage Foundation.
Peterson, C. (2000). The future of optimism. American Psychologist, 57, 44-55.
Peterson, C., Bettes, B. A., & Seligman, M. E. P. (1985). Depressive symptoms
and unprompted causal attributions: Content analysis. Behaviour Research and
Therapy, 23, 379-382.
Peterson, C., & Bossio, L. M. (1991). Health and optimism. New York: Free Press.
Peterson, C., & Bossio, L. M. (1993). Healthy attitudes: Optimism, hope, and
control. In D. Goleman & J. Gurin (Eds.), Midbody medicine: How to we your
mind for better health (pp. 351-366). Yonkers, NY: Consumer Reports Books.
Peterson, C., Colvin, D., & Lin, E. H. (1992). Explanatory style and helplessness.
Social Behavior and Personality, 20, 1-14.
Peterson, C., & de Avila, M. E. (1995). Optimistic explanatory style and the
perception of health problems. Journal of Clinical Psychology, 51, 128-132.

142

PETERSON AND BOSS10

Peterson, C., Maier, S. F., & Seligman, M. E. P. (1993). Learned helplessness: A

theory for the age of personal control. New York: Oxford University Press.
Peterson, C., & Park, C. (1997). Implicit and explicit explanatory styles in the lifecourse
of college-educated women. Unpublished manuscript, University of Michigan,
Ann Arbor.
Peterson, C., Schulman, P., Castellon, C., & Seligman, M. E. P. (1992). CAVE:
Content analysis of verbatim explanations. In C. P. Smith (Ed.), Motivation
and personality: Handbook of thematic content analysis (pp. 383-392). New York:
Cambridge University Press.
Peterson, C., & Seligman, M. E. P. (1984). Causal explanations as a risk factor for
depression: Theory and evidence. Psychological Review, 91 , 347-374.
Peterson, C., & Seligman, M. E. P. (1987). Explanatory style and illness. Journal
of Personality, 55, 237-265.
Peterson, C., Seligman, M. E. P., & Vaillant, G. E. (1988). Pessimistic explanatory
style is a risk factor for physical illness: A thirty-five year longitudinal study.
Journal of Personality and Social Psychology, 55, 23-27.
Peterson, C., Seligman, M. E. P., Yurko, K. H., Martin, L. R., & Friedman, H. S.
( 1998). Catastrophizing and untimely death. Psychological Science, 9, 127-130.
Peterson, C., Semmel, A., von Baeyer, C., Abramson, L. Y., Metalsky, G. I., &
Seligman, M. E. P. (1982). The Attributional Style Questionnaire. Cognitive
Therapy and Research, 6, 287-299.
Peterson, C., & Stunkard, A. J. (1989). Personal control and health promotion.
Social Science and Medicine, 28, 819-828.
Peterson, C., & Villanova, P. (1988). An expanded Attributional Style Questionnaire. Journal of Abnormal Psychology, 97, 87-89.
Peterson, C., Villanova, P., & Raps, C. S. (1985). Depression and attributions:
Factors responsible for inconsistent results in the published literature. J o u d
of
Psychology, 94, 165-168.
Rabkin, J. G., Remien, R., Katoff, L., & Williams, J. B. (1993). Resilience in
adversity among long-term survivors of AIDS. Hospital and Community Psychiutry, 44, 162-167.
Reivich, K. (1995). The measurement of explanatory style. In G. M. Buchanan &
M. E. P. Seligman (Eds.), Explanatory style (pp. 21-47). Hillsdale, NJ: Erlbaum.
Robins, C. J. (1988). Attributions and depression: Why is the literature so inconsistent?Journal of Personality and Social Psychology, 54, 880-889.
Sagan, L. A. (1987). The health of nations: True causes of sickness and well-being.
New York: Basic Books.
Scheier, M. F., 6r Bridges, M. W. (1995). Person variables and health: Personality
predispositions and acute psychological states as shared determinants for disease. Psychosomatic Medicine, 57, 255-268.
Scheier, M. F., & Carver, C. S. (1985). Optimism, coping, and health: Assessment
and implications of generalized outcome expectancies. Health Psychology, 4,
219-247.
OPTIMISM AND PHYSICAL WELL-BEING

143

Scheier, M. F., & Carver, C. S. (1987). Dispositional optimism and physical wellbeing: The influence of generalized outcome expectancies on health. journal
of Persodity, 55, 169-210.
Scheier, M. F., & Carver, C. S. (1992). Effects of optimism on psychological
and physical well-being: Theoretical overview and empirical update. Cognitive
Therapy and Research, 16, 201-228.
Scheier, M. F., Matthews, K. A., Owen, J. F., Magovern, G. J., Lefebvre, R. C.,
Abbott, R. A., & Carver, C. S. (1989). Dispositional optimism and recovery
from coronary artery bypass surgery: The beneficial effects on physical and
psychological well-being. Journal of Personality and Social Psychology, 57, 10241040.
Schleifer, S. J., Keller, S. E., Siris, S. G., Davis, K. L., & Stein, M. (1985). Depression
and immunity. Archives of General Psychiatry, 42, 129-133.
Schulz, R., Bookwala, J., Knapp, J. E., Scheier, M. F., &Williamson, G. M. (1996).
Pessimism, age, and cancer mortality. Psychology and Agtng, 1 I , 304-309.
Seeman, J. (1989). Toward a model of positive health. American Psychologist, 44,
1099-1 109.
Seligman, M. E. P., Castellon, C., Cacciola, J., Schulman, P., Luborsky, L., Ollove,
M., & Downing, R. (1988). Explanatory style change during cognitive therapy
for unipolar depression. Journal of Abnormul Psychology, 97, 13-18.
Siegel, B. S. (1988). Love, medicine, and mirucks. New York: Perennial.
Strack, S., Carver, C. S., & Blaney, P. H. (1987). Predicting successful completion
of an aftercare program following treatment for alcoholism: The role of dispositional optimism. journal of Personality and Social Psychology, 53, 579-584.
Sweeney, P. D., Anderson, K., & Bailey, S. (1986). Attributional style in depression:
A meta-analytic review. journal of Personality and Social Psychology, 50, 974991.
Taylor, S. E. (1989). Positive illusions. New York: Basic Books.
Taylor, S. E., Kemeny, M. E., Aspinwall, L. G., Schneider, S. G., Rodriguez, R.,
& Herbert, M. ( 1992). Optimism, coping, psychological distress, and high-risk
sexual behavior among men at risk for acquired immunodeficiency syndrome
(AIDS). Journal of Personality and Social Psychology, 63, 460-473.
Tennen, H., & Herzberger, S. (1986). Attributional Style Questionnaire. In D. J.
Keyser & R. C. Sweetland (Eds.), Test critiques (Vol. 4, pp. 20-30). Kansas
City, KS: Test Corporation of America.
Tomakowsky, J., Lumley, M. A., Markowitz, N., &Frank, C. (1996). The relationships
of optimistic explanatory styk and dispositional optimism to health in HIV-infected
men. Unpublished manuscript, Wayne State University, Detroit, MI.
van der Velde, F. W., van der Pligt, J., & Hooykaas, C. (1994). Perceiving AIDSrelated risk: Accuracy as a function of differences in actual risk. Health Psychology, 13, 25-33.
Verbrugge, L. M. (1989). Recent, present, and future health of American adults.
Annual Review of Public Health, 10, 333-361.

144

PETERSON AND BOSS10

Wallston, K. A., & Wallston, B. S. (1978). Locus of control and health: A review
of the literature. Health Education Monographs, 6 , 107-117.
Weil, A. (1988). Health and healing (Rev. ed.). Boston: Houghton Mifflin.
Weinstein, N. D. (1989). Optimistic biases about personal risks. Science, 246, 12321233.

OPTIMISM AND PHYSICAL WELL-BEING

145

7
OPTIMISM, PESSIMISM, A N D DAILY
LIFE WITH CHRONIC ILLNESS
GLENN AFFLECK, HOWARD TENNEN, AND ANDREA APTER

Dispositional optimism has become a new focal point of theory and

research on adaptation to adversity and stressful events. Nowhere has this
formidable resource been inspiring more investigation than in the literature
on coping with illness and other medical stressors. Health problems and their
treatment offer an excellent opportunity to evaluate whether a generalized
expectancy for positive outcomes (Scheier & Carver, 1985) confers both
short- and long-term benefits for both emotional and physical well-being.
In this vein, people characterized as dispositional optimists have been shown
to display superior psychosocial adaptation to a host of medical stressors,
including coronary artery bypass surgery (Fitzgerald, Tennen, Affleck, &
Pransky, 1993; Scheier et al., 1989); childbirth (Carver & Gaines, 1987);
The authors studies described in this chapter were funded by the National Institute of Arthritis,
Musculoskeletal, and Skin Diseases, the American Lung Association of Connecticut, and the
University of Connecticut General Clinical Research Center. For their assistance with these studies,
we thank co-investigators Susan Urrows, Pamela Higgins, Judith Fifield, Richard ZuWallack, and
Micha Abeles.

147

failed in vitro fertilization (Litt, Tennen, Affleck, & Klock, 1992); bone
marrow transplantation (Curbow, Somerfield, Baker, Wingard, & Legro,
1993); arthritis (Tennen, Affleck, Urrows, Higgins, & Mendola, 1992);
tinnitus and hearing impairment (Anderson, 1996a); HIV-positive status
(Taylor et al., 1992); pregnancy termination (Cozzarelli, 1993); and cancer
in oneself (Carver et al., 1994); or in ones loved ones (Given et al., 1993).
In two reviews of the consequences of dispositional optimism for physical
well-being, Scheier and Carver (1987, 1992) summarized effects from the
appearance of physical symptoms in healthy individuals to milestones of
physical recovery from coronary artery bypass surgery, and of mediators such
as optimists better health habits, more positive mood, coping strategies,
competent immune functioning, and diminished cardiovascular reactivity
to stress.
Vexing theoretical questions about the mechanisms by which dispositional optimism can prevent illness, foster recovery, and assist adaptation
to health-related stressorsremain, and many methodological problems in this
research area require creative solutions. This chapter offers some conceptual
refinements and methodological innovations for investigation of the daily
consequences and constituents of optimism. Because few prospective studies
of daily experience on dispositional optimism have been published, we draw
heavily on the methods and empirical findings of our own research program
on daily life with chronic illness. Before summarizing this work as it relates to
literature on optimism, we review some of the theoretical and methodological
advantages of a daily-process paradigm for research on optimism and on
other personal dispositions and traits.

PERSONALITY AND THE STUDY OF EVERYDAY LIFE

Two decades ago, Lazarus (1978), Epstein (1979), and Csikszentmihalyi, Larson, and Prescott ( 1977) challenged personality psychologists to study
intensively peoples thoughts, feelings, and behaviors over time. Lazarus
believed that for the field to advance, investigators needed to shift from
examining a single individual difference on an isolated occasion to exploring
processes and patterns of relations that unfold in time. Epstein (1983)
argued that defining personality as relatively stable reaction tendencies that
distinguish individuals from one another obliges us to inspect responses of
multiple individuals on multiple measures over multiple occasions (p. 91).
Csikszentmihalyi and colleagues (Csikszentmihalyi & Larson, 1987) initiated studies of daily experience in which individuals recorded their momentary thoughts, feelings, and behavior in response to signals from radiocontrolled beepers. The experience sampling method presented a new

148

AFFLECK ET AL.

perspective on everyday behavior and revealed the benefits of studying

psychological processes in their natural context.
This call to study people over time and across occasions has had a
growing impact on the way personality research is conducted (Tennen, Suls,
& Affleck, 1991). Larsen ( 1990) surveyed articles appearing in three leading
personality journals that employed intensive time-sampling methods, most
of which took the day as their unit of analysis. The number of articles
employing these methods increased nearly threefold between 1976 and
1988, a trend that we suspect has continued at an accelerating pace. This
methodology has also been rapidly gaining popularity in the literature of
health psychology and behavioral medicine (Affleck & Tennen, 1996;Stone
& Shiffman, 1994).The compatibility of daily-process methods for research
on personality and health should also accelerate this specific area of investigation.
Tennen et al. (1991) identified several ways in which a daily-process
paradigm can advance research on personality, especially research at the
intersection of personality and health. The first promise of such a paradigm
lies in its giving personality theorists the ability to ask and answer the
traditional questions of interest in creative new ways. For example, do people
evidence consistency in their behavior and experience, or is daily variation
the norm? A long-standing debate among personality theorists is the extent
to which behavior is tied to its context and the extent to which it reflects
traits. In response to this debate, Epstein (1977) argued that because traits
refer to broad action tendencies, their adequate measurement requires aggregating data over time and across situations, a procedure executed best in the
context of peoples everyday activities and experiences. This issue remains
relatively unstudied in the literature on optimism. Is optimism stable across
time and situations when time intervals and situations flow from the rhythms
and events of everyday life? Although individuals might differ in the mean
levels of optimism across days and daily experiences, the deviations from
the mean may represent an important state component of optimism, with
its unique antecedents and consequences. The ebb and flow of symptoms
in chronically ill individuals, for example, might track state variations in
general optimism.
A second promise of studying daily life is that it will provide the
capability to test new models of how major events influence health and
well-being. It has been argued, for example, that major life events assume
some of their adaptational consequences by influencing everyday affairs
(Kanner, Coyne, Schaefer, & Lazarus, 1981; Zautra, Affleck, & Tennen,
1994). Thus, the death of a spouse is not only a significant loss, but also a
disruption of established recreational patterns, formerly shared responsibilities, and day-to-day social relations. A key question for future research is

OPTIMISM, PESSIMISM, AND DAILY LIFE

149

whether optimism could buffer individuals from the disruption in everyday

activities that typically stems from major life changes.
A third and most alluring promise is that studying people intensively
may allow us to address questions that can only be answered by studying
individuals on a daily basis. For example, do daily stressors affect mood
beyond the day of their occurrence? Although Stone and Neale (1984a)
reported that daily events affect mood only on the day they occur, Bolger,
DeLongis, Kessler, and Schilling (1989) found that mood was actually
brighter on a day following a stressful event than on stress-free days. The
relation between fleeting events and vacillating emotions is undoubtedly
complex and difficult to capture, and this complexity demands that we study
individuals daily or even more intensively. Such studies could document
the ability of optimism to modify the strength-or even direction-of the
dynamic relations between events and emotions.
Another question that can only be addressed by studying daily experiences is whether certain emotions or symptoms co-occur within the same
individual. For example, when people are depressed, are they also anxious?
Are they also angry? That depression, anxiety, and hostility are positively
correlated across individuals (e.g., Gotlib, 1984; Watson & Clark, 1984)
has no bearing on these questions. Rather, this positive correlation shows
that individuals who are less anxious than their peers are also less angry
and less depressed (see Epstein, 1986). Intraindividual correlations over
time and circumstances, which address these questions directly, actually
reveal that when someone is sadder than usual, she or he is less angry (Epstein,
1983) and less anxious (Larsen & Ketelaar, 1991). To best understand the
emotional and other processes related to psychological and physical wellbeing, we need to examine them as they unfold. Would the modest betweenperson relation between optimism and pessimism (e.g., Marshall, Wortman,
Kusulas, Hervig, & Vickers, 1992) be replicated at the within-person level
of the day-to-day covariation in the state components of these constructs?
Individual differences in this association might be fertile territory for
investigation.
These are only some of the questions that demand intensive time
sampling. Another set of questions relate to the study of patterns in experience and behavior. For example, the daily experience paradigm allows us
to address prospectively the duration of emotional distress (Epstein, 1983)
and physical symptoms (Larsen & Kasimatis, 1991). It provides a context
in which to study when and how individuals habituate to events (Bolger
et al., 1989), and to study who is more likely to do so. It encourages us to
examine the frequency of change in our adaptation to events (Larsen, 1987)
and individual differences in reactivity to events (Affleck, Tennen, Urrows,
&. Higgins, 1994; Bolger & Schilling, 1991). Most clearly, it forces us to

150

AFFLECK ET AL.

consider time as a psychological variable. Might optimistic people be

distinguished by their ability to interrupt periods of distress? Might they
recover more quickly from illness episodes? Might they be better able to
limit the number of consecutive days in which their chronic illness symptoms
are unusually severe?
The promise of a daily-process paradigm for research on optimism
remains largely unmet. We are aware of only three published studies on
optimism that used prospective daily measures for any purpose. Shifren and
Hooker (1995) examined the characteristics of a state version of the Life
Orientation Test (LOT; Scheier &. Carver, 1985) in a sample of 30 spouse
caregivers of individuals with probable Alzheimers disease. Both the state
LOT and mood were assessed daily for 30 days. The authors demonstrated
significant day-to-day variability in state optimism. For approximately one
third of the participants, state optimism changes were linked with changes
in positive and negative affect from day to day. Shifren (1996) followed
this up with a small study of patients with Parkinsons disease, who for 70
consecutive days completed the state version of the LOT and reported
indicators of disease severity. For 3 of the 12 patients studied, heightened
levels of optimism on a given day predicted declines the next day in perceived
severity of the illness. More optimistic individuals, that is, those individuals
who scored higher in average daily optimism, also saw themselves as less
in need of assistance with activities of daily living.
Tennen et al. (1992) examined how dispositional optimism and finding
benefits in adversity relate to one another and figure in the aggregate levels
of daily symptoms, mood, and functioning of individuals with rheumatoid
arthritis. After completing the LOT and measures of perceived control over
and benefits from their chronic pain drawn from the Inventory of Perceived
Control Beliefs (Mendola, 1990), research participants reported their pain
intensity, mood, and pain-related activity limitations (e.g., missing work,
cutting back on planned social activities) each day for 75 consecutive days.
Those participants scoring higher on the LOT were significantly more likely
to construe benefits from their illness, and they also reported significantly
higher levels of average positive daily mood and fewer pain-related activity
limitation days.
The study by Tennen et al. (1992) reports a few findings from our
broader research program and sets the stage for summarizing many other
effects of optimism or pessimism on prospective accounts of daily life with
chronic illness. We begin an overview of our work with a more comprehensive look at the daily correlates of optimism and pessimism in our
diary study of patients with rheumatoid arthritis. We then review selected
findings from a full idiographic-nomothetic analysis of the effects of dispositional optimism and pessimism on daily life with participants experiencing

OPTIMISM, PESSIMISM, AND DAILY LIFE

151

either fibromyalgia or asthma. We close with an exploration of a possible

constituent of optimism in daily life with chronic illness: the expectation
of symptom changes from day to day.

OPTIMISM, PESSIMISM, AND DAILY LIFE WITH ARTHRITIS

Our first attempt to document the daily consequences of dispositional
optimism or pessimism was in a prospective study of the daily life experiences
of individuals with rheumatoid arthritis (RA),a chronic and painful disease.
The 75 patients recruited for this study began their participation by completing questionnaire measures of optimism and pessimism with the LOT and
a host of other background characteristics, including neuroticism, depressive
symptoms, social support, pain appraisals, pain coping strategies, and catastrophizing responses to pain. Then, for 75 consecutive days, within 1 hour
of bedtime they completed a structured diary that they returned to us by
mail the next morning. The diary measures included the intensity of pain
in 20 joints, the quality of the previous nights sleep, positive and negative
mood reported on the POMS-B (Lorr & McNair, 1982), desirable and
undesirable events reported with the Daily Life Experience Checklist (Stone
& Neale, 1982), and the extent of pain-related daily limitations in work,
child care, social and household responsibilities.
Aggregating these daily measures afforded an uncommonly reliable
picture of patients average days. As we have documented elsewhere (Affleck
& Tennen, 1996), recall errors can be expected in retrospective accounts
of events, symptoms, mood, and appraisals. More important than inaccuracy
per se is the source of the error. We need to be concerned not only with
random error, but especially with systematic error in the recall of pain,
mood, and events. Individuals who differ on other study variables (or more
important, in key ways not measured) may report differentially accurate
data or use different cognitive heuristics to assist their recall (Linton, 1991).
The correlations presented in Table 7.1 taught us two lessons about
the role of generalized outcome expectancies in daily life with RA. First
was the independence of pessimism as a predictor of negative experiences.
Regardless of their level of optimism, the more pessimistic patients reported
more negative daily mood, more pain-related activity limitations, more
negative daily events, and poorer sleep. Of interest is that the only pessimismadjusted daily correlate of optimism was its association with positive mood.
These findings underscore the value of a two-factor model of generalized
expectancies and daily life, evidenced especially by the parallelism between
pessimism and negative experiences. They also reinforce the desirability of
treating optimism and pessimism as separate constructs, instead of as opposite

152

AFFLECK ET AL.

Table 7.1.
Statistically Significant Correlates of Dispositional Optimism or Pessimism
in 75 Patients With Rheumatoid Arthritis
Optimism

Pessimism

Education
Neuroticism
Depression symptoms
Perceived social support
Personal control over pain
Perceived efficacy of pain coping

-.45***
(-.3,,**)
-.31**
(-.21)
.27*
(.23*)
.43***
(.39***)
-.24*

Pain catastrophizing
Perceived pain-related benefits
Pain-coping strategies
Positive self-statements
Prayer and hope
Distraction
Reinterpret pain sensations

(-. 16)

.2v*
(.26*)

-.23*
(-.35**)
.58***
(.5,,**)
.42***
(.38***)
-.35**
(-.31**)
-.25*
(-.28*)
.38***
(.36**)

.29**
(.28**)
.27*
(.24*)
.33'*
(.27*)
.25*

Daily diary aggregates

Pain-related activity limitations
Undesirable events
Negative mood
Positive mood
Sleep quality

.46***
(.4,,**)

.23*
(.23*)
.30**
(.28*)
.38***
(.33**)
-.3a***
(-.31**)
-.29**
(-.23*)

Note. Nonparenthesized correlations are unpartialled; parenthesized correlations with optimism partial the
effects of pessimism, parenthesized correlations with pessimism partial the effects of optimism.
* p < .05.'* p < .01. *** p < .001.

ends of a bipolar continuum. There is compelling evidence that optimism,

at least as measured by total scores on the LOT, is not a unitary construct.
Instead, the LOT appears to measure two relatively orthogonal constructs:
optimism and pessimism (Chang, DZurilla, & Maydeu-Olivares, 1994; Robinson-Whelen, Kim, MacCallum, & Kiecolt-Glaser, 1997). Our findings

OPTIMISM, PESSIMiSM, AND DAILY LIFE

153

agree with those establishing that optimism and pessimism may not have
equivalent relations with other personality and adaptational outcome measures (e.g., Marshall et al., 1992; Mroczek, Spiro, Aldwin, Ozer, & Bosse,

1993).

Other correlations appearing in Table 7.1 taught us a second lesson,

this one concerning the specificityof these findings. Optimism and pessimism
were correlated with other person and situation variables that themselves
figure in the quality of daily life. Once again, the more pessimistic individuals
differed in other negative traits (neuroticism) and states (depression, poor
social support) regardless of their level of optimism, and the more optimistic
individuals were more likely to report cognitive coping strategies, good social
support, and more personal control over their pain, regardless of their level
of pessimism. The critical issue, which is more conceptual than statistical,
concerns the status of these variables in the relation to generalized expectancies and daily experiences. Are they to be viewed as potential confounds
or as mediators? If the former, they need to be controlled statistically. If
the latter, they need to be incorporated as explanatory variables in fuller
models, as has been accomplished in research on coping strategies as mediators of the relation between optimism and adaptation to health problems
(Scheier & Carver, 1992).
A controversial point concerns the shared variance between the personality trait of neuroticism and that of optimism or pessimism. Smith,
Pope, Rhodewalt, and Poulton (1989) were among the first to advocate
partialling the effects of neuroticism from relations between optimism or
pessimism and assessments of physical and psychological well-being. Scheier,
Carver, and Bridges (1994) countered this argument with two observations.
First, whereas the relation between optimism or pessimism and adaptational
outcomes can often be explained by neuroticism, other important mediating variables linking personality and well-being, for example, coping behaviors, appear to be independently related to optimism or pessimism. Second,
the construct of neuroticism itself consists partly of a pessimistic orientation,
introducing a conceptual overlap between the two constructs. This complicates the interpretation of neuroticism-adjusted associations between optimism or pessimism and other adaptational processes and outcomes. Furthermore, Marshall et al. (1992) found that when optimism and pessimism
are treated as independent constructs, only pessimism is a correlate of
neuroticism.
The findings just summarized from our RA daily experience study fail
to take full advantage of the variation in the person-day data set. By
aggregating the daily reports, we lost information about the within-person
relations between the daily-process variables. When both differences between persons and differences within persons over time are considered in

154

AFFLECK ET AL.

a mixed idiographic-nomothetic design, one can use powerful statistical

methods and draw stronger inferences about the role of optimism in the
dynamics of daily life (Affleck & Tennen, 1996; Epstein, 1983; Larsen &
Kasimatis, 1991).

EXAMINING RELATIONS WITHIN AND ACROSS INDIVIDUALS:

MIXED IDIOGRAPHIC-NOMOTHETIC DESIGN
The dominant tradition in behavioral research is the nomothetic approach, which asks whether there are lawful relations among variables across
individuals. The idiographic approach examines relations among variables
within an individual, across time or situations in that individuals life. The
combination of these approaches in the mixed idiographic-nomothetic design enables investigators to ask: Are there relations between variables within
individuals over time that generalize across individuals or that relate to difierences
between individuals! Answering this question requires that we first examine
relations among variables over time for each individual we study. We must
then return to the population level in two ways: by determining if the
within-person relations generalize across persons, and by discerning how
they relate to differences between individuals. As we demonstrate next,
relating within-person associations to between-person differences yields rich
insights into the processes by which optimism and pessimism are expressed
in the dynamics of daily life.

Daily Life With Chronic Illness: An Idiographic-Nomothetic Analysis

The two prospective daily studies we describe below were conducted
with adult patients with asthma and women with primary fibromyalgia
syndrome (PFS). The asthma sample consisted of 48 individuals with moderate-to-severe asthma who were taking inhaled steroids for disease management. The PFS sample comprises 89 women who experienced pain in all
body quadrants of at least 3 months duration, pain in 11 of 18 tender point
sites upon moderate digital pressure exerted by a clinical examiner, and the
absence of other musculoskeletal pain disorders that could cause secondary
fibromyalgia (e.g., rheumatoid arthritis).
The intensive self-monitoring methods used in these studies were
nearly identical. For 30 days (fibromyalgia) or 21 days (asthma) participants
filled out paper-and-pencil diaries before retiring to sleep. Variables measured
in the nightly booklet included a rating of that days perceived control over
illness symptoms; a rating of the perceived efficacy of that days strategies
of coping with symptoms; and a checklist of pain or asthma symptom coping

OPTIMISM, PESSIMISM, AND DAILY LIFE

155

strategies adapted from the Daily Coping Inventory (Stone & Neale, 1984b);
and for fibromyalgia patients only, a checklist of undesirable and desirable
daily events culled from the Inventory of Small Life Events (Zautra, Guarnaccia, Reich, & Dohrenwend, 1988).
Pain, asthma symptoms, and mood are likely to change over the course
of a day, so even end-of-day summaries may be contaminated by systematic
retrospection errors in subjective averaging (Hedges, Jandorf, & Stone,
1985). To gather within-day reports more reliably, we employed an electronic interviewer (ELI) modeled on Shiffmans (Shiffman et al., 1994)
innovative procedure for ecological momentary assessment (Stone & Shiffman, 1994). Participants carried Psion Organizer I1 palm-top computers
(Psion Corporation, Concord, MA), which were programmed to request,
and then time stamp, information about their previous nights sleep quality
once a day after awakening, and to request and then time stamp their
symptoms, mood, and the extent to which their attention was focused on
their symptoms three times a day, at randomly selected times during the
morning, afternoon, and evening. (For further procedural details and other
findings from our ELI studies, see Affleck et al., 1998; Affleck, Urrows,
Tennen, Higgins, & Abeles, 1996; Apter et al., 1997).
The analyses summarized below concern the ability of dispositional
optimism or pessimism to predict (a) mean levels of daily symptoms;
attention to symptoms, mood, daily events, coping strategies, and symptomrelated cognitions; and (b) variation in the within-person relations of
changes in symptom severity with other measures, that is, symptomreactivity, and within-person relations of changes in undesirable events
with other measures, that is, stress-reactivity.
Statistical modeling of nomothetic-idiographic relations requires solutions that partition the dual sources of variance in person-day data sets. For
our analyses we have adopted a simultaneous multilevel modeling (MLM)
strategy. MLM treats both day-level and person-level observations as random
sampling units so that inferences can be made to both days and persons
from whom the sample was drawn. It allows estimation of the random effects
for each person of one daily variable on another and then models the random
effects of person attributes that may modify the strength and direction of
within-person relations.
The specific MLM technique we use here is Hierarchical Linear Modeling (Bryk & Raudenbush, 1992). The vernacular of HLM calls daily observations Level 1 variables that are nested within persons, the differences
between whom are called Level 2 variables (e.g., differences in personal
attributes and instable situational factors). Level 2 variables can be used to
simultaneously model individual differences in intercepts (means) of the
level variables and to model their slopes, that is, the relations between Level
1 variables.
156

AFFLECK E T AL.

Table 7.2.
HLM Level 2 Analysis of Effects of Dispositional Optimism or Pessimism
on Daily Variable Intercepts for 89 Fibromyalgia and 48 Asthma Patients
Fibromyalgia
Dailv variable

ODtimism

Intensity of paidasthma
symptoms
Attention to paidasthma
symptoms
Pleasant mood
Unpleasant mood
Personal control of paidasthma
symptoms
Perceived efficacy of pairdasthma
symptoms coping
For pain/symptom
improvement
For mood enhancement
Desirable daily events
Undesirable daily events
Coping strategies to contend with
pain/symptoms
Take more medication
Behavioral action to reduce
pain/symptoms
Relaxation
Distraction
Redefinition
Vent emotions
Seek spiritual comfort
Seek emotional S U.D. D O ~ ~

Asthma

Pessimism

ODtimism

Pessimism

.513

-.213

.I84

-. 156*

.069

-.005
.148*
-.071

.ooo

-.026

.I09

.049
-.171*
.059

.030
.094*
.321*
-.010

-.038
-.073
-. 194
-.028

.609
.313

-.014
-.343

-.016

-.004

-.025*

.005

-.005
-.003

.005
.003
.003
.002
-.004
-.029*
.002

-.008
-.024
-.015
.012
-.009
.022*

.008
.034*

-.227
-.067
.207***
-. 194**

-.001

.011

-.001

.027*
-.002

.138*

-.010

.009

.010

.ooo
.01o*
.005

.011*

Note. Tabular values are unstandardized maximum likelihood estimates.

p c .05.'* p < .01. *** p c ,001.

Optimism or Pessimism and Within-Person Intercepts

The ability of our Level 2 measures of optimism and pessimism to
predict observation-level intercepts (means) was evaluated first. Table 7.2
presents the findings.
Unexpectedly, dispositional optimism and pessimism played no direct
role in individuals' intercepts for momentary pain or asthma symptoms.
Neither did it predict attention to pain or asthma symptoms or perceived
control over pain or asthma symptoms. It did predict the daily moods of
participants in both studies. In both groups, optimism predicted pleasant
daily mood, and among those participants with fibromyalgia, so did lower
levels of pessimism. Among those participants with fibromyalgia, pessimism
was also a predictor of unpleasant mood and optimism a predictor of the
frequency of desirable daily events.
OPTIMISM, PESSIMISM, AND DAILY LIFE

157

Patients differing in levels of optimism or pessimism also used certain

daily coping strategies to a greater or lesser extent. The more optimistic
asthma patients were less likely to take extra medication to contend with
that days symptoms, and the more pessimistic asthma patients were more
likely to seek emotional support and to vent distressing emotions. A more
consistent finding across samples concerned differences in seeking spiritual
comfort through prayer and other devotional activities. The more optimistic,
or the less pessimistic, their general orientation, the more participants sought
spiritual comfort as a way of coping with that days pain or asthma symptoms.
Optimistic participants also tended more to view their coping strategies as
mood regulators; they were more likely to conclude that their days coping
efforts helped to improve their mood.
Optimism or Pessimism and Within-Person Slopes
Our next set of analyses answers the general question: Do optimists
and pessimists differ in their unique within-person relations between one
daily-process measure and another?We examined first those linkages expressing each persons symptom-reactivity, including the relation between symptom change and changes in mood, attention to symptoms, perceived control
over symptoms, and perceived efficacy of coping strategies to improve symptoms. Only one of these relations could be predicted by levels of optimism
or pessimism: Asthma patients who were more optimistic were significantly
less likely to pay attention to their symptoms when they increased in severity
(b = -.02; t = -2.42; p = .02), and this was true even when their degree of
pessimism was taken into account in the Level 2 analysis.
We also examined in the sample composed of patients with fibromyalgia
the within-person assessment of daily stress reactivity (Affleck et al.,
1994), as evidenced by the magnitude of the within-person relation between
changes in undesirable daily events and in mood, pain, and attention to pain.
Again, there was an optimism-associated difference involving attentional
processes. Greater optimism was linked with a significantly weaker association between a rise in undesirable events and an increase in attention to
pain ( b = -.01; t = -2.80; p = .006). This was the case even when the
extent of pessimism was incorporated in the Level 2 analysis.
Discussion
In a review of research on the effects of dispositional optimism on
psychological and physical well-being, Scheier and Carver ( 1992)concluded
that compared to pessimists, optimists manage difficult and stressful events
with less subjective distress and less adverse impact on their physical wellbeing (p. 224). They add that in part, this optimistic advantage seems

158

AFFLECK ET AL.

due to differences between optimists and pessimists in the manner in which

they cope with stress (p. 224).
The findings of our two prospective daily studies of patients with
fibromyalgia or asthma support this conclusion concerning the advantages
of optimism for psychological well-being but not for physical well-being. In
neither sample was dispositional optimism or pessimism a predictor of physical symptoms recorded three times a day across 30 or 21 days. In a broader
conception of psychological well-being, the link between optimism and the
experience of more desirable daily events by patients with fibromyalgia is
noteworthy. In both groups, there were predictable effects of an optimistic
or pessimistic orientation on daily mood. We also confirmed with daily
reports of coping with pain or asthma symptoms that optimists and pessimists
use some coping strategies to a greater or lesser degree. The tendency for
optimists to rely more on spiritual or religious coping with chronic illness
symptoms has not been found or addressed previously.
Are the daily mood differences associated with optimism and pessimism
mediated by differences in coping strategy use (Scheier & Carver, 1992)?
When we examined the within-person relations between those coping strategies associated with optimism or pessimism and mood the following day,
there were no significant findings. Thus, we could find no strong evidence
that the more positive mood of optimists, or the more negative mood
of pessimists, was due to differences in their daily coping attempts. The
mechanisms linking dispositional optimism or pessimism with daily mood
are unclear from our data. That this relation is due to shared variance with
neuroticism (Smith et al., 1989), an important facet of which is emotional
distress across time and situations, remains a tenable conclusion until specific
mediators are identified.
The ability of dispositional optimism or pessimism to explain individual
differences in the within-person relations between daily-process variables
was examined in the context of symptom reactivity and stress reactivity.
Symptom focus, operationalized as the extent to which participants were
attending to their symptoms during the half-hour before they were signalled
by the electronic interviewer, was implicated in the only significant findings
emerging from these analyses. Optimistic asthma patients were less likely
to focus on their symptoms when they were worsening, and optimistic
fibromyalgia patients were less likely to focus on their symptoms when they
were experiencing undesirable events. This diminished attentional focus on
symptoms when symptoms worsen or when undesirable events occur is likely
an adaptive stress response associated with optimism. It extends to the
dynamics of daily life with illness the conclusion that optimists are advantaged by their reluctance to dwell on problems or stressors (Carver et al.,
1993). And, as our findings demonstrate, it is the presence of optimism,
not the absence of pessimism, which confers this advantage.
OPTIMISM, PESSIMISM, AND DAILY LIFE

159

In each of these studies, we explored yet another way in which optimism

and pessimism might be expressed in daily life with chronic illness. This
concerns the expectation that tomorrow will be a better (or worse) day
than today. The findings summarized below reveal an optimistic bias
underlying the expectation of tomorrows pain or asthma symptoms.

DAILY SYMPTOM EXPECTANCIES:

WILL TOMORROW BE A BETTER DAY?
Included in the nightly booklets completed by both fibromyalgia and
asthma patients was a 0-10 rating (0 = much worse than today; 5 = same as
today; 10 = much better than today) of how severe they expected tomorrows
pain (or asthma symptoms) to be. The mean on the 0-10 expectancy scale
was 5.54 (SD = 1.85) for the fibromyalgia sample and 5.78 (SD = 1.68) for
the asthma sample (recall that a value greater than 5 would be in the
direction of expecting a better day tomorrow). There were nearly two times
as many optimistic evenings (37.4%) as pessimistic evenings (19.0%) when
fibromyalgia participants predicted tomorrows pain, and more than three
times as many optimistic (35.8%) as pessimistic (9.9%) evenings when
asthma patients predicted tomorrows symptoms. Optimistic days exceeded
the number of pessimistic days in 62.9% of the fibromyalgia participants
and in 72.9% of the asthma participants. For further analyses, we identified
an optimistic day as a score of 6 or greater on the symptom expectancy scale
and a pessimistic day as a score of 4 or less.
Do Symptom Expectancies Predict Next-Day Symptoms?

Is the prediction of a better or worse day tomorrow borne out by the

actual change in tomorrows symptoms? It was for those participants with
fibromyalgia, but not those with asthma. This was disclosed by an HLM
Level 1 analysis that modeled the next days pain from the previous days
actual pain along with the previous days prediction of the next days pain.
A more painful day today predicted a more painful day tomorrow (b = .309;
p < .001); but in addition so did a pessimistic prediction about tomorrows
pain (b = 1.12; p < .01). Similarly, an optimistic prediction today predicted
an improvement in tomorrows pain when compared with todays pain (b =
-.87; p < .01). We have identified only one other variable in our daily data
set that is able to predict the change in pain from one day to the next;
namely, the judgment that todays coping strategies were effective in reducing
pain. Yet even when this variable is taken into account, the ability of pain
expectancies to predict pain changes from day to day remains statistically
significant. Because symptom expectancies did not predict next-day asthma
160

AFFLECK ET AL..

symptoms, more study is needed to evaluate the generalizability of these

findings across the spectrum of chronic illness.
Does Dispositional Optimism Predict Daily Symptom Expectancies?

The possibility that daily pain or symptom expectancies might be

shaped in part by dispositional optimism or pessimism was examined through
an HLM Level 2 analysis of intercepts for symptom expectancies. Neither
optimism nor pessimism predicted daily symptom expectancies.

Within-Person Predictors of Symptom Expectancies

If optimism or pessimism had no direct effect on daily symptom expectancies, they could still have an indirect effect through their effects on other
daily processes that might shape symptom expectancies. HLM analyses of the
within-person correlates of optimistic and pessimistic days were conducted at
the day level (Level 1). Table 7.3 summarizes these results for that days
pain or symptom severity, that days mood, that days perceived control
over pain or symptoms, and that days judgment of the efficacy of coping
efforts to reduce pain or symptoms.
Thus far we have uncovered no description of a given day that appears
to predict an optimistic outlook for the next days symptoms. More can be
Table 7.3.
HLM Analysis of Level 1 Covariates of Optimistic and Pessimistic
Expectancies for Tomorrows Pain or Symptoms in Fibromyalgia Patients
( N = 8010 days) and Asthma Patients (N= 1008 days)
Fibromyalgia
patients
Daily variable

Asthma
patients

Optimistic Pessimistic Optimistic Pessimistic

expectancy expectancy expectancy expectancy

Intensity of paidasthma
symptoms
-.001
Attention to painlasthma
.012
symptoms
Pleasant mood
.009
-.002
Unpleasant mood
Personal control of paidasthma
symptoms
-.001
Perceived efficacy of paidasthma
symptoms coping
For painkymptom
improvement
.005
.009
For mood enhancement

.005*
.022

-.005
.007

-.006
-.014*
-.004

-.003
.018

-.001
.001

.011*
.048*

.003
-.008

.019

-.032**

.006
.019

-.033*
.001

Note. Tabular values are unstandardized maximum likelihood estimates.

* p c .05. ** p c .01. *** p c .001.

OPTIMISM, PESSIMISM, AND DAILY LIFE

161

said about the same-day processes that are linked with a pessimistic forecast.
For both the group of patients with asthma and that of patients with
fibromyalgia, a day with more severe pain or symptoms predicted a pessimistic
outlook for the next days symptoms. A pessimistic expectation was also
more likely for those patients with fibromyalgia when they judged that days
coping to be less effective in reducing their pain. For asthma patients, a
pessimistic prediction was also more likely when they had perceived less
personal control over that days symptoms, as well as when they had paid
more attention to that days symptoms.
A multivariate analysis of the significant Level 1 predictors of fibromyalgia pain expectancies showed that almost 7% of the day-to-day variation
in pessimistic expectancies could be explained by that days pain severity and
perceived coping efficacy, although neither was an independent predictor of
a pessimistic day. Approximately 20% of the within-person variance in
pessimistic expectancies for tomorrows asthma symptoms was due to that
days symptom severity, perceived control over symptoms, and attention to
symptoms. Both perceived control and symptom attention, but not symptom
severity, remained significant variables in a days pessimistic expectancy
when all three were considered together in a multivariate HLM analysis.

Discussion
Just as general optimism colors peoples expectations of the future, so
does it color the outlook for the next days symptoms in individuals living
with either of two chronic illnesses. These individuals generally expressed
a more optimistic than pessimistic outlook. In fact, optimistic daily forecasts
outnumbered pessimistic ones by a ratio of 2:l for fibromyalgia patients and
3:l for asthma patients. Whereas asthma patients expectancies did not
predict what tomorrow would actually bring, fibromyalgia patients were as
a group significantly more likely to experience the change they predicted
for tomorrows symptoms, whether they were optimistic or pessimistic.
It was surprising that neither dispositional optimism nor dispositional
pessimism was implicated in patients 24-hour expectancies. While puzzling,
this lack of association between situational expectancies and generalized
expectancies replicates the findings of both Scheier et al. (1989) and Fitzgerald, Tennen, Affleck, and Pransky (1993), who found that domain-specific
and generalized expectancies were inconsistently related among patients
undergoing coronary bypass surgery. The channels through which situational
expectancies might be altered by experiences unrelated to dispositional
optimism and the unique and overlapping contributions of situational and
generalized expectancies require closer scrutiny, particularly because some
investigators have demonstrated that situational expectancies account for
the majority of variance in adjustment (e.g., Bandura, 1977), and others

162

AFFLECK ET AL.

have demonstrated that it depends on the context and the adaptational

outcome selected (Scheier et al., 1989).
We find it noteworthy that an optimistic anticipation of tomorrows
symptoms could not be predicted by the severity of todays symptoms, todays
mood, or todays perception of control or coping efficacy. At the same time,
several of these daily processes could predict a pessimistic forecast. Negative
appraisals of todays ability to achieve personal control over symptoms and
the perception that todays coping was ineffective led patients to expect
tomorrows symptoms to be even worse. Thus, whereas personal traits may
not predict symptom expectancies, there are cognitive processes related to
personal agency that figure in the waxing and waning of a pessimistic outlook.
These processes, however, could not themselves be predicted by either
dispositional optimism or pessimism.

CONCLUSION AND FUTURE DIRECTIONS

Anderssons (199613) meta-analysis of 56 studies using the LOT found
consistent effects of dispositional optimism on the reporting of physical
symptoms, coping strategies, and negative affect. The new analyses presented
here are the first to consider the effects of dispositional optimism on each
of these processes when they are measured on a daily basis to better capture
their fluctuations and moments of change. We conclude, as did Anderson,
that emotional states are the most reliable consequence of optimism and
pessimism. This was replicated in our studies of rheumatoid arthritis, asthma,
and fibromyalgia and suggests that pessimism is mainly a predictor of daily
sadness and optimism a predictor of daily happiness. The mood-regulatory
function of optimism was evidenced as well in optimistic fibromyalgia patients day-to-day judgments of the efficacy of their pain-coping strategies:
Although optimists were not more likely than pessimists to find these strategies successful in alleviating their pain, they were more likely to find them
helpful in improving their mood.
The major departure of our findings from those that have been generated in the larger optimism literature was the lack of association with
momentary pain or symptom reports. We suspect that this discrepancy is
due in part to the considerable difficulty that individuals display in recalling
their illness symptoms and the possibility that optimism and pessimism are
sources of systematic bias in symptom recall. Perhaps optimists recall the
past more favorably than is warranted and pessimists less favorably than is
deserved. Future research should combine retrospective with prospective
daily reports to shed light on the role that optimism and pessimism might
play in shaping recall heuristics.
OPTIMISM, PESSIMISM, AND DAILY LIFE

163

Our idiographic-nomothetic design allowed us to assess the effects of

optimism and pessimism on dynamic bivariate relations that measure the
daily processes of symptom reactivity and stress reactivity at the withinperson level. The yield from these analyses was not great, suggesting that
individual differences other than levels of optimism and pessimism might
underlie these processes. We did find two dynamic effects of optimism and
pessimism, however, that add substantial support to the hypothesis that
optimists cope better with adversity than do pessimists because they are less
likely to dwell on their problems. Despite the fact that optimists and pessimists did not differ in their overall levels of attention to their symptoms,
they did differ in the likelihood that they would be more focused on them
when they worsened or when they were experiencing undesirable events.
This Person x Situation interaction concerning attention deployment merits
more study in other samples and with other contextual variables.
Finally, we introduced another construct for optimism research in the
form of daily expectations for symptom change, which were unrelated to
generalized outcome expectancies. Pessimistic expectations were fewer in
number but could be predicted by negative daily experiences. The days on
which our participants were optimistic about tomorrows symptoms, on
the other hand, could not be predicted. Extending the construct of daily
expectancies to other samples and elaborating it with the anticipation of
other daily experiences should occupy more attention in future research on
optimism and pessimism.

REFERENCES
Affleck, G., & Tennen, H. (1996). Daily processes in coping with chronic pain:
Methods and analytic strategies. In M. Zeidner & N. S. Endler (Eds.),Handbook
of coping (pp. 151-180). New York: Wiley.
Affleck, G., Tennen, H., Urrows, S., & Higgins, P. (1994). Person and contextual
features of stress reactivity: Individual differences in relations of undesirable
daily events with mood disturbance and chronic pain intensity. Journal of
Personality and Social Psychology, 66, 329-340.
Affleck, G., Tennen, H., Urrows, S., Higgins, P., Hall, C., Abeles, M., Newton,
C., & Karoly, P. (1998). Fibromyalgia and womens pursuit of personal goals:
A daily process analysis. Health Psychology, 17, 40-47.
Affleck, G., Urrows, S., Tennen, H., Higgins, P., & Abeles, M. (1996). Sequential
daily relations of sleep, pain intensity, and attention to pain among women
with fibromyalgia. Pain, 68, 363-368.
Anderson, G. (1996a). The role of optimism in patients with tinnitus and patients
with hearing impairment. Psychology and Health, 1 1 , 696-707.

I 64

AFFLECK ET AL.

Andersson, G. (1996b). The benefits of optimism: A meta-analytic review of the

Life Orientation Test. Personality and Individual Differences, 21, 719-725.
Apter, A., Affleck, G., Reisine, S., Tennen, H., Barrows, E., Willard, A., & ZuWallack, R. (1997). Perception of airways obstruction in asthma: Sequential daily
analyses of symptoms, PEFR, and mood. Journal of Allergy and Clinical Immunology, 99, 605-612.
Bandura, A. (1977). Self-efficacy: Toward a unifying theory of behavioral change.
Psychological Review, 84, 191-215.
Bolger, N., DeLongis, A., Kessler, R. C., & Schilling, E. A. (1989). Effects of
daily stress on negative mood. Journal of Personality and Social Psychology,
57, 808-818.
Bolger, N., 6.Schilling, E. A. (1991). Personality and the problems of everyday
life: The role of neuroticism in exposure and reactivity to daily stressors.Journal
of Personality, 59, 335-386.
Bryk, A. S., & Raudenbush, S. W. (1992). Hierarchical linear models. Newbury Park,
CA: Sage Publications.
Carver, C. S., & Gaines, J. G. (1987). Optimism, pessimism, and post-partum
depression. Cognitive Therapy and Research, 11, 449-462.
Carver, C., Pozo, C., Harris, S. D., Noriega, V., Scheier, M. F., Robinson, D. S.,
Ketcham, A. S., Moffat, F. L., Jr., & Clark, K. C. (1993). How coping mediates
the effect of optimism on distress: A study of women with early stage breast
cancer. Journal of Personality and Social Psychology, 65, 375-390.
Carver, C., Pozo-Kaderman, C., Harris, S. D., Noriega, V., Scheier, M. F., Robinson,
D. S., Ketcham, A. S., Moffat, F. L., Jr., & Clark, K. C. (1994). Optimism
versus pessimism predicts the quality of womens adjustment to early stage
breast cancer. Cancer, 73, 1213-1220.
Chang, E. C., DZurilla, T. J., & Maydeu-Olivares, A. (1994). Assessing the dimensionality of optimism and pessimism using a multi-measure approach. Cognitive
Therapy and Research, 18, 143-160.
Cozzarelli, C. (1993). Personality and self-efficacy as predictors of coping with
abortion. Journal of Personality and Social Psychology, 65, 1224-1 236.
Csikszentmihalyi, M., & Larson, R. (1987). The experience sampling method:
Toward a systematic phenomenology. Journal of Nervous and Mental Diseuse,
175, 526-536.
Csikszentmihalyi, M., Larson, R., & Prescott, S. (1977). The ecology of adolescent
experience. Journal of Youth and Adolescence, 6, 281-294.
Curbow, B., Somerfield, M. R., Baker, F., Wingard, J. R., & Legro, M. W. (1993).
Personal changes, dispositional optimism, and psychological adjustment to
bone marrow transplantation. Journal of Behvioral Medicine, 16, 423-443.
Epstein, S. (1977). Traits are alive and well. In D. Magnusson & N. S. Endler
(Eds.), Personality at the crossroads: Current issues in interactional psychology (pp.
83-98). Hillsdale, NJ: Erlbaum.
OPTIMISM, PESSIMISM, AND DAILY LIFE

165

Epstein, S. (1979). The stability of behavior, I: On predicting most of the people

much of the time. Journal of Personality and Social Psychology, 37, 1097-1 126.
Epstein, S. (1983). A research paradigm for the study of personality and emotions.
In M. M. Page (Ed.), Nebraska Symposium on Motivation: Vol. 1 . PersonalityCurrent theory and research (pp. 91-154). Lincoln: University of Nebraska Press.
Epstein, S. ( 1986). Does aggregation produce spuriously high estimates of behavior
stability? Journal of Personality and Social Psychology, 50, 1199-1210.
Fitzgerald, T. E., Tennen, H., Affleck, G., & Pransky, G. S. (1993). Quality of life
after coronary artery bypass surgery: The importance of initial expectancies
and control appraisals. Journal of Behavioral Medicine, 16, 25-43.
Given, C. W., Stommel, M., Given, B., Osuch, J., Kurtz, M. E., & Kurtz, J. C.
(1993). The influence of cancer patients symptoms and functional states on
patients depression and family givers reaction and depression. Health Psychology, 12, 277-285.
Gotlib, I. H. (1984). Depression and general psychopathology in university students.
Journal of Abnormal Psychology, 90, 521-530.
Hedges, S. M., Jandorf, L., & Stone, A. A. (1985). Meaning of daily mood assessments. journal of Personality and Social Psychology, 48, 428-434.
Kanner, A. D., Coyne, J. C., Schaefer, C., & Lazarus, R. S. (1981). Comparison
of two modes of stress measurement: Daily hassles and uplifts versus major life
events. Journal of Behavioral Medicine, 4, 1-39.
Larsen, R. J. (1987). The stability of mood variability: A spectral analytic approach
to daily mood assessments.Journal of Personality and Social Psychology, 52, 11951204.
Larsen, R. J. (1990, June). Second-order consistency: Patterns of change as units
of personality. In J. K. Norem (Chair), Modern units of personality research and
their implications for the issue of consistency. Symposium presented at the meeting
of the American Psychological Society, Dallas, TX.
Larsen, R. J., & Kasimatis, M. (1991). Day-to-day physical symptoms: Individual
differences in the occurrence, duration, and emotional concomitants of minor
daily illnesses. Journal of Personality, 59, 387-424.
Larsen, R. J., & Ketelaar, T. (1991). Personality and susceptibility to positive
and negative emotional states. Journal of Personality and Social Psycholoa,
61, 132-140.
Lazarus, R. S. (1978). A strategy for research on psychological and social factors
in hypertension. Journal of Human Stress, 4, 35-40.
Linton, S. J. (1991). Memory for chronic pain intensity: Correlates of accuracy.
Perceptual and Motor Skills, 72, 1091-1095.
Litt, M. D., Tennen, H., Affleck, G., 61 Klock, S. (1992). Coping and cognitive
factors in adaptation to in vitro fertilization failure. Journal of Behvioral Medicine, 15, 119-126.
Lon, M., & McNair, D. M. (1982). Profile of Mood States-B. San Diego, CA:
Educational and Industrial Testing Service.

166

AFFLECK ET AL.

Marshall, G. N., Wortman, C. B., Kusulas, J. W., Hervig, L. K., & Vickers, R. R.
(1992). Distinguishing optimism from pessimism: Relations to fundamental
dimensions of mood and personality. Journal of Personality and Social Psychology,
62, 1067-1074.
Mendola, R. A. ( 1990). Coping with chronic pain: Perceptions of control and dispositional
optimism as moderators of psychological distress. Unpublished doctoral dissertation, University of Connecticut, Storrs.
Mroczek, D. K., Spiro, A., Aldwin, C. M., Ozer, D. J., & BossC, R. (1993). Construct
validation of optimism and pessimism in older men: Findings from the normative aging study. Health Psychology, 12, 406-409.
Robinson-Whelen, S., Kim, C., MacCallum, R. C., & Kiecolt-Glaser, J. K. (1997).
Distinguishing optimism from pessimism in older adults: Is it more important
to be optimistic or not to be pessimistic? Journal of Personality and Social
Psychology, 73, 1345-1353.
Scheier, M. F., & Carver, C. S. (1985). Optimism, coping, and health: Assessment
and implications of generalized outcome expectancies. Health Psychology, 4 ,
2 19-247.
Scheier, M. F., & Carver, C. S. (1987). Dispositional optimism and physical wellbeing: The influence of generalized outcome expectancies on health. Journal
of Personality, 55, 169-2 10.
Scheier, M. F., & Carver, C. S. (1992). Effects of optimism on psychological
and physical well-being: Theoretical overview and empirical update. Cognitive
Therapy and Research, 16, 201-228.
Scheier, M. F., Carver, C. S., & Bridges, M. W. (1994). Distinguishing optimism
from neuroticism (and trait anxiety, self-mastery, and self-esteem): A reevaluation of the Life Orientation Test. Journal of Personality and Social Psychology,
67, 1063-1078.
Scheier, M. F., Matthews, K. A., Owens, J. F., Magovern, G. J., Sr., Lefebvre, R. C.,
Abbot, R. A., & Carver, C. S. (1989). Dispositional optimism and recovery
from coronary artery bypass surgery: The beneficial effects on physical and
psychological well-being. Journal of Personality and Social Psychology, 57, 10241040.
Shiffman, S., Fischer, L. A., Paty, J. A., Gnys, M., Hickcox, M., & Kassel, J. D.
(1994). Drinking and smoking: A field study of their association. Annuls of
Behavioral Medicine, 16, 203-209.
Shifren, K. (1996). Individual differences in the perception of optimism and disease
severity: A study among individuals with Parkinsons Disease. Journal of Behavioral Medicine, 19, 24 1-27 1.
Shifren, K., & Hooker, K. (1995). Stability and change in optimism: A study of
spouse caregivers. Experimental Aging Research, 21 , 59-76.
Smith, T. W., Pope, M. K., Rhodewalt, F., & Poulton, J. L. (1989). Optimism,
neuroticism, coping, and symptom reports: An alternative interpretation of the
Life Orientation Test. Journal of Personality and Social Psychology, 56, 640-648.
OPTIMISM, PESSIMISM, AND DAILY LIFE

167

Stone, A. A., & Neale, J. M. (1982). Development of a methodology for assessing

daily experiences. In A. Baum & J. E. Singer (Eds.), Adwances in enwironmental
psychology, environment and health (Vol. 4, pp. 49-83). New York: Erlbaum.
Stone, A. A., & Neale, J. M. (1984a). The effects of severedaily events on mood.
Journal of Personality and Social Psychology, 46, 137-144.
Stone, A. A., & Neale, J. M. (1984b). New measure of daily coping: Development
and preliminary results. Journal of Personality and Social Psychology, 46,892-906.
Stone, A. A., & Shiffman, S. (1994). Ecological momentary assessment (EMA) in
behavioral medicine. Annals of Behwioral Medicine, 16, 199-202.
Taylor, S. E., Kemeny, M. E., Aspinwall, L. G., Schneider, S. G., Rodriguez, R.,
& Herbert, M. ( 1992). Optimism, coping, psychological distress, and high-risk
sexual behavior among men at risk for acquired immunodeficiency syndrome
(AIDS). Journal of Personality and Social Psychology, 63, 460-473.
Tennen, H., Affleck, G., Urrows, S., Higgins, P., & Mendola, R. (1992). Perceiving
control, construing benefits, and daily processes in rheumatoid arthritis. Canadian Journal of Behavioral Science, 24, 186-203.
Tennen, H., Suls, J., & Affleck, G. (1991). Personality and daily experience: The
promise and the challenge. Journal of Personality, 59, 313-338.
Watson, D., & Clark, L. A. (1984). Negative affectivity:The disposition to experience aversive emotional states. Psychological Bulletin, 96, 465-490.
Zautra, A. J., Affleck, G., & Tennen, H. (1994). Assessing life events among older
adults. In L. M. Powell, J. A. Teresi, (Eds.), Annual rewiew of gerontology
and geriatrics: Focus on assessment techniques. Annual Review of gerontology and
geriatrics, Vol. 14 (pp. 324-352). New York: Springer.
Zautra, A. J., Guamaccia, C. A., Reich, J. W., & Dohrenwend, B. P. (1988). The
contribution of small events to stress and distress. In L. H. Cohen (Ed.),
Life ewents and psychological functioning (pp. 123-148). Newbury Park, CA:
Sage Publications.

168

AFFLECK ET AL.

OPTIMISM AND PESSIMISM:

BIOLOGICAL FOUNDATIONS
MARVIN ZUCKERMAN

Thinking rosy futures is as biological as sexual fantasy. Optimistically calculating the odds is as basic a human action as seeking food when hungry
or craving fresh air in a dump. Making deals with uncertainty marks
us [as a species] as plainly as bipedalism.
(L. Tiger, 1979, p. 35)

Recently there was a lottery drawing with an extraordinary payoff of

many millions of dollars. As I drove home I noticed long lines of people
standing outside of newsstands. My first thought was that some momentous
world event had occurred, and they were lining up to buy newspapers. Only
later did I realize they were all buying lottery tickets. The odds of anyone
winning were more than those of being struck by lightning. In fact if the
bettors responded pessimistically to the shorter odds of getting struck by
lightning, they would never leave their houses when it was raining. This
tells us something about the biological value of optimism as contrasted with
pessimism. Optimism leads to action with a possibility of reward no matter
how remote the possibility, but pessimism results in inaction and paralysis

169

in the necessary jobs of life, whether hunting woolly mammoths or starting

a new business. Nothing ventured, nothing gained is the motto of capitalism and perhaps it is wired in to our species.
Tigers quotation may be startling to many people who think of optimism as a purely learned attitude conditioned by the outcomes of events
in life. Tiger speculates that an optimistic attitude had adaptive value in
the struggles for life of our hominid ancestors over the million or so years
of evolution. Seligman (1991) has shown that it still is an advantage for
success in current life, whether in politics, business, or sports. Risk taking
in novel or unpredictable situations is a necessity for exploration. A species
such as homo supiens, which came out of Africa and colonized the entire
globe within the relatively short time span of 100,000 years, needed an
optimistic outlook for success in adaptation to sometimes hostile new environments. A pessimistic species would have stayed in one locale, gathering
fruit and nuts and hunting small animals until exhaustion of their food
resources led to their extinction. They would have avoided the dangers of
seeking mates outside of their own group and hastened their demise as a
species through the deleterious effects of inbreeding.
The evolutionary hypothesis suggests that something genetic underlies
the trait that is selected, but it does not guarantee that variations in the
genetic structure are strong factors in variation in the trait among current
members of the species (heritability). Negative outcomes of life experiences,
particularly uncontrollable outcomes, reinforce a pessimistic attitude (Seligman, 1991). Those people strongly disposed toward optimism, however,
seem to rebound quickly after the worst experiences. Voltaires character,
Candide, is a caricature of optimism created by his cynical, pessimistic
author. Seligman (1991) has pointed out that normal, happy optimists are
less realistic in their expectations than are depressed pessimists. However,
if they persist long enough, optimists do get reinforced, unlike the pessimists who attempt nothing and therefore miss many rewards.
Of course one could argue that because optimism fosters risky choices
and pessimism often leads to caution (rather than complete inaction) the
pessimist would be more likely to survive in a risky, hostile environment.
The optimist takes on the sabre-toothed tiger, while the pessimist runs for
the safety of the cave. It would seem that an optimal balance between
optimism and pessimism that is sensitive to the odds of risk would probably
be the best strategy for survival. Fortunately, as with most traits, optimism
and pessimism are normally distributed in the population. The extremes
within a group may provide a balance in group decision making.
Evolutionary psychology provides explanations for distal causes of the
general characteristics of a species relative to other species in terms of
the environmental challenges faced by the species and of their physical
capabilities in dealing with those challenges. Novel open environments are

170

MARVIN ZUCKERMAN

generally aversive to species such as rats, often used for an animal model
of human behavior. This is understandable in terms of the dangers from
predators both in the air and on the ground when the rat is out of its burrow.
But rats must forage to survive, so they do so with caution. We find marked
variations in the relative balance between exploratory activity and fearful
freezing between subspecies of rats and mice when they are placed in an
open field (Plomin, DeFries, McClearn, & Rutter, 1997). Similar problems
were faced by our hominid ancestors, but unlike rodents, they did not hide
in burrows but freely roamed the savannahs of Africa and the tundras of
the Arctic pursuing large, dangerous game.
Evolutionary psychology is addressed to the general, if not universal,
characteristics of a species. But this approach does not necessarily tell us
much about the sources of individual differences within a species. To understand what biological factors might explain the variations in optimism and
pessimism in the current members of our species, we must turn to fields such
as behavior genetics, neuropsychology, and neuropharmacology. I attempt to
do that in this chapter.
Tiger (1979) and Gould (1981) tend to denigrate psychologys attempt
to operationally define and measure traits such as intelligence in humans.
Without quantification of traits and related behavior one cannot begin to
study the phenomena in any meaningful manner. In this respect, psychology
is not different than any other science. One can question the reliability
and construct validity of our measures, but these are scientific questions
capable of answer, not imponderables only approachable from the philosophers armchair. The first measures of personality and intellectual ability
traits emerged in this century, and improved methods will undoubtedly be
developed in the future. But our current measures have already provided
some answers to the nature-nurture questions, even at the level of molecular genetics.
Personality traits such as optimism and pessimism may be studied at
many levels, from the generalized trait level to their foundations in the
DNA, or genetic level (Zuckerman, 1991, 1993). In between are other
levels: of behavioral expression, and of learning and cognition, physiology,
biochemistry, and neurology. In the section of this chapter discussing traits,
I attempt to place optimism and pessimism within the context of personality
and emotional trait models. I will illustrate this placement with data from
a recent study (Zuckerman, Joireman, Kraft, & Kuhlman, 1999). From the
top (traits) I go directly to the bottom (genetics). In that section I will
describe two studies of the genetics of optimism and pessimism traits.
Certain forms of psychopathology represent abnormal extremes of normal traits. Anxiety disorders, for instance, can be regarded as extreme
expressions of the general trait of neuroticism. Optimism and pessimism find
their most extreme expressions in the bipolar and unipolar mood disorders.
BIOLOGICAL FOUNDATIONS

171

Although optimism is usually a healthy attitude in normally functioning

people, in those experiencing mania it is expressed in wild, impulsive, and
risky behavior and grandiose schemes. In the depressive phase of the bipolar
disorder or in the episodic depressions of the unipolar mood disorder, we
see a severe kind of pessimism expressed in a generalized hopelessness. Some
of the genetics and biochemistry of these disorders may be relevant to the
biological bases of optimism and pessimism.
In the next part of the chapter, I deal with the normal range of
personality traits. A general discussion of the psychobiology of personality
(see Zuckerman, 1991) is beyond the scope of this chapter; however, I
will suggest the relatedness of pessimism to the biological traits underlying
neuroticism, and the shared biological foundations of optimism with those
for extraversion, impulsivity, and sensation seeking traits.

THE TRAITS
The reader may have noticed that I use the conjunction and when
referring to optimism and pessimism rather than using a hyphen. Because
optimism and pessimism are grammatical antonyms it has been assumed
that they constitute a bipolar trait, similar to extraversion-introversion or
dominance-submission. Of course, the nonpsychologist thinks in terms of
types, so that a person is either an optimist or a pessimist, and they betray
this type of thinking when they ask the question: What proportion of the
population are optimists? But like most traits, optimism and pessimism are
continuously and normally distributed. What was not expected is that they
are distributed on two different, even if correlated, dimensions (Chang,
DZurilla, & Maydeu-Olivares, 1994). Ball and Zuckerman ( 1990) developed
the Generalized Reward and Punishment Expectancy Scales (GRAPES) by
factor analyzing items designed to assess generalized reward expectancy and
other items for generalized punishment expectancies. The analysis yielded
two factors rather than a bipolar factor. Torrubia, Avila, Molt6, and Grande
(1995) developed scales for susceptibility to cues for punishment (SP) and
susceptibility to cues for reward (SR) based on Grays (1982) model for
personality. The scales assess short-term motivational and expectational
effects of such cues. The two scales are uncorrelated.
Traits that are originally conceived of as bipolar (such as positive and
negative affect), but turn out to be distinct and separate factors require a
new conceptualization because what they measure, in terms of behavioral
expressions and their biological and social origins, may be different. One
explanation for their relative independence may be that they are situationspecific, rather than general traits, so that an individual may be optimistic
in one class of situations, such as social interactions, but pessimistic in

172

MARVIN ZUCKERMAN

another, such as financial outcomes. There is certainly some specificity in

these traits, but their good internal reliabilities also suggest there is a latent
trait across situations. Another possibility is that these traits are subtraits
of broader ones, such as extraversion-introversion and neuroticism, which
are independent dimensions.
These are the kinds of questions that led us to an examination of the
factorial relationships among personality, motivational, cognitive, and affect
traits (Zuckerman, Joireman, Kraft, & Kuhlman, 1999). The personality
traits included extraversion, neuroticism, and psychoticism from the Eysenck
Personality Questionnaire-Revised (EPQ-R; Eysenck, Eysenck, & Barratt,
1985) and sociability, neuroticism, impulsive sensation seeking, aggression,
and activity from the Zuckerman-Kuhlman Personality Questionnaire
(ZKPQ; Zuckerman, Kuhlman, Joireman, Teta, & Kraft, 1993).
The GRAPES scales (Ball & Zuckerman, 1990) are optimism and
pessimism scales similar to those in the LOT scale (Scheier & Carver, 1985).
An example of a generalized reward expectancy or optimism item is: I
expect I will rise to the top of any field of work I am or will be engaging
in. A punishment expectancy item is: It is likely that most of us will have
a serious car accident at some point in our lives.
The GRAPES assesses general expectancies of outcomes in the distant
future. Torrubia et al.s (1995) SP and SR scales represent immediate expectancies and their motivational effects. For instance, DOyou often do things
to be praised? (SR), or DOyou often refrain from doing something because
you are afraid of it being illegal? (SP).
The affect traits assessed by the Multiple Affect Adjective Check ListRevised (MAACL-R; Lubin & Zuckerman, 1999; Zuckerman & Lubin,
1985) includes three negative affects (anxiety, depression, hostility) and
two positive affects (positive affect and sensation seeking, a kind of active
or surgent type of positive affect).
The personality, cognitive, motivational, and affect traits were intercorrelated and factor analyzed in one sample and then subjected to a Procrustes rotation in a replication sample. The results in the Procrustes rotation
are shown in Table 8.1. Three factors were derived and labeled in terms of
the personality factors defining them: Neuroticism (N), Extraversion (E),
and Psychoticism-Impulsive Unsocialized Sensation Seeking (P-ImpUSS).
The last factor has been previously obtained in factor analyses using threefactor solutions (Zuckerman, Kuhlman, & Camac, 1988; Zuckerman, Kuhlman, Thornquist, & Kiers, 1991). It includes the EPQs Psychoticism, and
the ZKPQs Impulsive Sensation Seeking and Aggression-Hostility scales.

A Procrustes rotation is one that is constrained by conditions established in advance so that the
loadings of variables on the factors cannot exceed the maximum amounts set beforehand.

BIOLOGICAL FOUNDATIONS

173

Table 8.1.
Factor Analysis and Procrustes Rotation of Personality, Cognitive,
Motivational, and Affect Traits
Variables

N-Anxiety

E-Sociability

P-ImpUSS

.81

-.02

.34
-.04

EPQ: Neuroticism
EPQ: Extraversion
EPQ: Psychoticism
ZKPQ: Neuroticism/Anxiety
ZKPQ: Sociability
ZKPQ: Impulsive SS
ZKPQ: Aggression/Hostility
ZKPQ: Activity
Gen. Punishment
Expectancy
Gen. Reward expectancy
Sens. signals of punishment
Sens. signals of reward
MAACL: Anxiety
MAACL: Depression
MAACL: Hostility
MAACL: Positive Affect
MAACL: SS (Surgent) Affect
FCC/TCC

-.I7
-.36
.75
-.01

-.35
.I3
-.07
.65

-.38
.78

.I9

.74
.53
.46

-.25
-.21
.97

.87
.10
.01
.68
.41

.23

.76

.25
-.I3
.54
.60

.37
.02

-.02
.13

.55

-.I4
.I5
.37
.44

-.30
.59

-.08
-.32
.02
.44
.83

.98

.56
.66
-.57

.05
.97

VCC

.99
1 .oo
.98
.98
1 .oo
.99
1 .oo
.82
.95
.93
.99
.98
.96
.97
.91
.99
.95
.97

Nofe. N = neuroticism; E = extraversion; P-ImpUSS = psychoticism-impulsiveunsocialized sensation seeking; EPQ = Eysenck Personality Questionnaire; ZKPQ = Zuckerman-Kuhlman Personality Questionnaire;
Gen. = Generalized; Sens. = Sensitivity to; MAACL = Multiple Affect Adjective Check List-Revised; SS =
Sensation Seeking; VCC = variable congruence coefficient; FCC = factor congruence coefficient; TCC =
total congruence coefficient. Boldface numbers indicate the primary loadings of variables on one or two of
the three factors.

The GRAPES punishment expectancy (pessimism) and the Torrubia et

al. ( 1995) susceptibility to punishment scales had strong relationships with
the Neuroticism factor (N), and the GRAPES reward expectancy (optimism)
and the susceptibility to reward scales had moderate relationships with the
Extraversion factor. The MAACL-R negative affects, anxiety in particular,
were strongly related to theN factor, whereas the positive affect scales, particularly the active, surgent type of positive affect (SSPA; e.g., adventurous, daring, enthusiastic, energetic, wild), were highly related to the E factor.
Because only one of the tests in the study, the ZKPQ, measured five
personality factors, the factor analysis was able to produce only three significant factors. Table 8.2, however, shows the correlations between the Generalized Reward and Punishment Expectancy scales and all of the other personality and affect trait scales in the derivation and replication samples. Only
the replicated correlations are described below. In both samples generalized
reward expectancy correlated positively with EPQ extraversion; ZKPQ activity; MAACL positive and sensation-seeking (surgent) affects; and negatively
with EPQ and ZKPQ measures of neuroticism, and MAACL anxiety and

1 74

MARVIN ZUCKERMAN

Table 8.2.
Correlations Between Generalized Reward and Punishment Expectancy
Scales and Personality and Affect Trait Scales
Reward expectancy
Expectancy
EPQ: Extraversion
EPQ: Neuroticism
EPQ: Psychoticism
ZKPQ: Sociability
ZKPQ: Neuroticism-Anxiety
ZKPQ: Impulsive Sensation
Seeking
ZKPQ: Aggression-Hostility
ZKPQ: Activity
MAACL: Anxiety
MAACL: Depression
MAACL: Hostility
MAACL: Positive Affect
MAACL: Sensation Seeking
Affect

Sample 1

.38*
-.31*
.03
.I8
-.44*
.I7

.oo

.32*
-.23*
-.23*
-.05
.31

.41

Sample 2

.48*
-.37*
.04
.I9
-.2r

.26*
-.04
.30*
-.29*
-.39*
-.I5
.44*
.45*

Punishment expectancy
Sample 1

-.03
.45*
-.09
-.07
.41*

-.I6
.08
-.06
.21*
.22*
.I2
-.I3
-.I9

Sample 2

-.I4
.50*
-.07
-.I 1
.39*

-.I3
.25'
-.05
.41*
.40*
.31*
-.21*

-.I0

Note. EPQ = Eysenck Personality Questionnaire; ZKPQ = Zuckerman-Kuhlman Personality Questionnaire;

MAACL: Multiple Affect Personality Questionnaire-Revised. Sample 1, n = 188; sample 2, n = 135
' p < .0001.

depression. Generalized punishment expectancy correlated positively with

EPQ and ZKPQ measures of neuroticism and with MAACL measures of
anxiety and depression in both samples.
Thus cognitive, motivational, and affect traits are organized primarily
within the two major and independent personality factors E and N. The
relative independence of the GRAPES reward and punishment expectancy
scales (they correlated -28) and susceptibility to reward and punishment
scales ( r = .05) resembles the relative independence of optimism and pessimism scales found by Chang et al. ( 1994).The generalized reward expectancy
scale is positively associated with extraversion and positive affects and negatively associated with neuroticism and the negative affects of anxiety and
depression. Generalized punishment expectancy is positively associated with
neuroticism and the negative affects of anxiety and depression. The associations of optimism and pessimism with personality dimensions of E and N and
positive and negative affect traits suggest that these attitudinal dimensions
may share some of the genetic and other biological factors with E and N.

GENETICS OF OPTIMISM AND PESSIMISM

Although personality dimensions such as E and N have been subjected
to genetical analyses in many twin and adoption studies, only two twin
BIOLOGICAL FOUNDATIONS

1 75

studies have been published on the genetics of optimism and pessimism.

Genetic-twin studies compare similarities or differences of a trait for identical
and fraternal twins. Identical twins raised in the same families have all their
genes in common and also share the same family environment. Fraternal
twins also share the same family environment but have, on average, only
half of their genes in common. The difference in similarity, as expressed in
correlations, should control for the effects of shared environments, age, and
gender (if only same-gendered fraternal twins are used). If genetic effects
play no role relative to that for shared environment, then the correlations
should be the same in identical and fraternal twins, but if genetic effects
are significant, then the correlation should be significantly higher in identical
twins than in fraternal twins. The ratio of the correlation of identical twins
to that of fraternal twins should be about 2:l if only polygenetic additive
effects are involved, but the ratio may be more than 2:l if other types of
genetic mechanisms, such as dominance or epistasis (unique combinations
of genes rather than simple additive effects) are involved in the trait.
Heritability of a trait is the proportion of variance that can be attributed to
genetic factors alone.
Schulman, Keith, and Seligman ( 1993)gave the Attributional Style
Questionnaire (ASQ; Peterson et al., 1982) to 115 pairs of identical and
27 pairs of fraternal twins. On a composite score summing reactions to
positive and negative events, the correlations were .48for identical and 0 for
fraternal twins. The difference between correlations indicates a substantial
hereditary effect, but the 0 correlation in fraternal twins either indicates
that the trait is entirely epistatic in genetic mechanism or that the correlation
of the fraternal twins is unreliable because of the low number of participants
in this group. The latter interpretation is more likely, but another anomalous
finding emerged when the good and bad event scales were analyzed separately. For positive events, the analysis also showed a moderate correlation
of S O for identical twins, but the correlation for fraternal twins was nearly
as high (.41),which would ordinarily indicate a substantial effect for shared
environment. In contrast, the scale for bad events showed the same pattern
as the composite scale, with correlations of .43 for identical and -.03 for
fraternal twins. Schulman et al. (1993)note that the optimism score for
positive events is more responsive to environmental experience and that it is
the respondents attributions about negative events that affect their behavior
more. They suggest that optimism is not directly heritable and that the
heritable component is a function of other traits, such as intelligence,
attractiveness, or certain temperament traits, which are more heritable and
affect the amount of failure or success in life experience. The actual outcomes
of events then determine the generalized attitudes of optimism or pessimism
toward future events. In another study, Seligman and his colleagues (Seligman et al., 1984) found that childrens optimistic or pessimistic styles for

176

MARVlN ZUCKERMAN

interpreting bad events were correlated (T = .39) with their mothers styles,
but not with their fathers styles. A relationship with the mothers style and
not the fathers is not readily explainable in terms of genetics unless a major
gene involved is on the X chromosome. In that case boys could only inherit
it from their mothers.
The twin study by Plomin et al. (1992) used the LOT scale, which is
scored separately for optimism and pessimism. Unlike earlier studies in which
the negative correlations between the two scales were moderate, optimism
and pessimism scales were uncorrelated ( r = .02). In addition to substantial
numbers of identical (126) and fraternal (146) twins raised together, the
study also included substantial groups of identical (72) and fraternal (178)
twins raised apart. The correlations for identical twins raised apart constitute
direct measures of heritability because heredity is presumably the only factor
that could make them alike. These correlations were .28 for LOT optimism
and .41 for LOT pessimism. However estimates of heritability based on all
twins in the study using a model-fitting method yielded heritability values
of .23 for LOT optimism and .27 for LOT pessimism. Some evidence of
nonadditive types of genetic variance and shared environmental influences
were found for optimism but not for pessimism.
Heritabilities for optimism and pessimism were lower than those typically found for personality traits such as extraversion and neuroticism, which
average about .40 in recent studies and S O in older studies (Bouchard, 1993;
Loehlin, 1992). They are closer to those found for nonreligious social attitude
items (Bouchard, Lykken, McGue, Segal, &. Tellegen, 1990).
The lower heritabilities of optimism and pessimism suggest that their
heritability may be indirect, as suggested by Schulman et al. (1993). Based
on our previous analyses of our own measures of optimism and pessimism
(GRAPES), I would suggest that optimism is partly based on the genetic
bases for extraversion, and that pessimism is partly a function of genes for
neuroticism. The influence of some effects of nonadditive genetic mechanisms and shared environmental effects for optimism is interesting in view
of the findings in two studies of twins raised together and those raised apart
in which there was evidence of nonadditive genetic effects and shared
environments for extraversion but not for neuroticism (Pederson, Plomin,
McClearn, Csr Friberg, 1988; Tellegen et al., 1988).
But even if there is some evidence for shared environmental effects
in learned optimism (Seligman, 1991) the larger source of optimism is
in the nonshared environment or the people and experiences that affect
one twin but not the other. This is generally true for broader personality
traits as well. Particular life experiences of loss and disappointment in
relationships or at work may attenuate optimism and increase pessimism
for future events. The effects of such events are probably influenced by
temperament. It takes many and severe bad happenings to discourage optiBIOLOGICAL FOUNDATIONS

I77

mists like Candide and Don Quixote for whom it always remains the best
of all possible worlds.

PSYCHOPATHOLOGY
Optimism and pessimism have not been studied directly in terms of
their biological correlates. In a literature search of the psychological abstracts
by database, I found few conjunctions of the terms optimism or pessimism
with biological terms. Therefore, nearly all of what is to be said about the
topic of this chapter depends on extrapolation from studies of mood and
anxiety disorders in which the attitudes are found in extreme forms. Of
course such extrapolation is dangerous, because even within the disorders,
there is a great deal of variation in these cognitive phenomena. The biological factors underlying the disorders may not be the same as those influencing
the attitudes of optimism and pessimism. Despite this limitation there is
ample evidence of a close relationship between the disorders and the attitudes
toward the future. When the depression is gone the pessimistic attitude
usually disappears with it.
Nearly all forms of psychopathology involve genetically influenced
vulnerabilities and stressful life experiences (Zuckerman, 1999). The drugs
that are effective for treating certain disorders provide us with clues to the
biological bases of the disorders and their prominent symptoms. Antipsychotic drugs reduce the hallucinations and delusions of schizophrenia, and
antidepressant drugs reduce the depressive cognitive outlook of the severely
depressed patient. The actions of these drugs in the brain may help us
understand some symptoms of the disorders.
Optimism and pessimism are important cognitive traits in many disorders, but are particularly salient symptoms in the mood disorders. The shortterm unrealistic optimism of the grandiose and impulsive manic episode
can be transformed within days or hours to the pessimism and hopelessness
of the depressive phase in patients with the rapid cycling bipolar (manicdepressive) disorder (Cutler & Post, 1982). Changes in the biochemistry
during such mood shifts illustrate how the central neurochemical balances
may affect mood and cognition.
Bipolar disorder is highly heritable (Gershon, 1990; Kendler, Pederson,
Johnson, Neale, & Mathe, 1993). All of the primary brain monoamine
neurotransmitters (norepinephrine, dopamine, and serotonin) have been
implicated in some studies of the bipolar disorder (Goodwin & Jamison,
1990). The catecholamines (norepinephrine and dopamine) have been implicated in the switches from depression to mania in rapid cycling cases,
although it is not clear whether these are causes or effects of the activity
changes going from one state to the other. However drugs that increase

178

MARVIN Z UCKERMAAJ

activity in these systems can precipitate the switch in state. One of the
factors that may be important in the stability or lability of these systems,
particularly dopamine, is the general level of the enzyme monoamine oxidase
(MAO). This enzyme regulates the level of monoamines by catabolic reduction of the neurotransmitter within the neuron. MAO-B is low in the blood
platelets of patients with bipolar disorders, where they are in the ill or the
well state. The association of MA0 with bipolar disorder is interesting in
view of the similar association of the sensation seeking trait with bipolar
disorder and MA0 (Zuckerman, 1985).
People with bipolar disorder tend to score high on the Sensation
Seeking Scale whether they are in the manic, normal, or even depressed
states. MA0 levels are low in high sensation seekers and are also low in
people whose behaviors and habits are associated with the trait of sensation
seeking, such as tobacco, drug, and heavy alcohol use; gambling; and criminality (Zuckerman, 1994; Zuckerman, Buchsbaum, & Murphy, 1980). Low
M A 0 levels are also found in patients with other disorders characterized
by disinhibition, such as those with borderline and antisocial personalities
and drug abuse disorders. M A 0 is nearly totally heritable and a gene for
the B type has been located on the X chromosome.
Just as the manic state is associated with optimism, the depressive state
is accompanied by pessimism. One of Becks (1972) three major depressive
schemas is negative views of the future, or hopelessness. Abramson, Metalsky, and Alloy (1989) describe a hopelessness depression characterized by
retarded behavior, lack of energy, apathy, lack of motivation, sleep disturbance, difficulty in concentration, negative cognitions, sad affect, and suicidal preoccupation. Hopelessness is regarded as the proximal cause of depression, whereas attributional styles, such as attribution of negative events to
internal, stable, and global causes are conceived of as the predisposing causes
for hopelessness and depression. It must be emphasized that the hopelessness
or extreme pessimism of depressed individuals is a state that accompanies
the mood change, and does not precede the depression, and usually dissipates
when the depression is in remission. The attribution styles, however, may
be more traitlike, and there is evidence that they remain elevated after
treatment (Eaves & Rush, 1984).
Unipolar major depression, particularly that of the more severe hopelessness or melancholic type, shows substantial heritability, but that of the
milder or neurotic type shows little or no evidence of heritability (Nurnberger
& Gershon, 1992; Zuckerman, 1999). Deficits in brain norepinephrine (NE)
or serotonin have been suggested as underlying the pathology of major
depression (Maes & Meltzer, 1995; Schatzberg & Schildkraut, 1995). Drugs
that activate either NE, serotonin, or both systems have been efficacious
in the treatment of depression (Burke & Preskom, 1995). Some patients
respond more to NE-potentiating drugs, whereas others respond better to
BIOLOGICAL FOUNDATIONS

179

serotonin-selective drugs, suggesting that there may be two types of etiology

for depression.
Neuroticism and anxiety traits are related negatively to reward expectancy, and they are related positively to punishment expectancy, as shown
in Tables 8.1 and 8.2, therefore, anxiety disorders would be expected to show
a similar relationship to generalized expectancies. Like depressive disorders,
anxiety disorders are characterized by negative expectations for the future.
People suffering from panic disorder believe that there will be harmful
physical consequences of their panic attacks (McNally, 1990), and people
with agoraphobia expect to have panic attacks if they venture outside of
their homes. Patients with generalized anxiety disorder spend nearly every
day worrying about possible dire events. In the specific phobias the expectations of negative consequences are less generalized and are confined to
specific situations but play a prominent role in the avoidance patterns that
develop out of the fearful expectations. These people are pessimistic about
their abilities to cope with the feared situation and about what will happen
to them if they enter into the situation or encounter the feared object.
Heritabilities for most of the anxiety disorders estimated from twin
studies are modest (about 23-39%) but are even lower if not absent for
general anxiety disorder and situational specific phobias (Kendler, Neale,
Kessler, Heath, & Eaves, 1992a, 199213). Panic disorder and phobic panic
reactions in the phobic situations are mediated through a system in which
norepinephrine is the principal neurotransmitter. At low levels of activity,
this system mediates the sensitivity to signals of punishment (Gray, 1982)
or short-term negative expectancies of punishment triggered by immediate
cues in the environment. The system has also been described as an alarm
system (Redmond, 1987). At high levels of activity, the system plays a
role in panic attacks. Drugs that cause sudden increases in the activity of
the system precipitate panic attacks in persons with panic disorders, and
drugs that reduce activity in the system reduce anxiety (Price, Goddard,
Barr, & Goodman, 1995). However other drugs, such as sodium lactate,
that do not potentiate the noradrenergic system have the capacity to elicit
panic reactions in patients with panic disorders. It is also interesting that
drugs such as the benzodiazepines that inhibit the noradrenergic neurons
by potentiating the GABA-inhibitory system are of benefit in generalized
anxiety disorders but of little help in panic disorders. O n the other hand,
drugs that activate the serotonergic system have been efficacious in treating
panic disorder, mixed anxiety and depression, and obsessive-compulsive
disorder (Shader & Greenblatt, 1995).
Most of the drugs that have strong effects on panic disorder do not
show these effects in normal control participants or in patients suffering
most other types of anxiety disorders. Whatever is responsible for the initial
spontaneous panic attacks, later attacks could be triggered by anything that

180

MARVlN ZUCKERMAN

causes arousal of autonomic activity and bodily cues associated with panic
attacks. Expectations of attacks may be a self-fulfilling prophecy based on
conditioning of external and internal cues associated with the initial
attacks.

PERSONALITY
Extraversion, activity, sensation seeking, and impulsivity all may involve a general approach mechanism (Gray, 1982, 1987; Zuckerman, 1984,
1995),whereas neuroticism and anxiety are aspects of a behavioral inhibition
mechanism (Gray, 1982). Theory and comparative research with nonhuman species have linked the approach mechanism to activity of the
mesolimbic dopaminergic system involved in reward effects in the brain.
At the cognitive level, the tendency to approach, or the relative balance
between approach and inhibition, might be optimism and pessimism in
terms of generalized expectancies of outcomes. This is particularly true in
reaction to novel stimuli or situations where there has been no previous
experience that could be used to predict outcomes. In this case one would
have to fall back on generalized expectancies of outcomes in the decision
to approach or enter into the situation. Reactivity of the dopaminergic
system might be what underlies optimism as a trait or state. Of course
previous experience in stressful situations where there was unavoidable stress
may produce the pessimistic attitude of learned helplessness (Seligman,
1975) and a weakened approach and powerful inhibition mechanism. T o
the extent that the strengths of the approach and inhibition mechanisms
depend upon genetically influenced brain systems, they may influence the
generalized expectancies of reward and punishment. However, data from
the two studies of the genetics of optimism and pessimism suggest only weak
heritability influence relative to that in the personality traits. Some shared
environmental influences were suggested for optimism, but not for pessimism.
This could indicate that although parents and life experiences influence
optimism, only life experiences outside of the family (nonshared) and the
genetic predisposition affect pessimism.
The genetic component of pessimism may be shared with that for
neuroticism, just as the genetic component of optimism is a joint function
of shared genetic influences from extraversion and neuroticism. Plomin et
al. (1992) chose not to control for neuroticism in their multivariate genetic
analyses because it so lowly correlated with optimism and pessimism, but
correlations in our study are much higher. If we can assume that the same
biological factors influence pessimism as influence neuroticism, this would
be the place to start if there is really a biology of pessimism. Neuroticism
and trait anxiety are so closely related that they cannot be distinguished at
BIOLOGICAL FOUNDATIONS

181

the phenomenal level. The biological mechanisms involved in state anxiety

are a best clue to those in trait anxiety, because a trait is more or less the
summation of states over time (Zuckerman, 1976).
The central organizing structure for fear in the brain is the amygdala.
This structure receives input from systems at all levels, from primary sensory
signals to polymodal centers of organization in the thalamus and higher
cortical centers, and it probably assigns significance to the stimuli (LeDoux,
1987). The output of the amygdala to the central gray, hypothalamus,
trigeminal factor motor nucleus, locus coeruleus, and frontal cortex results
in the typical autonomic arousal, facial expressions, startle, freezing of behavior, and increased vigilance associated with the emotion of fear (Davis,
1986; Davis, Hitchcock, & Rosen, 1987). The noradrenergic ascending
bundle from the locus coeruleus carries the arousal to all areas of the cortex
in a positive feedback loop. Individual differences in the sensitivity or
reactivity of any parts of this system may underlie individual differences in the
proneness to experience anxiety and to develop conditioned fear responses.
This kind of analysis of fear at the biological level does not minimize
the role of life experiences and associated cognitive mechanisms, such as
pessimism, on mood. But individuals having the same kind of negative
experiences do not necessarily transform those experiences into a longlasting expectation of future harm. Some of the differences in reactions to
stressful experience may depend on the genetic or biological factors involved
in individual differences.
Risk appraisal is a central mechanism in pessimism. If an activity is
perceived as risky, this perception implies a pessimistic attitude toward
outcome of that activity. Experience in particular activities is a factor in
developing optimistic or pessimistic expectations for future outcomes. If no
harm has resulted from the experience, then risk appraisal is reduced. But
a harmful experience may increase risk appraisal and decrease the likelihood
of repeating that experience if there is any choice. Personality disposition
may predispose toward risk appraisal even for activities that were experienced. Sensation seeking is a trait that correlates highly with risk appraisal for
novel situations never before experienced; low sensation seeking individuals
tend to rate such situations as riskier than high sensation seeking individuals
(Zuckerman, 1979). Neuroticism and extraversion were not related to the
general risk appraisal tendency.
Impulsivity is another factor related to the tendency to enter into
novel but potentially risky situations. However, impulsivity is often a result
of a lack of appraisal of potential outcomes rather than of an inherently
optimistic appraisal of situations. The impulsive person sometimes acts and
thinks about the possible negative outcomes only after the act if ever. It is
possible to be a cautious sensation seeker as is the mountain climber who
plans his or her ascent very carefully, seeking to minimize all foreseeable

1a2

MARVIN ZUCKERMAN

risks. Sensation seekers have an optimistic bias in their beliefs in their own
capacities to deal with the risks involved. But it is the unforeseen risks that
are often fatal.
However, extraversion more so than sensation seeking is associated
with general reward expectancy (see Tables 8.1 and 8.2). Extraverts are not
necessarily risk takers or impulsive about activities with physical risks, but
they probably do take more social risks than do introverts. The biological
basis of extraversion is not as clear as that for neuroticism-anxiety. Extraversion, like sensation seeking, probably has a strong dopaminergic reactivity
to novel situations or people. There may also be a lack of strength in
inhibitory mechanisms. Both sensation seekers and extraverts have a tendency to become easily disinhibited in social situations. Of course, alcohol
is a pharmacological agent widely used to promote disinhibition of behavior
at social occasions. Extraverts and sensation seekers need less of it, and
sometimes act as if they had reached an optimal level of intoxication when
they have actually consumed very little alcohol. Alcohol or other drugs at
moderate doses may increase optimism and decrease pessimism, although I
have not seen any experimental studies of this effect. Gamblers who drink
too much become recklessly optimistic in their betting, ignoring the reality
of odds and bluffing their way to bankruptcy.

Mood, Arousal, and Arousability

Earlier in this book (see chap. 6, this volume) Peterson and Bossio
reviewed the findings linking optimism and pessimism to physical wellbeing and outcomes of illness. Although there is some evidence that these
attitudes show correlations with the immune system functions, these authors
conclude that the major mechanism linking optimism and health is a mundane behavioral pathway. Strangely, it is the pessimists who tend to engage
in unhealthy behavior and neglect the possible ways of coping with an
illness. It is as if they are resigned to their fate. The optimists, in contrast,
tend to be confident that they can avoid illness through proper diet and
health measures, and if they do become ill, that they can recover through
appropriate treatment.
There is another way that pessimism may have deleterious effects on
physical health. Like anxiety, pessimism is associated with negative moods,
such as anxiety and depression, and these moods are associated with chronic
arousal, that could be expressed in elevated blood pressure. One study
recorded ambulatory blood pressures over a 3-day period with periodic assessments of mood (Raikkonen, Matthews, Flory, Owens, & Gump, 1999).
Pessimists, particularly those low on optimism, had higher blood pressure
over the 3 days of the experiment than did optimists, whereas the optimists
blood pressures only increased during negative mood states. The same findBZOLOGICAL FOUNDATIONS

183

ings were obtained for anxiety, but analysis showed independent effects of
anxiety and optimism or pessimism. The results suggest a chronic elevation
of blood pressure in pessimists even though only participants in the normal
range of blood pressure (BP) were used in the study. In contrast, the optimists
showed elevations of BP only during periods of negative mood states, perhaps
an expression of attempts at coping. If chronic, nonadaptive elevations in
blood pressure are characteristic of pessimists and anxious persons, it could
predispose them to certain types of health problems, particularly those
involving the cardiovascular system.

CONCLUSIONS
An optimistic attitude may be an adaptive trait selected in evolution
for our species. But, as with personality and intellectual traits, there is wide
individual variation in both optimism and pessimism. Heritability measures
of the sources of this individual variation suggest a low (relative to personality
traits) but significant heritability for optimism and pessimism. Some of this
heritability may be due to the association of these cognitive traits with
broader personality traits: optimism with extraversion and pessimism with
neuroticism. Some of the same genes involved in the personality traits may
influence the genetic part of the tendency to optimism and pessimism.
Optimism is also influenced by shared familial factors and nonshared life
events, but pessimism seems to be primarily learned by events outside of
the shared family environment.
Optimism and pessimism play an important role in many disorders,
particularly the mood disorders, where they appear to be state dependent.
When in a depressed state, the patient tends to have a pessimistic outlook
for the future, whereas during manic states, an unrealistic optimism prevails.
Drug treatments can change these generalized expectancies, showing some
dependency of optimism and pessimism on the biological states underlying
the clinical conditions. But for less severe states of depression, cognitive
and other therapies can also change the expectancy attitudes.
Pessimism may affect health through the chronic arousal associated
with anxiety and helplessness in the pessimist. Optimism is adaptive because
it encourages attempts to cope and achieve, whereas pessimism is nonadaptive because it discourages such attempts and tends to lead to passivity in
the face of stress or challenge. Whatever is wired in to our species is
probably due to the survival and mating skills of optimists among our
early hominid ancestors. However, the pessimist also has certain values
contributing to the survival of the species, such as avoiding excessive risks
that could lead to premature extinction. Much of the biological bases of
184

MARVIN ZUCKERMAN

these traits are probably shared with that for the associated personality and
mood traits.
Most of the inferences about the biological bases of optimism and
pessimism come from their associations with better-researched personality
and mood traits. However, this kind of indirect inference is not reliable.
More research is needed on the direct associations between optimism and
pessimism and biological traits. Experimental research is needed on the
biological effects of failure and success. To what extent do these effects vary
as a function of trait optimism and pessimism? What are the effects of drugs
on these attitudes? Although psychobiological research on human attitudes
is difficult, I am optimistic about the future for such research.
If optimism is a vital factor in health, it must work through biological
mechanisms. W e must learn what these mechanisms are and how they work.

REFERENCES
Abramson, L. Y., Metalsky, G. I., & Alloy, L. B. (1989). Hopelessness depression:
A theory-based subtype of depression. Psychological Review, 96, 358-372.
Ball, S. A., & Zuckerman, M. (1990). Sensation seeking, Eysenck's personality
dimensions and reinforcement sensitivity in concept formation. Personality and
lndividual Differences, 1 1 , 343-353.
Beck, A. T. (1972). Depression: Causes and treatment. Philadelphia, PA: University
of Pennsylvania Press.
Bouchard,T. J., Jr. (1993). Genetic and environmental influences on adult personality: Evaluating the evidence. In J. Hettema & I. J. Deary (Eds.), Foundations of
personality (pp. 15-44). Dordrecht, Netherlands: Kluwer Academic Publishers.
Bouchard, T. J., Jr., Lykken, D. T., McGue, M., Segal, N. L., & Tellegen, A. (1990).
Sources of human psychological differences. Science, 250, 223-228.
Burke, M. J., & Preskorn, S. H. (1995). Short-term treatment of mood disorders
with standard anti-depressants.In F. E. Bloom & D. J. Kupfer (Eds.),Psychophrm o l o g y : The fourth generation of prog-ress (pp. 1053-1065). New York: Raven Press.
Chang, E. C., DZurilla, T. J., & Maydeu-Olivares,A. (1994). Assessing the dimensionality of optimism and pessimism using a multimeasure approach. Cognitive
Therapy and Research, 18, 143-160.
Cutler, N. R., & Post, R. M. (1982). State related dyskinesias in manic-depressive
illness. Journal of Clinical Psychopharmacology , 2 , 350-354.
Davis, M. ( 1986). Pharmacological and anatomical analysis of fear conditioning
using the fear-potentiated startle paradigm. Behavioral Neuroscience, 100, 814824.
Davis, M., Hitchcock, J. M., & Rosen, J. B. (1987). Anxiety and amygdala: Pharmacological and anatomical analysis of the fear-potentiated startle paradigm. In
BIOLOGICAL FOUNDATIONS

1 85

G. H. Bower (Ed.), The psychology of learning and motivation (Vol. 21, pp.
263-305). New York: Academic Press.
Eaves, G., & Rush, A. I. (1984). Cognitive patterns in institutionalizedand remitted
major depression. Journal of Abnormal Psychology, 93, 31-40.
Eysenck, S. B. G., Eysenck, H. J., & Barratt, P. (1985). A revised version of the
psychoticism scale. Personality and lndividuul Differences, 6, 21-29.
Gershon, E. S. (1990). Genetics. In F. K. Goodwin & K. R. Jamison (Eds.), Manic
depressive illness (pp. 369-401). New York: Oxford University Press.
Goodwin, F. K., & Jamison, K. R. (1990). Manic-depressive illness. New York: Oxford
University Press.
Gould, S. (1981). The mismeasure of man. New York: Norton Press.
Gray, J. A. (1982). The neuropsychology of anxiety: An enquiry into the functions of
the septohippocumpul system. New York: Oxford University Press.
Gray, J. A. (1987). The neuropsychology of emotion and personality. In S. M.
Stahl, S. D. Iverson, & E. C. Goodman (Eds.), Cognitive neurochemistry (pp.
17 1- 190). Oxford: Oxford University Press.
Kendler, K. S., Neale, M. C., Kessler, R. C., Heath, A. C., & Eaves, L. J. (1992a).
The genetic epidemiology of phobias in women. Archives of General Psychiatry,
49, 273-281.
Kendler, K. S., Neale, M. C., Kessler, R. C., Heath, A. C., & Eaves, L. J. (199213).
Major depression and generalized anxiety disorder: Same genes (partly) different environments? Archives of General Psychiatry, 49, 7 16-722.
Kendler, K. S., Pederson, N., Johnson, L., Neale, M. C., & Mathe, A. A. (1993). A
pilot Swedish twin study of affective illness, including hospital- and population
ascertained subsamples. Archives of General Psychiatry, 50, 699-706.
LeDoux, J. E. (1987). Emotion. In F. Plum (Ed.), Handbook ofphysiology: The nervous
system (Vol. 5, pp. 419-459). Bethesda, MD: American Physiological Society.
Loehlin, J. C. (1992). Genes and environment in personality development. Newbury
Park, CA: Sage Publications.
Lubin, B., & Zuckerman, M. (1999). Manual for the Multiple Affect Adjective Check
List-Revised (MAACL-R) . San Diego, CA: Educational and Industrial Testing Service.
Maes, M., & Meltzer, H. Y. (1995). The serotonin hypothesis of major depression.
In C. E. Bloom & D. J. Kupfer (Eds.), Psychopharmacology:The fourth generation
of progress (pp. 933-944). New York: Raven Press.
McNally, R. J. (1990). Psychological approaches to panic disorder:A review. Psychological Bulletin, 108, 403-419.
Nurnberger, J. I., Jr., & Gershon, E. S. (1992). Genetics. In E. S. Paykel (Ed.),
Handbook of affective disorders (pp. 131-148). New York: Guilford.
Pederson, N. L., Plomin, R., McClearn, G. E., & Friberg, L. (1988). Neuroticism,
extraversion and related traits in adult twins reared apart and reared together.
Journul of Personality and Social Psychology, 55, 950-957.

186

MARVIN ZUCKERMAN

Peterson, C., Semmel, A., von Baeyer, C., Abramson, L. Y., Metalsky, G. I., &
Seligman, M. E. P. ( 1982). Attributional Style Questionnaire. Cognitive Therapy and Research, 6, 287-299.
Plomin, R., DeFries, J. C., McClearn, G. E., &Rutter, M. (1997). Behvioralgenetics.
New York: W. H. Freeman.
Plomin, R., Scheier, M. F., Bergeman, C. S., Pederson, N. L., Nesselroade, J. R.,
& McClearn, G. E. (1992). Optimism, pessimism and mental health: A twin/
adoption analysis. Personality and Individual Differences, 13, 921-930.
Price, L. H., Goddard, A. W., Barr, L. C., & Goodman, W. K. (1995). Pharmacological challenges in anxiety disorders. In F. E. Bloom & D. J. Kupfer (Eds.),
Psychopharmacology: The fourth generation of progress (pp. 1311-1324). New
York: Free Press.
Raikkonen, K., Matthews, K. A., Flory, J. D., Owens, J. F., & Gump, B. B. (1999).
Effects of optimism, pessimism, and trait anxiety on ambulatory blood pressure
and mood during everyday life. Journal of Personality and Social Psychology,
76, 104-128.
Redmond, D. E., Jr. (1987). Studies of locus coeruleus in monkeys and hypotheses
for neuropsychopharmacology.In H. Y. Meltzer (Ed.), Psychopharmacology: The
third generation of progress (pp. 967-975). New York: Raven Press.
Schatzberg, A. F., & Schildkraut, J. J. (1995). Recent studies on norepinephrine
systems in mood disorders. In F. E. Bloom & D. J. Kupfer (Eds.), Psychophrmacology: The fourth generation ofprogress (pp. 91 1-920). New York: Raven Press.
Scheier, M. F., & Carver, C. S. (1985). Optimism, coping, and health assessment
and implications of generalized outcome expectancies. Health Psychology, 5 ,
219-247.
Schulman, P., Keith, D., & Seligman, M. E. P. (1993). Is optimism heritable? A
study of twins. Behvior Research and Therapy, 31, 569-574.
Seligman, M. E. P. (1975). Helpkssness: O n depression, development and death. New
York: Freeman.
Seligman, M. E. P. (1991). Learned optimism. New York: Knopf.
Seligman, M. E. P., Peterson, C., Kaslow, N. J., Tannenbaum, R. L., Alloy, L. B.,
& Abramson, L. Y. (1984). Attribution style and depressive symptoms among
children. Journal of Abnormal Psychology, 93, 235-238.
Shader, R. I., & Greenblatt, D. J. (1995). The pharmacotherapy of acute anxiety:
A mini-update. In F. E. Bloom & D. J. Kupfer (Eds.), Psychopharmacology: The
fourth generation of progress (pp. 1341-1348). New York: Free Press.
Tellegen, A., Lykken, D. T., Bouchard, T. J., Wilcox, K., Segal, N., & Rich, A.
(1988). Personality similarity in twins reared together and apart. Journal of
Personality and Social Psychology, 54, 1031-1039.
Tiger, L. (1979). Optimism: The biology of hope. New York: Simon & Schuster.
Torrubia, R., Avila, C., Moltci, I.,
& Grande, I. (1995). Testing for stress and
happiness: The role of the behavioral inhibition system. In C. D. Spielberger,
I. G. Sarason, J. M. T. Brebner, E. Greenglass, P. Laungani, & A. M. ORoark
BIOLOGICAL FOUNDATIONS

187

(Eds.), Stress and emotion: Anxiety, anger, and curiosity (Vol. 15, pp. 191-211).
Washington, DC:Taylor and Francis.
Zuckerman, M. (1976). General and situation-specific traits and states: New approaches to assessment of anxiety and other constructs. In M. Zuckerman &
C. D. Spielberger (Eds.), Emotions and anxiety: New concepts, methods, and
applications (pp. 133-174). Hillsdale, NJ: Erlbaum.
Zuckerman, M. (1979). Sensation seeking and risk-taking. In C. E. Izard (Ed.),
Emotions in personality and psychopathology (pp. 163-197). New York: Plenum.
Zuckerman, M. (1984). Sensation seeking: A comparative approach to a human
trait. Behavioral and Brain Sciences, 7, 413-471.
Zuckerman, M. ( 1985). Sensation seeking, mania, and monoamines. Neuropsychobiology, 13, 121-128.
Zuckerman, M. (1991). Psychobiology of personality. Cambridge, UK: Cambridge
University Press.
Zuckerman, M. (1993). Personality from the top (traits) to bottom (genetics) with
stops at each level between. In J. Hettema & I. J. Deary (Eds.), Foundations
of personality (pp. xx-xx). Dordrecht, The Netherlands: Kluwer.
Zuckerman, M. ( 1994). Behavioral expressions and biosociul bases of sensation seeking.
New York: Cambridge University Press.
Zuckerman, M. (1995). Good and bad humors: Biochemical bases of personality
and its disorders. Psychological Science, 6, 325-332.
Zuckerman, M. (1999). Vulnerability to psychopathology: A biosocial model. Washington, DC: American Psychological Association.
Zuckerman, M., Buchsbaum, M. S., & Murphy, D. L. (1980). Sensation seeking
and its biological correlates. Psychological Bulletin, 88, 187-214.
Zuckerman, M., Joireman, J., Kraft, M., & Kuhlman, D. M. (1999). Where do
motivational and emotional traits fit within three factor models of personality?
Personality and Individual Differences, 26, 487-504.
Zuckerman, M., Kuhlman, D. M., & Camac, C. (1988). What lies beyond E and
N ? Factor analyses of scales believed to measure basic dimensions of personality.
Journal of Personality and Social Psychology, 54, 96-107.
Zuckerman, M., Kuhlman, D. M., Joireman, J., Teta, P., & Kraft, M. (1993). A
comparison of three structural models for personality: The big three, the big
five, and the alternative five. Journal of Personality and Social Psychology, 65,
757-768.
Zuckerman, M., Kuhlman, D., Thornquist, M., & Kiers, H. (1991). Five (or three)
robust questionnaire scale factors of personality without culture. Personality
and Individual Differences, 12, 929-94 1.
Zuckerman, M., & Lubin, B. (1985). Manual for the Multiple Affect Adjective Check
List-Revised (MAACL-R) . San Diego, CA: Educational and Industrial Testing Service.

188

MARVIN ZUCKERMAN

OPTIMISM, PESSIMISIVI,AND
PSYCHOLOGICAL WELLcBEING
MICHAEL F. SCHEIER, CHARLES S. CARVER,
AND MICHAEL W. BRIDGES

For the past several years, we and a number of other psychologists who
are interested in issues concerning stress, coping, and health have been
exploring the potential benefits of an optimistic orientation to life. The
primary purpose of this chapter is to present an overview of the research
that has accumulated on this issue. By design, our review is limited to effects
on psychological well-being. Evidence linking optimism to physical wellbeing is reviewed elsewhere in this book.
This chapter is organized as follows. We begin by commenting on a
distinction between two ways of assessing optimism and pessimism. Then
we review some of the empirical evidence linking positive thinking to wellbeing. Our review focuses primarily on studies that are prospective in design,
Preparation of this chapter was facilitated by support from the National Cancer Institute (CA64710,
CA62711, and CA78995) and the National Heart, Lung, and Blood Institute (HL65111,
PROHL65112, HL 65112, and PROHL65111).

189

because of their superior methodological rigor. Having reviewed the literature, we go on to consider why optimism might confer benefits, arguing that
the benefits are due, in part, to the way in which optimists and pessimists
cope with problems. We close by asking whether the effects of optimism
are always good and whether the effects of pessimism are always bad.

VARIATIONS IN CONCEPTIONS AND ASSESSMENT

OF OPTIMISM
Expectancies are pivotal in contemporary theories of optimism, but as
suggested by the opening chapters of this volume, there are at least a couple
of ways to think about expectancies and how to measure them. One approach
measures expectancies directly, asking people to indicate the extent to which
they believe that their future outcomes will be good or bad. This is the
approach weve taken in our work, as reflected in peoples responses to the
Life Orientation Test (LOT; Scheier & Carver, 1985) or to the briefer Life
Orientation Test-Revised (LOT-R;Scheier, Carver, & Bridges, 1994),which
we currently use. Expectancies that are generalized-expectancies that pertain more or less to the individuals entire life space-are what we mean
when we use the terms optimism and pessimism (cf. Armor & Taylor, 1998).
The other approach to assessing optimism is more indirect. It assesses
expectancies by examining attributional style, the characteristic manner in
which a person explains prior events (Peterson & Seligman, 1984; Seligman,
1991; see also chap. 3, this volume). Attributions for bad outcomes that
are explained in terms of causes that persist into the future, influence a
broad range of events, and involve an aspect of the self are seen as pessimistic
in nature. They are pessimistic because they carry the implication that
negative outcomes will continue to occur in the future. The opposite attributional style, explaining negative events in terms of causes that are more
time limited, narrow in their effects, and external to the self, is thought to
reflect a more optimistic orientation. Although in principle attributional
style can be assessed with respect to both negative and positive prior events,
in practice, assessment is usually limited to negative outcomes.
These two approaches to conceptualizing and measuring optimism
have led to their own research literatures, each of which sheds light on the
nature and function of optimism and pessimism (see also Snyder et al.s
discussion of hope, another variant in this general line of theories, in chap.
5, this volume). In what follows, we focus mostly on optimism as operationalized by Scheier and Carver (Scheier & Carver, 1985; Scheier & Carver,
1992; Scheier et al., 1994)-that is, in terms of self-reported generalized
expectancies.
190

SCHEIER ET AL.

EFFECTS OF OPTIMISM AND PESSIMISM

ON PSYCHOLOGICAL WELL-BEING
When people confront adversity or difficulty in their lives, they experience a variety of emotions, ranging from excitement and eagerness, to anger,
anxiety, and depression. The balance among these feelings appears to relate
to peoples degree of optimism or pessimism. Optimists are people who
expect to have positive outcomes, even when things are difficult. This
confidence should yield a mix of feelings that is relatively positive. Pessimists expect negative outcomes. This doubt should yield a greater tendency
toward negative feelings-anxiety, guilt, anger, sadness, or despair (Carver
& Scheier, 1998; Scheier & Carver, 1992).
Over the past decade or so, researchers have examined relationships
between optimism and distress in groups of people facing difficulty or adversity. The range of stressors involved in the research is very broad. For
example, researchers have examined the experiences of students entering
college (Aspinwall & Taylor, 1992); employees of businesses (Long, 1993);
and survivors of missile attacks (Zeidner & Hammer, 1992). Researchers
have measured the responses of people caring for medical patients of one
type or another (Given et al., 1993; Hooker, Monahan, Shifren, & Hutchinson, 1992; Shifren & Hooker, 1995).Researchers have examined experiences
of people dealing with medical procedures, such as childbirth (Carver &
Gaines, 1987; Park, Moore, Turner, & Adler, 1997); abortion (Cozzarelli,
1993); coronary artery bypass surgery (Fitzgerald, Tennen, Affleck, &
Pransky, 1993; Scheier et al., 1989); attempts at in vitro fertilization (Litt,
Tennen, Affleck, & Klock, 1992); and bone marrow transplantation
(Curbow, Somerfield, Baker, Wingard, & Legro, 1993). Yet other researchers
have looked at how people deal with a diagnosis of cancer (Carver et al.,
1993; Friedman et al., 1992); and the progression of AIDS (Taylor et
al., 1992).
Some of these studies are more limited than are others. In some projects,
the researchers assessed optimism and distress at the same point in time,
usually after the negative event occurred. Whats known from these crosssectional studies is that people considered pessimists tended to report more
distress than those considered optimists. What is not known is whether the
pessimists may have been more distressed even before the event occurred.
Additionally, cross-sectional studies make it difficult to tease apart causation.
The presumption in the studies is that the casual arrow runs from optimism
to distress, but the opposite possibility cannot be ruled out.
Given these inherent limitations of cross-sectional studies, it is far
more useful to conduct prospective studies-to see how levels of distress
shift and change over time and circumstance. Of course, its sometimes
difficult to recruit participants before the occurrence of the event in which
PSYCHOLOGICAL WELL-BEING

I91

youre interested (e.g., its hard to know when a hurricane will occur, so
its hard to recruit people 2 weeks beforehand). But even in cases where
participants were not recruited before the event, its useful to examine the
process of adaptation to the event across an extended period. Some of the
studies that make up this literature on stress and well-being did assess
participants at multiple time points, and we will focus on the studies here
(for a representative sampling of cross-sectional research, see Table 9.1).
The roots of current research on optimism can be traced back to earlier
work within the field of health psychology. For this reason, it is perhaps
not surprising that much of what we know about the effects of optimism
on psychological well-being comes from research conducted in healthrelated contexts.

Pregnancy and Childbirth

One early study of the effect of optimism on emotional well-being
(Carver 6r Gaines, 1987) examined the development of depressed feelings
after childbirth. Women in this study completed the LOT and a depression
scale in the last third of their pregnancy. They then completed the depression
measure again 3 weeks after their babies were born. Optimism was related
to lower depression symptoms at the initial assessment. More important,
optimism predicted lower levels of depressive symptoms postpartum, even
when researchers controlled for the initial levels. Thus optimism seemed
to confer a resistance to the development of depressive symptoms after
having a baby.
A similar point is made by two more recent studies. As did Carver
and Gaines, Fontaine and Jones (1997) administered the LOT and a measure
of depression to women before and after childbirth. Optimism was once
again associated with fewer depressive symptoms, both during pregnancy
and 2 weeks postpartum. Park et al. (1997) also examined the effects of
optimism on psychological adjustment during pregnancy. Women completed
the LOT during their first prenatal clinic visit. During the last trimester of
their pregnancy, the womens anxiety was assessed, as well as the extent to
which they were maintaining positive states of mind. Optimism was a
significant predictor of both. The predictive association of optimism with
anxiety was negative, whereas the association with positive states of mind
was positive.
Another medical situation that has been studied with respect to optimism effects is infertility-a problem that causes a good deal of unhappiness
to a good many people (Stanton & Dunkel-Schetter, 1991). A procedure
called in vitro fertilization is one way to circumvent fertility problems, but
it doesnt always work. In this study, researchers focused on people whose
attempts were unsuccessful (Litt et al., 1992). Approximately 8 weeks before
192

SCHElER ET

AL.

College students (N= 726)

Chang, 1998a

Note. MI = myocardial infarction.

Sweetman, Munz, & Wheeler, 1993

Zeidner & Hammer, 1992

Long, Kahn, & Schutz, 1992

Long & Sangster, 1993
Marshall & Lang, 1990
Nelson, Karr, & Coleman, 1995
Neverlien & Backer, 1991
OBrien, VanEgeren, & Mumby, 1995
Schuller, 1995
Sumi, Horie, & Hayakawa, 1997

Khoo & Bishop, 1997

Lauver & Tak, 1995

Hooker et al., 1992

Desharnais, Godin, Jobin, Valois, & Ross 1990

Dunn, 1996
Fontaine & Seal, 1997
Friedman et al., 1992
Given et al., 1993
Guarnera & Williams, 1987
Alzheimers disease spouse caregivers
(N= 51)
Adult Singaporeans (N= 126)
Patients experiencing breast cancer
symptoms (N= 135)
Female business managers (N= 192)
Rheumatoid arthritis patients (N= 107)
Professional women (N= 192)
College students (N= 241)
10-1 2-year-old dental patients (N= 163)
Undergraduates (N= 131)
College students (N= 75)
Female Type A Japanese college
students (N= 144)
Male and female attorneys (N = 80)
Victims of missile attacks ( N = 261)

Greater general well-being

Lower anxiety, less depression, less fear

Lower anxiety/higher job satisfaction

Better adjustment
Less depression
Fewer daily hassles
Less clinical dental anxiety
Lower stress levels
Lower trait anxiety
Lower anxiety

College students (N= 388)

Chang, 1998b

Bone marrow transplant patients

( N =135)
Post-MI patients (N= 158)
Amputation patients (N= 138)
Female adults (N= 101)
Mixed cancer patients ( N = 954)
Cancer patient caregivers (N= 196)
Elderly adults (N= 92)

Fewer perceived hassles/more positive

psychological adjustment
Less stress, fewer depressive symptoms/
higher satisfaction with life
Less depression/higher satisfaction with
life
Less negative mood/higher satisfaction
with life
Less perceived susceptibility/less fear
Less depression/higher self-esteem
Fewer mood-related symptoms
Lower levels of daily stress
Lower levels of depression
Less external locus of control/more internal
locus of control
Less stress, less depressiodgreater
positive well-being
Less stresdgreater positive well-being
Less anxiety in seeking care

College students (N= 408)

Ausbrooks. Thomas. & Williams. 1995

Barron, Foxall, von Dollen, Shull, & Jones,
1992
Blankstein, Flett, & Koledin, 1991

Curbow et al., 1993

Association With LOT scores

Less chronic anger/less anger suppression
Less loneliness
~~~

Sample
Colleae students UV = 720)
Eldefy low-visionwomen (N= 56)

Author(s)

Table 9.1.
Cross-Sectional Studies Documenting Benefits of Optimism on Psychological Well-Being

the attempt, the researchers measured optimism, specific expectancies for

fertilization success, coping strategies, distress levels, and the impact of the
infertility on participants lives. Later, 2 weeks after notification of a negative
pregnancy test, distress was assessed. Neither demographics, obstetric history,
marital adjustment, nor the rated effect of infertility on participants lives
predicted this Time 2 distress-but pessimism did. When Time 1 distress
was controlled, pessimism was the strongest predictor of Time 2 distress.
Yet another study examined the influence of optimism on adjustment
to abortion (Cozzarelli, 1993). Women completed measures of optimism,
self-esteem, self-mastery, self-efficacy, and depression 1 hour prior to an
abortion. Their depression and psychological adjustment were assessed 30
minutes after the abortion and again 3 weeks later. Optimists had less
preabortion depression, better postabortion adjustment, and better 3-week
adjustment than did pessimists. The authors of the study concluded that
optimism relates to psychological adjustment both directly and also indirectly
through a sense of personal efficacy.
The importance of optimism in adjusting to abortion has recently been
reaffirmed in research by Major, Richards, Cooper, Cozzarelli, & Zubek
(1998). Major et al. were interested in examining personal resources that
might protect women from the distress due to an abortion. Optimism was
identified in their research as a key component in a style of personal resilience
that facilitated postabortion adjustment.
Coronary Artery Bypass Graft Surgery
Other research investigated the reactions of men who were undergoing
and recovering from coronary artery bypass surgery (Scheier et al., 1989).
Patients completed questionnaires the day before surgery, 6-8 days after
surgery, and 6 months postsurgery. Prior to surgery, optimists reported lower
levels of hostility and depression than did pessimists. A week after surgery,
optimists reported feeling more relief and happiness, greater satisfaction
with their medical care, and greater satisfaction with the emotional support
they had received from friends. Six months after surgery, optimists reported
higher quality of life than did pessimists. In a follow-up of the same patients
5 years after surgery (cited in Scheier & Carver, 1992), optimists continued
to experience greater subjective well-being and general quality of life compared with pessimists. All these differences remained significant when medical factors were statistically controlled.
Another study on optimism and quality of life after coronary artery
bypass surgery (Fitzgerald et al., 1993) assessed participants 1 month prior
to surgery and 8 months after surgery. Analysis revealed that optimism was
negatively related to presurgical distress. Furthermore, when the researchers

194

SCHEIER ET AL.

controlled for presurgical life satisfaction, optimism was positively related

to postsurgical life satisfaction. Further analysis revealed that the general
sense of optimism appeared to operate on feelings of life satisfaction through
a more focused sense of confidence about the surgery. That is, the general
sense of optimism about life seems to have been funneled into a specific
kind of optimism regarding the surgery, and from there to satisfaction with
life. All of the effects involving optimism reported by Fitzgerald et al. (1993)
were found to be independent of disease severity.
Similar salutary effects of optimism have been observed in a group of
women undergoing coronary artery bypass surgery (King, Rowe, Kimble, &
Zerwic, 1998). In this study, data were collected at 1 week, and 1, 6, and
12 months after surgery. At all points of time, optimism demonstrated
significant adaptive relationships with positive mood, negative mood, and
satisfaction with life. More importantly, optimism assessed at 1 week was
associated with more positive mood and less negative mood at 1 month
independent of initial levels of positive and negative mood.

Cancer
Optimism has also been studied in the context of adjusting to the
diagnosis and treatment for cancer. One study examined the effect of optimism on psychological adaptation to treatment for early-stage breast cancer
(Carver et al., 1993).Diagnosis and treatment for breast cancer is a traumatic
experience, and cancer is a life-threatening disease. However, prognosis for
persons with early-stage cancer is relatively good. This experience thus
provides sufficient ambiguity about what will happen in the future to permit
individual differences in optimism to be readily expressed. Patients in this
study were interviewed six times: at the time of diagnosis, the day before
surgery, 7-10 days after surgery, and 3, 6, and 12 months later. Optimism
was assessed (using the LOT) at the time of diagnosis and was used to
predict distress levels at the subsequent time points.
Optimism inversely predicted distress over time, above and beyond
the effect of relevant medical variables and beyond the effects of earlier
distress. That is, the prediction of distress at 3, 6, and 12 months after
surgery was significant even when the immediately prior level of distress
was controlled. Thus, optimism predicted not just lower initial distress, but
also resilience to distress during the year following surgery.
In another study involving cancer patients, Johnson (1996) demonstrated the beneficial effects of optimism among a group of men receiving
radiation therapy for prostate cancer. Optimism was assessed using the LOT
prior to the first radiation treatment. Mood was assessed throughout the
treatment period and 2 weeks, 1 month, and 3 months following the end

PSYCHOLOGICAL WELL-BEING

195

of treatment. Optimism was a strong predictor of patients emotional responses both during and after treatment, with less optimistic patients being
particularly vulnerable to negative moods.
Also relevant here is a study by Christman (1990), which examined
adjustment among a group of patients undergoing radiotherapy for a variety
of cancers. Patients were enrolled into the project at one of three points
in time, on the first day of treatment, on the 15th day of treatment, and
on the last day of treatment. Multiple assessments were made on those
patients who were enrolled prior to the conclusion of their treatment.
Optimism was negatively related to adjustment problems at all three assessment points, as well as to illness uncertainty at the first and second assessment
points. These associations involving optimism could not be accounted for
in terms of differences between optimists and pessimists in their preferences
for health-related locus of control.
Other Health-Related Contexts

Still further evidence of the association between optimism and psychological well-being was found by Taylor et al. (1992) among a cohort of
gay and bisexual men at risk of developing Acquired Immunodeficiency
Syndrome (AIDS). Among those men who were seropositive (HIV+) as
well as those who were seronegative (HIV-) greater optimism was associated
with lower levels of subsequent psychological distress as measured by a
composite index of negative affect and by specific worries and concerns
about AIDS.
In a somewhat different medical population, Chamberlain, Petrie, and
Azaria (1992) demonstrated the beneficial effects of optimism among a
group of patients undergoing joint replacement surgery and rehabilitation.
Data were collected before surgery, the third day following surgery, and
approximately 6 weeks following surgery. Optimism assessed presurgically
was positively associated with follow-up levels of life satisfaction and positive
well-being, and negatively associated with psychological distress and pain.
Not only does optimism have a positive effect on the psychological
well-being of people dealing with medical conditions, but it also influences
the psychological well-being of people who are acting as caregivers to patients. In one project supporting this conclusion, researchers studied a group
of cancer patients and their caregivers and found that caregivers optimism
related to a number of caregiver well-being variables (Given et al., 1993).
Optimism was related to lower symptoms of depression, less impact of caregiving on physical health, and less impact on caregivers daily schedules.
Caregiver optimism thus predicted caregiver reactions to the burdens of
caring for a family member with cancer, and did so independent of patient variables.
196

SCHEIER ET AL.

Similar results have been found in research on caregivers of Alzheimers

patients (Hooker et al., 1992; Shifren & Hooker, 1995) and caregivers of
patients suffering from stroke (Tompkins, Schulz, & Rau, 1988). In each
case, optimism was associated with lower levels of depression and higher
levels of psychological well-being.

Non Health-Related Contexts

Although much of the evidence for the relationship between optimism
and psychological well-being comes from people facing health threats and
people caring for seriously ill relatives, other researchers have looked at
groups of people undergoing events that are difficult but far less extreme.
For example, the start of a persons college years is a difficult and stressful
time. Researchers have examined students making the adjustment to their
first semester of college (Aspinwall & Taylor, 1992). Optimism, self-esteem,
and a number of other variables were assessed when the students first arrived
on campus. Measures of psychological and physical well-being were obtained
at the end of the semester. Higher levels of optimism upon entering college
predicted lower levels of psychological distress at the end of the semester.
The relationship was independent of effects due to self-esteem, locus of
control, desire for control, and baseline mood.
Another study of adjustment to college life (cited in Scheier & Carver,
1992) included some additional variables, such as perceived stress and loneliness. This project assessed most of the adjustment variables both at the
beginning of the semester and at the end of the semester (rather than just
at the end), and looked at changes in these measures over time. The data
revealed that optimists became less stressed, less depressed, less lonely, and
more socially supported over the semester, compared with their pessimistic counterparts.
In a similar vein, Stewart et al. (1997) examined adjustment among
a group of first-year medical students. Students were surveyed prior to the
beginning of their medical training. Two months later, students who initially
reported high baseline levels of anxiety and low levels of optimism were
more likely to develop symptoms of depression and anxiety.
In a study involving first-year law students, Segerstrom, Taylor, Kemeny, and Fahey (1998) assessed dispositional optimism and mood disturbance prior to the beginning of classes. The same measures were readministered at midsemester, approximately 8 weeks later. Baseline optimism was
associated with fewer mood disturbances at follow-up. Relevant to our discussion here, this relation remained significant even after controlling for initial
level of mood disturbance.
Among a community population of middle-age adults, Raikkonen,
Matthews, Flory, Owens, and Gump (1999) assessed ambulatory blood presPSYCHOLOGICAL WELL-BEING

197

sure for 3 days and asked participants to keep a concurrent diary of daily
moods and daily stressful events. In examining the average rating of mood
across the three assessment days, the authors found that optimism (assessed
before monitoring began) was significantly associated with lower negative
mood and higher positive mood. Kaikkonen et al. also found that optimists
rated the stressfulness of the days most negative event as less stressful than
did pessimists.
In another study involving a middle-aged sample, Bromberger and
Matthews ( 1996) focused on women approaching menopause. Optimism,
depressive symptoms, and life stress were assessed at baseline and again 3
years later. Women with a more pessimistic orientation at baseline were at
higher risk for symptoms of depression at follow-up. This effect was particularly pronounced among women who reported experiencing an ongoing
stressor at the time at which pessimism was assessed. All of these effects
remained significant after baseline depression was statistically controlled.
Finally, Cohen (1990)examined the effects of optimism on psychological well-being in a study of community nurse executives. Participants initially
completed a survey that included the LOT. Later, these participants were
scheduled for an interview and were asked to provide information regarding
psychological symptoms. Optimism proved to be a significant predictor of
psychological well-being, and did so independent of the number of years
participants spent in nursing.

OPTIMISM, PESSIMISM, AND COPING

Differences in Coping Responses

As the evidence reviewed in the previous section makes clear, optimists

experience less distress than do pessimists when dealing with difficulties in
their lives. Is this just because optimists are more cheerful than pessimists?
Apparently not, because the differences often remain even when statistical
controls are incorporated for previous levels of distress. There must be other
explanations. Do optimists do anything in particular to cope that helps
them adapt better than pessimists? Many researchers are now investigating
this possibility as a potential mechanism through which optimism confers
psychological benefits. In this section we consider the strategies that optimists and pessimists tend to use, as well as the broader meaning of these
strategies.
In many respects, this discussion is a more detailed depiction of the
broad behavioral tendencies we discussed in chapter 2 of this volume, where
we described expectancy-value models of motivation. That is, people who
are confident about the future exert continuing effort, even when dealing
198

SCHEIER ET AL.

with serious adversity. People who are doubtful about the future are more
likely to try to push the adversity away as though they can somehow escape
its existence by wishful thinking. They are more likely to do things that
provide temporary distractions but dont help solve the problem. Sometimes
they even give up trying. Both the effort and the removal of effort can be
expressed in a variety of ways, and those expressions-coping reactions and
coping strategies-are the focus of this section.
Differences in coping methods used by optimists and pessimists have
been found in a number of studies. In one early project, Scheier, Weintraub,
and Carver (1986) asked undergraduates to recall the most stressful event
that had happened to them during the previous month and complete a
checklist of coping responses with respect to that event. Optimism related
positively to problem-focused coping, especially when the stressful situation
was perceived to be controllable. Optimism also related to the use of positive
reframing and (when the situation was perceived to be uncontrollable) with
the tendency to accept the reality of the situation. In contrast, optimism
related negatively to the use of denial and the attempt to distance oneself
from the problem.
These findings provided the first indication that optimists dont use
just problem-centered coping. They also use a variety of emotion-focused
coping techniques, including striving to accept the reality of difficult situations and putting the situations in the best possible light. These findings
hint that optimists may enjoy a coping advantage over pessimists even in
situations that cannot be changed.
Other researchers have studied differences in dispositional coping styles
among optimists and pessimists (Billingsley, Waehler, & Hardin, 1993;
Carver, Scheier, & Weintraub, 1989; Fontaine, Manstead, &Wagner, 1993).
As with situational coping, optimists reported a dispositional tendency to
rely on active, problem-focused coping, and they reported being more planful when confronting stressful events. Pessimism was associated with the
tendency to disengage from the goals with which the stressor is interfering.
Although optimists reported a tendency to accept the reality of stressful
events, they also reported trying to see the best in bad situations and to
learn something from them. In contrast, pessimists reported tendencies
toward overt denial and substance abuse, strategies that lessen their awareness of the problem. Thus, in general terms, optimists appear to be active
copers and pessimists appear to be avoidant copers.
Other projects studied relationships between optimism and coping
strategies in specific contexts. For example, Strutton and Lumpkin (1992)
studied coping in the work environment. They found that optimists used
problem-focused coping (such as directed problem solving) more than pessimists. Pessimism was related to the use of avoidant coping (self-indulgent
escapism including sleeping, eating, and drinking). Another study focusing
PSYCHOLOGICAL WELL-BEING

199

on executive women (Fry, 1995) found that optimists reacted to daily hassles
differently than did pessimists. Optimistic women expected gain or growth
from stressful events and reported coping indicative of acceptance, expressiveness, and tension reduction. They also reported using their social support
in stressful circumstances rather than withdrawing, distancing, or engaging
in self-blame.
Finally, the study of AIDS patients described earlier (Taylor et al.,
1992) provides information regarding coping. In general, optimism was
associated with active coping strategies. Optimism predicted positive attitudes and tendencies to plan for recovery, seek information, and reframe
bad situations so as to see their most positive aspects. Optimists made less
use of fatalism, self-blame, and escapism, and they didnt focus on the
negative aspects of the situation or try to suppress thoughts about their
symptoms. Optimists also appeared to accept situations that they could not
change, rather than trying to escape those situations.
It is important to realize that the various associations that have emerged
between optimism and coping do not seem to be simply due to a difference
in the manner in which optimists and pessimists appraise events. Chang
(1998a), for example, examined the impact of optimism and appraisals on
coping with an upcoming course exam. Optimists and pessimists did not
differ in their primary appraisals of the event, but they did differ in terms
of their secondary appraisals. That is, optimists perceived the exam to be
more controllable and their efforts to cope with it more effective than did
pessimists. On the other hand, significant differences still emerged between
optimists and pessimists on several varieties of coping even after the effect
of secondary appraisals on coping was statistically controlled.
Promotion of Well-Being
In describing the manner in which optimists and pessimists cope with
adversity, we should note several other studies. These studies do not deal
with coping per se, but they make points that are closely related to the
points we have made regarding coping. Many of the studies involve reactions
to particular health threats or illness episodes. As a group, these studies
demonstrate that optimists are more likely than pessimists to face health
threats head-on, and do whatever they can to improve the situation they
are confronting. In this respect, the behavior seems to reflect a type of
problem-focused coping, an attempt to engage in proactive processes that
promote good health and well-being (Aspinwall, 1997; Aspinwall & Tay-

lor, 1997).
In one study, researchers looking at the possibility of individual differ-

ences in health promotion followed a group of heart patients who were

participating in a cardiac rehabilitation program (Shepperd, Maroto, 6r
200

SCHEIER ET AL.

Pbert, 1996). Optimism among participants was related to greater success

in lowering levels of saturated fat, body fat, and global coronary risk. Optimism was also related to increases in level of exercise across the rehabilitation
period. In another study (cited in Scheier & Carver, 1992), researchers
investigated the lifestyles of coronary artery bypass patients 5 years after
their surgery. This study found that optimists were more likely than pessimists
to be taking vitamins, to be eating low-fat foods, and to be enrolled in a
cardiac rehabilitation program.
Heart disease isnt the only aspect of health-related behavior that has
been related to optimism. Another obvious health risk related to peoples
behavior is HIV infection. By avoiding certain sexual practices (e.g., sex
with unknown partners), people reduce their risk of infection. One study
of HIV-negative gay men revealed that optimists reported having fewer
anonymous sexual partners than did pessimists (Taylor et al., 1992). In
another study, the intentions of a group of sexually active inner-city minority
adolescents to engage in unsafe sex were examined (Carvajal, Garner, &
Evans, 1998). Optimists reported stronger intentions of avoiding unsafe sex.
Taken together, these findings suggest that optimists were making efforts
or intending to make efforts to reduce their risk, thereby safeguarding
their health.
One final study is also relevant here. Friedman, Weinberg, Webb,
Cooper, & Bruce (1995) examined health behavior and health behavior
intentions among a group of hospital employees at risk for skin cancer.
Optimists were significantly more likely than pessimists to report intentions
to engage in skin cancer-relevant health prevention behaviors (e.g., regular
sunscreen use). In addition, among those participants identified at screening
with suspicious lesions, those who were more optimistic were more likely
to comply with recommended follow-up care.
Other studies have examined the health-related habits reported by
groups of people with no particular salient health concerns. At least two
such projects found associations such that optimists reported more healthpromoting behaviors than did pessimists (Robbins, Spence, & Clark, 1991;
Steptoe et al., 1994). In sum, the various studies described here suggest
that optimism is associated with behaviors aimed at promoting health and
reducing health risk.
Consistent with this pattern is research showing conclusively that
optimists are not simply people who stick their heads in the sand and ignore
threats to their well-being. Rather, they display a pattern of attending
selectively to risks-risks that are both applicable to them and related to
serious health problems (Aspinwall & Brunhart, 1996). If the potential
health problem is minor, or if it is unlikely to bear on them personally,
optimists do not show this elevated vigilance. Only when the threat matters
does it emerge. This pattern of selective vigilance fits the idea that optimists
PSYCHOLOGICAL WELL-BEING

20 1

scan their surroundings for threats to their well-being, but save their active
coping efforts for threats that are truly meaningful.

Coping Differences and the Link to Emotional Distress

The foregoing studies help to establish that optimists cope differently
than do pessimists. However, these studies do not show that such differences
in coping tendencies are responsible for those differences we see in psychological well-being. Fortunately, a number of the studies reviewed earlier in the
section on optimism and distress contained measures of coping tendencies
in addition to measures of well-being. This allowed those researchers to
examine whether the differences they observed in well-being were mediated
by differences in coping. We report on the results of those studies in the
following paragraphs. Several other studies are relevant to the discussion
as well.
One of the studies reviewed earlier assessed the use of attentionalcognitive strategies as ways of dealing with the experience of coronary artery
bypass graft surgery (Scheier et al., 1989). Before surgery, optimists were
more likely than pessimists to report that they were making plans for their
future and setting goals for their recovery. Optimists, as compared to pessimists, also tended to report being less focused on the negative aspects of
their experience-their distress emotions and physical symptoms. Once the
surgery was past, optimists were more likely than pessimists to report seeking
out and requesting information about what the physician would be requiring
of them in the months ahead. Optimists were also less likely to report
trying to suppress thoughts about their physical symptoms. Results from
path analyses suggested that the positive impact of optimism on quality of
life 6 months postsurgery occurred through the indirect effect of differences
in coping.
Similarly, King et al. (1998) also assessed coping in their study of
women undergoing coronary bypass graft surgery. Although their results
were not entirely consistent across all of the assessment points in their study,
King et al.s optimists displayed more positive thinking during the week
following surgery, engaged in more attempts at finding meaning at 1 month,
and employed less escapism at 12 months. Mediational analyses demonstrated that finding meaning and escapism were responsible, at least in part,
for the relation observed between optimism and negative mood.
The study of adaptation to failed in vitro fertilization described earlier
(Litt et al., 1992) also examined coping. Although the researchers did not
find a relationship between optimism and instrumental coping, they did
find that pessimism related to escape as a coping strategy. Escape, in turn,
related to greater distress after the fertilization failure. In addition, optimists

202

SCHEIER ET AL.

were more likely than pessimists to report feeling that they benefited somehow from the failed fertilization experience (e.g., by becoming closer to
their spouse).
Several studies have examined the relationship between optimism,
coping, and distress among cancer patients. One study followed women
undergoing breast biopsy (Stanton & Snider, 1993).Optimism, coping, and
mood were assessed the day before biopsy in all women participating in the
study. Women who received a cancer diagnosis were then reassessed 24
hours before surgery and 3 weeks after surgery. Women with a benign
diagnosis completed a second assessment that corresponded to either the
second or the third assessment of the cancer group. Pessimists used more
cognitive avoidance in coping with the upcoming diagnostic procedure than
did optimists. This contributed significantly to distress prior to biopsy, and
also predicted postbiopsy distress among women with positive diagnoses.
Another study of cancer patients, mentioned earlier in the chapter,
examined the ways women cope with treatment for early stage breast cancer
during the first full year after treatment (Carver et al., 1993). Optimism,
coping (with the diagnosis of cancer), and mood were assessed the day
before surgery. Coping and mood were also assessed 10 days postsurgery,
and at three follow-up points during the next year. Both before and after
surgery, optimism was associated with a pattern of reported coping tactics
that centered on accepting the reality of the situation, placing as positive
a light on the situation as possible, trying to relieve the situation with
humor, and (at presurgery only) taking active steps to do whatever there
was to be done. Pessimism was associated with denial and behavioral disengagement (giving up) at each measurement point.
The coping tactics that related to optimism and pessimism also related
strongly to the distress that participants reported. Positive reframing, acceptance, and the use of humor all related inversely to self-reports of distress,
both before and after surgery. Denial and behavioral disengagement were
positively related to distress at all measurement points. At the 6-month
point a new association emerged, such that distress was positively related
to another kind of avoidance coping: self-distraction. Not unexpectedly,
given the pattern of the correlations, further analyses revealed that the
effect of optimism on distress was largely indirect through coping, particularly
at postsurgery.
Other studies have been designed to examine the mediational role of
coping in the relationship between optimism and psychological well-being.
In the college adaptation study described earlier (Aspinwall &Taylor, 1992),
optimistic students were more likely than pessimistic students to engage in
active coping and less likely to engage in avoidance coping. Avoidance
coping was associated with poorer adjustment, and active coping was

PSYCHOLOGICAL WELL-BEING

203

(separately) associated with better adjustment. The paths from optimism to

well-being through coping were significant. Thus, as in the health studies
just reviewed, the beneficial effects of optimism in this context seemed to
be operating at least in part through the differences in coping.
Similarly, in the study described earlier concerning adjustment to
pregnancy (Park et al., 1997), optimistic women were more likely than
pessimistic women to engage in constructive thinking (i.e., the tendency
to think about and solve daily problems in an effective way). Furthermore,
as with optimism, constructive thinking also correlated negatively with later
anxiety and positively with later positive states of mind. Subsequent analyses
revealed that the association between optimism and each of these markers
of psychological adjustment was mediated through the tendency of optimists
to engage in constructive thinking.
Finally, recall the study by Segerstrom et al. (1998) examining mood
disturbances among students enrolled in law school. Additional data gathered from that study showed that optimists engaged in less avoidant coping
than did pessimists. Mediational analyses demonstrated that the differences
between optimists and pessimists in the degree of mood disturbance they
experienced were at least partially due to the differences between them in
their use of avoidant coping strategies.
In sum, these studies indicate that optimists differ from pessimists both
in their stable coping tendencies and in the kinds of coping responses that
they spontaneously generate when they are confronting stressful situations.
Optimists also differ from pessimists in the manner in which they cope with
serious disease and with concerns about specific health threats. In general
(see Exhibit 9.1), findings from this research suggest that optimists tend
to use more problem-focused coping strategies than do pessimists. When
problem-focused coping is not a possibility, optimists turn to adaptive emotion-focused coping strategies such as acceptance, use of humor, and positive
reframing. Pessimists tend to cope through overt denial and by mentally and
behaviorally disengaging from the goals with which the stressor is interfering.

Exhibit 9.1.
Coping Tendencies of Optimists and Pessimists
Optimists
Information seeking
Active coping and planning
Positive reframing
Seeking benefit
Use of humor
Acceptance

204

SCHEIER ET AL..

Pessimists
Suppression of thoughts
Giving up
Self-distraction
Cognitive avoidance
Focus on distress
Overt denial

Acceptance, Overcoming Adversity, and Continued Goal-Engagement

It is particularly noteworthy to us that optimists turn toward acceptance

in uncontrollable situations, whereas pessimists tend to turn toward the use
of overt denial. Although both tactics seem to reflect emotion-focused
coping, there are important qualitative differences between them that may,
in turn, be associated with different qualities of outcomes. More concretely,
denial (the refusal to accept the reality of the situation) means attempting
to adhere to a worldview that is no longer valid. In contrast, acceptance
implies a restructuring of ones experience so as to come to grips with the
reality of the situation that one confronts. Acceptance thus may involve a
deeper set of processes, in which the person actively works through the
experience, attempting to integrate it into an evolving worldview.
The active attempt to come to terms with the existence of problems
may confer special benefits to acceptance as a coping response. We want
to be very clear here, however, about the sort of acceptance we are talking
about. The kind of acceptance we have in mind is a willingness to admit
that a problem exists or that an event has happened-even an event that
may irrevocably alter the fabric of ones life. W e are not talking about a
stoic resignation, a fatalistic acceptance of the negative consequences to
which the problem or event might lead, despite how likely those consequences might be. The latter kind of acceptance does not confer a benefit
at all.
Consider, for example, someone diagnosed with an illness such as a
terminal cancer or AIDS. The ultimate outcome in both cases may be death.
Yet, the person can accept the fact that he or she is terminally ill without
simultaneously succumbing to the feeling that he or she is as good as dead.
The latter sort of acceptance, or resignation, may well promote a kind of
functional death, in which the person prematurely disengages from the
opportunities of life. Consistent with this idea there are some findings to
suggest that people who react to illness diagnoses with stoic resignation or
passive acceptance of their own impending deaths, actually die sooner than
do those people who exhibit less of these qualities (Greer, Morris, & Pettingale, 1979; Greer, Morris, Pettingale, & Haybittle, 1990; Pettingale, Morris,
& Greer, 1985; Reed, Kemeny, Taylor, Wang, & Visscher, 1994; for further
discussion of this issue, see Scheier & Carver, in press).
In contrast to resignation to the ultimate consequence of the diagnosis,
limited acceptance of the diagnosis per se may have very different consequences. It may cause people to reprioritize their lives, to realistically revise
and cut back on long-term goals, and to use what time they have left in
constructive and optimal ways. By accepting the fact that life may be
compromised (but not over), people may be impelled to develop a more

PSYCHOLOGICAL WELL-BEING

205

adaptive set of guidelines within which to live the life they have left. W h y
is this revised outlook on life so important, and why should it lead to
enhanced well-being?
T o answer these questions fully, we need to revisit some of the ideas
we introduced in chapter 2 of this volume, describing the theoretical underpinnings of optimism (see also Carver & Scheier, 1998).In particular, recall
the central role that is played by goals in the theory discussed there. In that
view, people live life by identifying goals for themselves and working and
behaving in ways to attain those goals (see Figure 9.1, Loop 1). Obviously,
the goals themselves can take a variety of forms (Carver & Scheier, 1998).
Regardless of their nature, however, goals provide the structure that defines
peoples lives and imbues them with meaning, both in the short run and
in the long run.
Two factors are important in keeping people goal-engaged. One is the
persons ability to identify goals that are valued. People dont take up goals
that dont matter to them, and if they did, they wouldnt persist at them
very long when things got difficult. Valued goals provide the purpose for
living. The second factor is the sensed attainability of the goal. If a goal
seems unattainable at the outset, effortful behavior will never even begin.
If people continually fail to make progress toward goals theyve committed
themselves to, they will begin to withdraw effort and start to perceive those
goals as out of reach. In contrast, hope enables people to hold onto valued
goals, remain engaged in the process of goal striving, and stay committed
to the attempt to move forward (see Figure 9.1, Loop 2).
The final part of the overall picture is that the loss of engagement in
a goal does not necessarily lead to a vacuum of goallessness. Elsewhere,
we have discussed the ebb and flow of engagement and disengagement in
terms of competition among goals (Carver & Scheier, 1998).We suggested
that disengagement from one goal may reflect a weakening of commitment
to it. Because people usually have many goals at once, however, this weakening of commitment permits another goal to become prepotent and thus
permits behavior to shift toward its pursuit. Thus, the loss of one goal is
often followed directly by pursuit of another one.
Sometimes this means scaling back to a less ambitious goal in the
same domain as the abandoned goal. By taking up an attainable alternative,
the person remains engaged in goal pursuit and forward movement (see
Figure 9.1, Loop 3 ) . This is particularly important when the blocked path
concerns a value that is central to the self. People need multiple paths to
these core values (cf. Linville, 1985,1987;Showers & Ryff, 1996; Wicklund
& Gollwitzer, 1982). If one path is blocked, people need to be able to divert
to another one.
Consider a young woman, diagnosed with breast cancer, who has a
great personal investment in family ties and in her career. If chemotherapy

206

SCHEZER ET AL.

Q
Selfregulation

'/alj
Pursue

0
Disergag?
totally

Figure 9.1. Successful self-regulation is a continuing process of identifying goals,

pursuing them, attaining them, and identifying further ones (Loop 1). Given adversity,
a step of evaluating chances of success may be added (Loop 2), but sufficient
confidence places the person back into Loop 1. If confidence is low enough, the
person may seek an alternative goal (Loop 3); if available, this loop returns the person
to goal pursuit and attainment. If, however, the original goal is seen as unattainable
and no alternative is available, the person may disengage completely. From On the
Self-Regulation of Behavior (p. 349) by C. S. Carver and M. F. Scheier, 1998, New
York: Cambridge University Press. Copyright 1998 by Cambridge University Press.
Reproduced with permission.

PSYCHOLOGICAL WELL-BEING

207

destroys her ability to have children, positive family experiences can be

obtained in other ways (cf. Clark, Henry, & Taylor, 1991). If her long-term
career ambitions are threatened by the progression of the disease (e.g., by
impending death), those ambitions might be realized in smaller ways in the
time that remains. Alternatively, she might decide to reprioritize various
aspects of her life and spend less time working and more time in close relat ionships.
In any case, it seems apparent that the ability to shift to a new goal,
or to find a new path to a continuing goal, is a very important part of
remaining goal-engaged. What happens if there is no alternative to take
up? In such a case disengagement from an unattainable goal is not accompanied by a shift, because there is nothing to shift to. This is the worst situation,
where there is nothing to pursue, nothing to take the place of what is seen
as unattainable (cf. Moskowitz, Folkman, Collette, & Vittinghoff, 1996;
Scheier &Carver, in press). If commitment to the unattainable goal remains,
the result is great distress. If the commitment wanes, the result is emptiness.
There is reason to suspect that such a state might also be implicated in
premature death (Carver & Scheier, 1998).

IS OPTIMISM ALWAYS GOOD: IS PESSIMISM ALWAYS BAD?

The picture that we have painted throughout this chapter is one in
which optimists are virtually always better off than are pessimists. The
evidence we have reviewed indicates that they are less distressed when times
are tough, they cope in ways that foster better outcomes for themselves,
and are more proactive in their responses to adversity. Although there are
certainly times and situations in which optimists are only minimally better off
than pessimists, and there are probably cases where they have no advantage
at all, remarkably little evidence exists that optimists are ever worse off
than pessimists.
Several people have suggested the possibility that such situations may
exist, that optimism may be potentially damaging (Schwarzer, 1994; Tennen
& Affleck, 1987). And, indeed, there is logic behind this hypothesis. For
example, too much optimism might potentially lead people to ignore a
threat until its too late, or might lead people to overestimate their ability
to deal with an adverse situation, resulting in poorer outcomes.
Most of the data reviewed in the preceding sections indicate that this
is generally not the case. On the other hand, there are a couple of studies
that suggest the possibility that optimists may not always take action to
enhance their future well-being. Goodman, Chesney, and Tipton ( 1995)
studied the extent to which a group of adolescent girls at risk for HIV
infection sought out information about HIV testing and agreed to have a
208

SCHElER ET A L

test performed. Those scoring higher in levels of optimism were less likely
to expose themselves to the information about HIV-testing, and were less
likely to follow through with an actual test than were those who scored lower
in levels of optimism (see also, Perkins, Leserman, Murphy, & Evans, 1993).
These findings contradict the bulk of evidence reviewed earlier. Why
the findings diverge in this way is not clear. Goodman et al. (1995) noted
that the average level of optimism among the adolescent girls they studied
was substantially lower than the levels typically seen in other sample groups.
This difference may somehow have played a role in the results. Alternatively,
perhaps the results do not really contradict previous findings at all. Perhaps
it seems so only because of the absence of other data that would render the
findings more understandable. For example, no information was gathered
in this study about the knowledge the girls had concerning the serostatus
of their sexual partners. Perhaps optimists had gone to greater lengths than
pessimists to verify that their partners were HIV-negative. If so, they would
have had less need to seek out HIV-relevant information or have their HIV
status tested.
In a similar vein, perhaps the adolescent girls in the Goodman et al.
(1995) study took greater steps to avoid engaging in unsafe sex, and as a
result of their cautiousness, had less need to pursue information about their
HIV status. Consistent with this possibility are recent findings reported in
a study by Carvajal et al. (1998), in which optimistic adolescent girls
expressed stronger intentions to avoid unsafe sex than did their more pessimistic counterparts. Unfortunately, there was no evidence in the Goodman
et al. (1995) study that optimistic participants were being any more cautious
in actual sexual behavior. Moreover, Perkins et al. (1993) found that optimistic gay men were more likely than pessimistic gay men to report engaging
in high-risk sexual behavior, results contradicting those of Carvajal et al.
(1998). Obviously, more work is needed for these questions to be resolved.
The idea that optimists may fail to take steps to protect themselves
against threats is one potential way in which optimism can work against a
person. Another possibility worth considering is that the worldview of an
optimist might be more vulnerable to the shattering impact of a traumatic
event than that of a pessimist. Such an event is more consistent with
the pessimists worldview than with the optimists. Given the diagnosis of
metastatic cancer, the experience of a violent rape, or having ones home
destroyed by fire or flood, who will react worse, the optimist or the pessimist?
Will optimists be less able to rebuild the shattered assumptions of their
lives? All these possibilities are legitimate. Any might occur. However, we
are aware of no evidence that any of them systematically does occur.
Perhaps the lack of support for the idea that optimists respond worse
to a shattering event reflects a more general lack of information about how
qualities of personality relate to responses to trauma, or to experiences such
PSYCHOLOGICAL WELL-BEING

209

as terminal illness. In reality, there is not a great deal of information on

these questions. However, we do not expect optimists to respond adversely.
Rather, we expect them to reset their sights on their changed realities, and
to continue to make the best of the situations they are facing. Pessimists
may find that their worldviews are confirmed by trauma or disaster, but we
doubt that they will take much satisfaction in this confirmation. Rather,
their experience will center on the anticipation of continued adversity, and
on the feelings of distress to which such expectancies are attached.

CONCLUSION
Only relatively recently have researchers given rigorous attention to
the possibility that an optimistic orientation to life is beneficial. But it does
seem to be so. Put simply, optimists emerge from difficult circumstances
with less distress than do pessimists. In part, this optimistic advantage is
due to differences in the manner in which optimists and pessimists cope
with the difficulties they confront. That is, optimists seem intent on facing
problems head-on, taking active and constructive steps to solve their problems; pessimists are more likely to abandon their effort to attain their goals.
When something cannot be changed, optimists try to cast the situation in
the best way possible and accept its reality, rather than trying to will the
event away. In contrast, pessimists are much more likely to engage in
different varieties of avoidance coping.
The evidence obtained thus far provides an interesting window in how
the worlds of the optimist and pessimist differ. O n the other hand, the
picture is not yet complete, or as clear as one might like. For example,
although some forms of acceptance seem to promote well-being and health,
the form that reflects resignation appears more detrimental. Systematic work
needs to be done to understand the nature of the difference between these
forms of acceptance, as well as to determine whether optimists use only one
form and not the other. Additionally, we have speculated that a limited
disengagement from goals can be helpful, in that it allows a person to remain
goal-engaged in the same general life domain that was of interest initially.
Future research should be sensitive to the fact that disengagement from
goals is not always bad-indeed, is sometimes a necessity of living (Carver
& Scheier, 1998; Wrosch, Scheier, Carver, & Schulz, 2000). This research
should attempt to determine the conditions under which goal-engagement
most adaptively gives way to disengagement in order to maximize outcomes
and psychological health. Clearly, aspects of the puzzle linking optimism
and pessimism to psychological well-being need further attention. In light
of what has been revealed thus far, the challenge of refining and clarifying
the picture would seem to be a worthwhile endeavor.

210

SCHEIER ET AL.

REFERENCES
Armor, D. A., & Taylor, S. E. (1998). Situated optimism: Specific outcome expectancies and self-regulation. In M. Zanna (Ed.), Advances in experimental social
psychology (Vol. 30, pp. 309-379). San Diego, CA: Academic Press.
Aspinwall, L. G. (1997). Where planning meets coping: Proactive coping and the
detection and management of potential stressors. In S. L. Friedman & E. K.
Scholnick (Ed.), The developmental psychology ofplanning: W h y , how, and when
do we plan? (pp. 285-320). Mahwah, NJ: Erlbaum.
Aspinwall, L. G., & Brunhart, S. M. (1996). Distinguishing optimism from denial:
Optimistic beliefs predict attention to health threats. Personality and Social
Psychology Bulletin, 22, 993-1003.
Aspinwall, L. G., &Taylor, S. E. (1992). Modeling cognitive adaptation: A longitudinal investigation of the impact of individual differences and coping on
college adjustment and performance.journal of Personality and Social Psychology,
63, 989- 1003.
Aspinwall, L. G., & Taylor, S. E. (1997). A stitch in time: Self-regulation and
proactive coping. Psychological Bulletin, 121, 417-436.
Ausbrooks, D. P., Thomas, S. P., &Williams, R. L. (1995). Relationships among selfefficacy, optimism, trait anger, and anger expression. Health Values, 19, 46-54.
Barron, C. R., Foxall, M. J., von Dollen, K., Shull, K. A., &Jones, P. S. (1992).
Loneliness in low-vision older women. Issues in Mental Health Nursing, J 3,387401.
Billingsley, K. D., Waehler, C. A., & Hardin, S. I. (1993). Stability of optimism
and choice of coping strategy. Perceptual and Motor Skills, 76, 91-97.
Blankstein, K. R., Flett, G. L., & Koledin, S. (1991). The brief College Student
Hassles Scale: Development, validation and relation with pessimism. journal
of College Student Develownt, 32, 258-264.
Bromberger, J. T., & Matthews, K. A. (1996). A longitudinal study of the effects
of pessimism, trait anxiety, and life stress on depressive symptoms in middleaged women. Psychology and Agmg, 11, 207-213.
Carvajal, S. C., Gamer, R. L., & Evans, R. I. (1998). Dispositional optimism as a
protective factor in resisting HIV exposure in sexually active inner-city minority adolescents. journal of Applied Social Psychology, 28, 2196-221 1.
Carver, C. S., & Gaines, J. G. (1987). Optimism, pessimism, and postpartum
depression. Cognitive Therapy and Research, 1 1, 449-462.
Carver, C. S., Pozo, C., Harris, S. D., Noriega, V., Scheier, M. F., Robinson, D. S.,
Ketcham, A. S., Moffat, F. L., & Clark, K. C. (1993). How coping mediates
the effect of optimism on distress: A study of women with early stage breast
cancer. journal of Personality and Social Psychology, 65, 375-390.
Carver, C. S., & Scheier, M. F. (1998). On the self-regulation of behavior. New York:
Cambridge University Press.
PSYCHOLOGICAL WELL-BEING

21 1

Carver, C. S., Scheier, M. F., & Weintraub, J. K. (1989). Assessing coping strategies:
A theoretically based approach. journal of Personality and Social Psychology,
56, 267-283.
Chamberlain, K., Petrie, K., & Azaria, R. (1992). The role of optimism and sense
of coherence in predicting recovery following surgery. Psychology and Health,
7, 301-310.
Chang, E. C. (1998a). Dispositional optimism and primary and secondary appraisal
of a stressor: Controlling for confounding influences and relations to coping
and psychological and physical adjustment. Journal of Personality and Social
Psychology, 74, 1109-1 120.
Chang, E. C. (1998b). Does dispositional optimism moderate the relation between
perceived stress and psychological well-being? A preliminary investigation.
Personality and Individual Differences, 25, 233-240.
Christman, N. J. ( 1990). Uncertainty and adjustment during radiotherapy. Nursing
Research, 39, 17-20, 47.
Clark, L. F., Henry, S. M., & Taylor, D. M. (1991). Cognitive examination of
motivation for childbearing as a factor in adjustment to infertility. In A. L.
Stanton & C. Dunkel-Schetter (Eds.), Infertility: Perspectives from stress and
coping research (pp. 157-180). New York: Plenum Press.
Cohen, J. H. (1990). Community nurse executives psychologic well-being: Relationships among stressors, social support, coping, and optimism. Public Health
Nursing, 7, 194-203.
Cozzarelli, C. (1993). Personality and self-efficacy as predictors of coping with
abortion. Journal of Personality and Social Psychology, 65, 1224-1236.
Curbow, B., Somerfield, M. R., Baker, F., Wingard, J. R., & Legro, M. W. (1993).
Personal changes, dispositional optimism, and psychological adjustment to
bone marrow transplantation. Journal of Behavioral Medicine, 16, 423-443.
Desharnais, R., Godin, G., Jobin, J., Valois, P., & Ross, A. (1990). Optimism and
health-relevant cognitions after a myocardial infarction. Psychological Reports,
67, 1131-1135.
Dunn, D. S. (1996). Well-being following amputation: Salutary effect of positive
mean, optimism and control. Rehabilitation Psychology, 41, 285-302.
Fitzgerald, T. E., Tennen, H., Affleck, G., & Pransky, G. S. (1993). The relative
importance of dispositional optimism and control appraisals in quality of life
after coronary artery bypass surgery. Journal of Behavioral Medicine, 16, 25-43.
Fontaine, K. R., & Jones, L. C. (1997). Self-esteem, optimism, and postpartum
depression. journal of Clinical Psychology, 53, 59-63.
Fontaine, K. R., Manstead, A. S. R., & Wagner, H. (1993). Optimism, perceived
control over stress, and coping. European Journal of Personality, 7, 267-281.
Fontaine, K. R., & Seal, A. (1997). Optimism, social support, and premenstrual
dysphoria. Journal of Clinical Psychology, 53, 234-247.
Friedman, L. C., Nelson, D. V., Baer, P. E., Lane, M., Smith, F. E., & Dworkin,
R. J. (1992). The relationship of dispositional optimism, daily life stress, and

21 2

SCHEIER ET AL..

domestic environment to coping methods used by cancer patients. Journal of

Behavioral Medicine, 15, 127-141.
Friedman, L. C., Weinberg, A. D., Webb, J. A., Cooper, H. P., & Bruce, S. (1995).
Skin cancer prevention and early detection intentions and behavior. American
Journal of Preventive Medicine, 1 1, 59-65.
Fry, P. S. (1995). Perfectionism, humor, and optimism as moderators of health
outcomes and determinants of coping styles of women executives. Genetics,
Social, and General Psychology Monographs, 121, 211-245.
Given, C. W., Stommel, M., Given, B., Osuch, J., Kurtz, M. E., & Kurtz, J. C.
(1993). The influence of cancer patients symptoms and functional states on
patients depression and family caregivers reaction and depression. Health
rsychoiogy, 12, 277-285.
Goodman, E., Chesney, M. A., & Tipton, A. C. (1995). Relationship of optimism,
knowledge, attitudes, and beliefs to use of HIV antibody test by at-risk female
adolescents. Psychosomatic Medicine, 57, 541-546.
Greer, S., Morris, T., & Pettingale, K. W. (1979). Psychological response to breast
cancer: Effect on outcome. Lancet, 2, 785-787.
Greer, S., Morris, T., Pettingale, K. W., & Haybittle, J. L. (1990). Psychological
response to breast cancer and 15-year outcome. Lancet, 1, 49-50.
Guarnera, S., & Williams, R. L. (1987). Optimism and locus of control for health
and affiliation among elderly adults. Journal of Gerontology, 43, 594-595.
Hooker, K., Monahan, D., Shifren, K., & Hutchinson, C. (1992). Mental and
physical health of spouse caregivers: The role of personality. Psychology and
Agmg, 7, 367-375.
Johnson, J. E. (1996). Coping with radiation therapy: Optimism and the effect of
preparatory interventions. Research in Nursing and Health, 19, 3-12.
Khoo, S., & Bishop, G. D. (1997). Stress and optimism: Relationships to coping
and well-being. Psychologia: An International Journal of Psychology in the Orient,
40, 29-40.
King, K. B., Rowe, M. A., Kimble, L. P., & Zerwic, J. J. (1998). Optimism, coping,
and long-term recovery from coronary artery bypass in women. Research in
Nursing and Health, 21, 15-26.
Lauver, D., & Tak, Y. (1995). Optimism and coping with a breast cancer symptom.
Nursing Research, 44, 202-207.
Linville, P. (1985). Self-complexity and affective extremity: Dont put all of your
eggs in one cognitive basket. Social Cognition, 3 , 94-120.
Linville, P. (1987). Self-complexity as a cognitive buffer against stress-related illness
and depression. Journal of Personality and Social Psychology, 52, 663-676.
Litt, M. D., Tennen, H., Affleck, G., & Klock, S. (1992). Coping and cognitive
factors in adaptation to in vitro fertilization failure. Journal of Behavioral Medicine, 15, 171-187.
PSYCHOLOGICAL WELL-BEING

213

Long, B. C. (1993). Coping strategies of male managers: A prospective analysis of

predictors of psychosomatic symptoms and job satisfaction.Journal of Vocational
Behvim, 42, 184-199.
Long, B. C., Kahn, S. E., & Schutz, R. W. (1992). Causal model of stress and
coping: Women in management. Journal of CounselingPsychology, 39,227-239.
Long, B. C., & Sangster, J. I. (1993). Dispositional optimism/pessimism and coping
strategies: Predictors of psychosocial adjustment of rheumatoid and osteoarthritis patients. Journal of Applied Social Psychology, 23, 1069-1091.
Major, B., Richards, C., Cooper, M. L., Cozzarelli, C., & Zubek, J. (1998). Personal
resilience, cognitive appraisals, and coping: An integrative model of adjustment
to abortion. .Journal of Personality and Social Psychology, 74, 735-752.
Marshall, G. N., & Lang, E. L. (1990). Optimism, self-mastery, and symptoms of
depression in women professionals. Journal of Personality and Social Psychology,
59, 132-139.
Moskowitz, J. T., Folkman, S., Collette, L., & Vittinghoff, E. (1996). Coping and
mood during AIDS-related caregiving and bereavement. Annals of Behuvimal
Medicine, J 8, 49-5 7.
Nelson, E. S., Karr, K. A., & Coleman, P. K. (1995). Relationships among daily
hassles, optimism and reported physical symptoms. Journal of College Student
Psychotherapy, 10, 11-26.
Neverlien, P. O., & Backer, J. T. (1991). Optimism-pessimism dimensions and
dental anxiety in children aged to 10 to 12 years. Community Dental Oral
Epidemiology, 6, 342-346.
OBrien, W. H., VanEgeren, L., & Mumby, P. B. (1995). Predicting health behaviors
using measures of optimism and perceived risk. Health Values, J9( l ) , 21-28.
Park, C. L., Moore, P. J., Turner, R. A., & Adler, N. E. (1997). The roles of
constructive thinking and optimism in psychological and behavioral adjustment during pregnancy. Journal of Personality and Social Psychology, 73, 584592.
Perkins, D. O., Leserman, J., Murphy, C., & Evans, D. L. (1993). Psychosocial
predictors of high-risk sexual behavior among HIV-negative homosexual men.
AIDS Education and Prevention, 5, 141-152.
Peterson, C., & Seligman, M. E. P. (1984). Causal explanations as a risk factor for
depression: Theory and evidence. Psychological Review, 91 , 347-374.
Pettingale, K. W., Morris, T., & Greer, S. (1985). Mental attitudes to cancer: An
additional prognostic factor. Lancet, 1, 750.
Raikkonen, K., Matthews, K. A., Flory, J. D., Owens, J. F., & Gump, B. B. (1999).
Effects of optimism, pessimism, and trait anxiety on ambulatory blood pressure
and mood during everyday life. Journal of Personality and Social Psychology,
76, 104-128.
Reed, G. M., Kemeny, M. E., Taylor, S. E., Wang, H. J., & Visscher, B. R. (1994).
Realistic acceptance as a predictor of decreased survival time in gay men
with AIDS. Health Psychology, 13, 299-307.
214

SCHEIER ET AL..

Robbins, A. S., Spence, J. T., & Clark, H. (1991). Psychological determinants

of health and performance: The tangled web of desirable and undesirable
characteristics. Journal of Personality and Social Psychology, 61, 755-765.
Scheier, M. F., & Carver, C. S. (1985). Optimism, coping and health: Assessment
and implications of generalized outcome expectancies. Health Psychology, 4,
219-247.
Scheier, M. F., & Carver, C. S. (1992). Effects of optimism on psychological
and physical well-being: Theoretical overview and empirical update. Cognitive
Therapy and Research, 16, 201-228.
Scheier, M. F., & Carver, C. S. (in press). Adapting to cancer: The importance of
hope and purpose. In A. Baum & B. L. Andersen (Eds.), Psychosocial interventions for cancer. Washington, DC: American Psychological Association.
Scheier, M. F., Carver, C. S., & Bridges, M. W. (1994). Distinguishing optimism
from neuroticism (and trait anxiety, self-mastery, and self-esteem): A reevaluation of the Life Orientation Test. Journal of Personality and Social Psychology,
67, 1063-1078.
Scheier, M. F., Matthews, K. A., Owens, J. F., Magovem, G. J., Lefebvre, R. C.,
Abbott, R. A., & Carver, C. S. (1989). Dispositional optimism and recovery
from coronary artery bypass surgery: The beneficial effects on physical and
psychological well-being. Journal of Personality and Social Psychology, 57, 10241040.
Scheier, M. F., Weintraub, J. K., & Carver, C. S. (1986). Coping with stress:
Divergent strategies of optimists and pessimists. Journal of Personality and Social
Psychology, 51, 1257-1264.
Schuller, I. S. (1995). Cognitive style categorization width and anxiety. Studia
Psychologica, 37, 142-145.
Schwarzer, R. (1994). Optimism, vulnerability, and self-beliefs as health-related
cognitions: A systematic overview. Psychology and Health, 9, 161-180.
Segerstrom, S. C., Taylor, S. E., Kemeny, M. E., & Fahey, J. L. (1998). Optimism
is associated with mood, coping, and immune change in response to stress.
Journal of Personality and Social Psychology, 74, 1646- 1655.
Seligman, M. E. P. (1991). Learned optimism. New York: Knopf.
Shepperd, J. A., Maroto, J. J., & Pbert, L. A. (1996). Dispositional optimism as a
predictor of health changes among cardiac patients. Journal of Research in
Personality, 30, 517-534.
Shifren, K., & Hooker, K. (1995). Stability and change in optimism: A study among
spouse caregivers. Experimental Agmg Research, 21, 59-76.
Showers, C. J., & Ryff, C. D. (1996). Self-differentiation and well being in a life
transition. Personality and Social Psychology Bulletin, 22, 448-460.
Stanton, A. L., & Dunkel-Schetter, C. (Eds.). (1991). Infertility: Perspectives from
stress and coping research. New York: Plenum Press.
Stanton, A. L., & Snider, P. R. (1993). Coping with breast cancer diagnosis: A
prospective study. Health Psychology, 12, 16-23.
PSYCHOLOGICAL WELL-BEING

215

Steptoe, A., Wardle, J., Vinck, J., Tuomisto, M., Holte, A., & Wichstram, L.
( 1994). Personality and attitudinal correlates of healthy lifestyles in young
adults. Psychology and Health, 9, 331-343.
Stewart, S. M., Betson, C., Lam, T. H., Marshall, I. B., Lee, P. W., & Wong, C. M.
(1997). Predicting stress in first year medical students: A longitudinal study.
Medical Education, 3, 163-168.
Strutton, D., & Lumpkin, J. (1992). Relationship between optimism and coping
strategies in the work environment. Psychology Reports, 71 , 1179-1 186.
Sumi, K., Horie, K., & Hayakawa, S. (1997). Optimism, Type A behavior, and
psychological well-being in Japanese women. Psychological Reports, 80, 43-48.
Sweetman, M. E., Munz, D. C., & Wheeler, R. J. (1993). Optimism, hardiness,
and explanatory style as predictors of general well-being among attorneys.
Social Indicators Research, 29, 153-161.
Taylor, S. E., Kemeny, M. E., Aspinwall, L. G., Schneider, S. G., Rodriguez, R.,
& Herbert, M. ( 1992). Optimism, coping, psychological distress, and high-risk
sexual behavior among men at risk for Acquired Immunodeficiency Syndrome
(AIDS). Journal of Personality and Social Psychology, 63, 460-473.
Tennen, H., & Affleck, G. (1987). The costs and benefits of optimistic explanations
and dispositional optimism. Journal of Personality, 55, 377-393.
Tompkins, C. A., Schulz, R., & Rau, M. T. (1988). Post-stroke depression in
primary support persons: Predicting those at risk. Journal of Consulting and
Clinical Psychology, 56, 502-508.
Wicklund, R. A., & Gollwitzer, P. M. (1982). Symbolic self-completion. Hillsdale,
NJ: Erlbaum.
Wrosch, C., Scheier, M. F., Carver, C. S., & Schulz, R. (2000). The importance of
goal disengagement in a positive psychology. Unpublished manuscript, Carnegie
Mellon University, Pittsburgh, PA.
Zeidner, M., & Hammer, A. L. (1992). Coping with missile attack: Resources,
strategies, and outcomes. Journal of Personality, 60, 709-746.

216

SCHEZER ET AL.

10
UNDERSTANDING HOW OPTIMISM
WORKS: AN EXAMINATION OF
OPTIMISTS ADAPTIVE MODERATION
OF BELIEF AND BEHAVIOR
LISA G. ASPINWALL, LINDA RICHTER,
AND RICHARD R. HOFFMAN 111

A clear picture of the mental and physical health benefits associated

with optimism has emerged in the past 15 years. With few exceptions,
people considered optimists appear to achieve better outcomes in a wide
range of situations, including adjustment to life-threatening and chronic
illnesses (e.g., cancer, AIDS) and major life transitions (e.g., moving from
home to college, from one country to another; see Scheier, Carver, &
Bridges, chap. 9, this volume; Taylor & Aspinwall, 1996, for reviews).
However, optimistic beliefs do not appear to be associated with intelligence,
Lisa G. Aspinwall was supported in part by a grant from NSF (SBR-9709677) and by the
Department of Psychology, University of Maryland. Linda Richter was supported in part by a
National Research Service Award from NIMH (1F31 MH11240-01).

217

academic achievement, wealth, or other characteristics that might convey

advantages in life (see, e.g., Epstein & Meier, 1989). How, then, are optimistic beliefs translated into good outcomes in such a wide range of life domains?
That is, what do optimists think and do when encountering threatening,
challenging, or novel situations that may account for these good outcomes?
This chapter addresses the question: How does optimism work?
Throughout the chapter we use the term optimist and the phrase high
in optimism to denote research participants whose scores on standardized
measures of optimism are one standard deviation or more above the mean
in their respective samples. The term pessimist and the phrase low in optimism
denote participants whose scores are one standard deviation or more below
the sample mean. Most of the research described here assesses optimism
through Scheier and Carvers Life Orientation Test (LOT; 1985) of dispositional optimism (see the Introduction of this volume and Scheier, Carver,
& Bridges, 1994, for more detail). Related frameworks, such as Snyders
(1994) work on hope and Armor and Taylors (1998) examination of domain-specific optimistic beliefs, are also discussed.
We begin by reviewing what is known about how optimistic beliefs
are associated with good outcomes. Next, we examine some intriguing recent
evidence demonstrating greater flexibility in the cognitive processing and
behavior of optimists that may explain their ability to adapt successfully to
new situations, especially those situations that are negative or threatening.
In particular, optimists seem to be more able than pessimists to vary their
beliefs and behavior to match important features of the situation at hand.
We will discuss the potential adaptational benefits of being able to moderate
ones behavior in this way and identify some future research directions that
might elucidate the mechanisms that account for such moderation.

POTENTIAL MEDIATORS OF OPTIMISTS GOOD OUTCOMES:

ACTIVE COPING AND THE SELF-FULFILLING PROPHECY
There is strong evidence that the link between optimism and good
psychological adjustment may be explained by the different ways optimists
and pessimists cope with stress (see Taylor & Aspinwall, 1996, for review;
see Scheier et al., chap. 9, this volume). In their study of a large sample of
freshmen college students, Aspinwall and Taylor ( 1992) found that optimists
were more likely than pessimists to report active ways of coping (such as
problem-solving efforts), and they were less likely than pessimists to report
avoidant ways of coping (such as attempts to stop thinking about the problem
or substance use). These ways of coping, in turn, predicted better selfreported psychological adjustment to college for these optimistic students
3 months later. Similar findings have been reported in a number of other
218

ASPlNWALL ET AL.

contexts, ranging from emigration from East to West Germany (Jerusalem,

1993) to coping with surgery for breast cancer (Carver et al., 1993).
With respect to physical health,' there is evidence that optimists may
take better care of themselves, either in response to illness or in order to prevent it. For example, Taylor et al. (1992) found that among gay men who
were HIV-seropositive, those who were optimistic reported greater efforts to
maintain their health through diet and exercise. Similar findings have been
obtained in a study of residents of public housing in New Jersey, in which
project residents who were optimistic were more likely to take actions to protect their personal health and also to improve conditions in their neighborhood (Greenberg, 1997). Such efforts at self-care represent one important
pathway through which optimists are likely to fare better over time.
Thus far, the picture that emerges from studies of coping and health
is that optimists report better adjustment to a wide range of stressors because
they tend to cope actively instead of avoidantly. In other words, optimists
tend to respond to difficulty with continued efforts to solve their problems,
instead of denying the problems or wishing they would go away. Optimists
are also more likely to take constructive steps to safeguard their health.
This pattern of results may be understood as a sort of self-fulfillingprophecy:
Optimists,by definition, are people with favorable expectations about the
future. Such expectations should make success on a given problem seem
more likely and should thereby promote continued problem-solving efforts,
resulting in better outcomes (Carver & Scheier, 1990).
If all life problems and transitions were amenable to improvement
through one's efforts, we would have a complete explanation of the benefits
of optimism. However, many situations involve negative information and
events largely outside our control, such as a diagnosis of life-threatening
illness. In situations in which our problems cannot be solved or are not
entirely amenable to our own efforts, it would be reasonable to predict that
optimists' tendencies to cope actively would work against them, such that
they might expend a great deal of energy trying to solve the unsolvable
and thus expose themselves to disappointment, frustration, and failure.
Interestingly, however, optimists appear to be more likely than people low
in optimism to report disengaging from intractable situations and working
to accept them. In the next section, we review evidence from naturalistic
studies of stress and coping and from a laboratory study of persistence that
suggests that optimists are more-rather than less-likely to disengage from
uncontrollable stressors.
~~~

'There is accumulating evidence that optimism is associated with enhanced immunocompetence and
other health benefits among people under stress (such as first-year law students) and among people
managing serious illnesses (such as HIV infection). A review of this literature is beyond the scope of
this chapter. Please see Taylor, Kemeny, Reed, Bower, and Gruenewald (2000).

UNDERSTANDING HOW OPTIMISM WORKS

219

OPTIMISTS RESPONSES TO UNCONTROLLABLE STRESSORS:

PARADOXES OF DISENGAGEMENT AND ACCEPTANCE
Evidence From Studies of Stress and Coping
The first evidence that optimists appear to be able to moderate their
coping behavior based on whether a stressor is perceived to be controllable
came from two studies by Scheier, Weintraub, and Carver (1986). As described previously (see chap. 9, this volume), optimistic college students
who were asked to describe a prior stressor or to imagine one reported
that they would cope actively with problems that were perceived to be
controllable, but would disengage from active problem-solving efforts and
use coping strategies designed to manage their emotions for problems perceived to be uncontrollable. Similar findings were obtained by Carver and
his colleagues (1993) in a study of women adjusting to surgery for breast
cancer. Compared to pessimists, optimistic women were more likely to
indicate that they had accepted the reality of having had surgery for breast
cancer. Optimists also reported greater use of humor in coping, and they
reported lower levels of both denial (refusing to believe the surgery had
happened) and disengagement (giving up attempts to cope with the aftermath of the surgery). Each of these coping strategies was, in turn, related
to lower psychological distress in the year following the surgery (see chap.
9, this volume for additional detail).
Such moderation of behavior-coping actively with controllable stressors (the aftermath of the surgery) and accepting uncontrollable ones (having
had the surgery)-closely resembles the well-known Serenity Prayer. In the
Serenity Prayer, people wish for the courage to change the things they can
change, the serenity to accept the things they cannot change, and-most
important-the wisdom to know the difference. Embodied in this prayer is
the idea that the key to successful adaptation may lie in being able to
moderate ones beliefs and behaviors to match important objective features
of problems, such as whether these beliefs are amenable to personal control. If
ones situation is indeed uncontrollable, disengagement from active problemsolving efforts and acceptance of the reality of the situation might be adaptive
responses that preserve resources for future coping efforts and limit current
negative outcomes (Aspinwall & Richter, 1999; Aspinwall &. Taylor, 1997).
T o date, most research on optimism has focused on the first part of
the Serenity Prayer, that is, on changing the things one can change through
active coping. However, we have begun to suspect that the last two, largely
unexplored parts of the Serenity Prayer (accepting things one cannot change
and distinguishing things one can change from those one cannot) may play
an important role in optimists good outcomes. Thus far, the findings linking
optimism to greater acceptance of and disengagement from uncontrollable

220

ASPZNWALL ET AL.

situations are not well understood. As noted earlier, most theoretical accounts identify factors such as favorable expectations, generalized selfefficacy beliefs (Schwarzer & Jerusalem, 1995), and more favorable control
appraisals (Chang, 1998) as keys to optimists success. However, if optimists
are predisposed to see their problems as controllable and their likely outcomes
as favorable, why would they be more, rather than less, likely to disengage
from uncontrollable situations and accept negative realities? Carver et al.
(1993) speculated that belief in a good future may make current problems
easier to accept, and thereby, easier to deal with. This explanation may
account for optimists greater acceptance, but it does not seem to account
for how and whether optimists are able to distinguish controllable from
uncontrollable situations.
In trying to understand this phenomenon, we have conducted several
studies to test how optimists respond to negative events and information
of various kinds. As we will review, we find evidence in a number of domains
consistent with the idea that optimists beliefs and behaviors are responsive
to meaningful variations in the situations or problems presented to them.
We first describe a study of how optimists allocate effort to controllable
versus uncontrollable problems, and then examine optimists responsiveness
to other kinds of information.
Experimental Study of Optimism and Responses
to Unsolvable Problems

In an effort to create a laboratory analog to the kinds of situations

described in the Serenity Prayer, we examined how college students responded to a challenging test of verbal intelligence in which some problems
were solvable and some were unsolvable. We were especially interested in
how participants would respond to an initial set of problems that could not
be solved. Would those who were optimistic quickly identify the problems
as unsolvable within the time limit and move on to the rest, or would
they use most of their time trying to solve the unsolvable problems, thus
compromising their performance on the other problems?
We created this situation by modifying the widely used nonproductive
persistence paradigm. In a typical study, participants are given a set of
anagrams or other puzzles that, unbeknownst to them, cannot be solved,
and the time they spend working on these before giving up is measured.
Because the problems are unsolvable, longer solution times are interpreted
as nonproductive persistence and are said to represent maladaptive coping.
There are several potential difficulties with this procedure and with the
interpretation of greater persistence as maladaptive (see Aspinwall & Richter, 1999, for discussion). One problem is that it is usually reasonable to
assume that a problem one has been given can be solved, especially if it is
UNDERSTANDlNG H O W OPTlMISM WORKS

22 1

the only task one has been given and time has been allotted for one to
work on it. A related problem is that there is nothing else to do in the
experimental setting but work on the unsolvable task. Under such conditions, persistence cannot really be said to be maladaptive, because persistence
on the unsolvable task does not compromise efforts on other tasks.
To provide a better test of whether people can disengage from unsolvable problems and devote their efforts to solvable ones, we told college
student participants that they would be completing sets of anagram puzzles
as a test of verbal intelligence.* Participants were given a 20-minute time
limit and instructed to solve as many problems as possible. For all participants, the first seven anagrams were completely unsolvable. For some participants, the seven unsolvable anagrams were the only task they were given
(no alternatives condition), but for other participants there were two additional sets of seven anagrams each, and these alternative sets contained
solvable puzzles (although participants were not given any information about
the solvability of any of the problems). Additionally, some participants were
allowed to return to previously unsolved trials, whereas others were not
allowed to do so.
The results were striking. In the absence of alternative sets of puzzles, all
participants (regardless of their level of optimism) worked on the unsolvable
problems virtually until the end of the time limit. In contrast, when alternative tasks were available, participants high in optimism who were not allowed
to return to previous trials disengaged approximately 4 minutes sooner from
the unsolvable anagrams than did participants low in optimism. Note that
for participants who were not allowed to return to previous trials, moving
on to the solvable anagrams required admitting that one could not solve
the first seven consecutive problems in a test of verbal intelligence.
An additional finding of the study was that optimists tended to outperform participants low in optimism on the first set of subsequent solvable
trials when these trials were presented as alternative ways of demonstrating
high verbal intelligence. These results suggest that optimists are able to
disengage from unsolvable tasks in order to allocate effort to solvable tasks
and are somewhat more likely to be successful on the solvable tasks, perhaps
because they have preserved physical resources (e.g., time) or emotional
resources (e.g., positive mood) by spending less time on problems that cannot
be solved.
These findings are potentially important because they suggest that
optimists are able to recognize and disengage from unsolvable problems and
to devote problem-solving efforts to aspects of a situation that are controlla-

"or ease of presentation, we have simplified the design and results of this study. Please see
Aspinwall and Richter (1999) for a complete presentation.

222

ASPINWALL ET AL.

ble or solvable. They also provide complementary evidence to studies of

coping in naturalistic settings (Carver et al., 1993; Scheier et al., 1986). In
self-report studies of coping, several alternative explanations for optimists
superior adjustment are always present, such as the possibility that optimists
report better outcomes because their problems are objectively less serious
than those of pessimists or that optimists reports of superior adjustment
reflect a positive response bias or denial (see Aspinwall & Brunhart, 1996,
for discussion and evidence). The design of the anagram study allowed us
to present exactly the same (although, admittedly, much less important,
severe, or enduring) problems to all participants, and to obtain objective
measures of variations in problem-solving effort and performance. We
obtained results that are quite consistent with those of naturalistic studies
of coping among optimists, namely, that optimists who are exposed to
unsolvable tasks do not invariantly persist on them and instead allocate
effort to alternatives when they are available. Again, such findings are
intriguing because they cannot be explained solely by the more favorable
expectations optimists typically hold, or by other documented tendencies,
such as the tendency to appraise situations as more amenable to personal
control.

ADDITIONAL EVIDENCE FOR OPTIMISTS MODERATION OF

BELIEFS AND BEHAVIOR
In the following sections, we review two additional kinds of evidence
that suggest optimism is associated with the successful moderation of beliefs
and behavior, depending on important features of a situation or problem,
such as whether it is controllable. These findings suggest that optimists
greater ability to distinguish controllable from uncontrollable situations may
be part of a larger pattern of successful moderation of belief and behavior
in ways that correspond well to objective features of a situation, especially
situations that present the possibility of threat or loss.
Situated Optimism and the Moderation of Optimistic Beliefs
in Specific Domains
Additional evidence for the adaptive moderation of belief and behavior
among optimists comes from research on more domain-specific, rather than
general, optimistic beliefs. Armor and Taylor (1998) reviewed several ways
in which the beliefs and behavior of people with favorable expectancies
about specific tasks or outcomes appear to be responsive to real and important
features of the situation. Specifically, they examined cases in which the
well-documented optimistic bias-the tendency to believe that one is at
UNDERSTANDING HOW OPTIMISM WORKS

223

lower risk than others for negative events or that ones chances of success
are greater than those of others (Taylor & Brown, 1988;Weinstein, 1987)appeared to be tempered by reality.
The degree to which optimistic beliefs are responsive to reality is a
critical issue in understanding whether and how optimistic beliefs are adaptive, because the belief that one is at less risk than others for negative
outcomes across the board may interfere with important precautions, such
as prudent driving, safer sex, and other preventive health behaviors. Additionally, if optimists beliefs about their abilities or outcomes were not
responsive to reality, they would be in danger of entering situations for
which they were unprepared and of suffering dramatic disconfirmations of
their favorable expectations.
Armor and Taylor ( 1998) identified three major ways in which optimistic beliefs appear to be grounded or situated in reality. The first is that
people do not appear to be indiscriminantly optimistic, but, similar to the
dispositional optimists described earlier, seem to vary their beliefs depending
on whether they perceive events to be controllable. For events perceived
to be controllable, the optimistic bias is large and robust; that is, people
say they are much less likely than other people like them to experience
negative outcomes. However, for events perceived to be uncontrollable, the
optimistic bias is reduced or eliminated. A second major finding of Armor
and Taylors review was that the magnitude of the optimistic bias seemed
to depend on whether a participants performance could be verified. Unlike
height or income, some desired attributes are harder to define and measure
(e.g., sense of humor or savoir faire). When performance can be easily
measured, people tend to have realistic assessments of their abilities and
prospects, that is, they do not inflate their assessments relative to those of
other people. It is only when people can hold favorable beliefs that are less
subject to disconfirmation that they rate themselves as especially wonderful.
Finally, Armor and Taylor noted that although people seemed to be more
optimistic than would be warranted, their beliefs were not so distant from
reality as to get them into trouble.
These findings suggest that there is a strategic element to optimistic
beliefs, that people are able to hold them (and enjoy their motivational
benefits) when they will serve them well (situation is controllable) and
when they will not get them into trouble (they cannot be disconfirmed),
but people temper their optimistic beliefs when they may put them at
risk (failure, disconfirmation, embarrassment). Consistent with this point,
research on the role of optimistic beliefs at different points in making
decisions and implementing goals suggests that people seem to use their
optimism to fuel actions toward goals, but suspend it when they are making
plans about how to meet their goals, an activity that requires a more realistic
assessment of their capabilities and prospects (Taylor & Gollwitzer, 1995).

224

ASPINWALL ET AL.

From their review, Armor and Taylor concluded that optimistic beliefs
are bounded (i.e., they dont get too far out of line); strategic (i.e., they help
people meet their goals and are used selectively, rather than indiscriminantly); and responsive (i.e., they are adjusted to match features of a situation). Consistent with our research on coping and problem solving, all three
of these properties of optimistic beliefs suggest that optimists may be skilled
in knowing which kinds of situations are amenable to change (and which
are not) and in modifying their beliefs and behavior accordingly.

Optimism and the Moderation of Cognitive Processing of Information

A third kind of evidence in support of the idea that optimists are more
flexible and adaptive in their consideration of information about potential
problems and stressors comes from studies of optimists processing of different
kinds of information about threats to health and well-being (Aspinwall &
Brunhart, 1996, in press). The results of these studies suggest that optimists
differ from pessimists in two potentially important ways: (a) Optimists pay
more attention to negative information, remember more of it, and show
evidence of greater elaborative processing of it, and (b) rather than devoting
attention to all of the information presented, optimists pay particularly close
attention to the most useful information available. In contrast, pessimists
often pay less attention to negative information, and they do not vary their
attention to such information as a function of its relevance to the self or
to other potentially important properties.
Before describing these studies, it may be useful to review the impetus
for them. They were initially undertaken to test a common alternative
interpretation of the finding that optimism is consistently associated with
better reported adaptation to negative life events, namely, that optimists
report better outcomes not because they are better adjusted, but because
they deny the gravity of their situations. The present studies tested whether
optimism functioned as does denial or other defensive processes by assessing the prospective relation of optimism to attention to threatening information as a function of increasing severity or self-relevance. If, as the
information became more threatening, optimists attention to such information decreased, such a finding would support the idea that optimism functions similarly to denial. In contrast, if optimists attention increased as
the information became more threatening, such a finding would provide
evidence that optimism was potentially adaptive in confronting negative
information.
In one of our first studies in this area, we presented college students
with information about the risks and benefits of their own health behaviors.
Participants selected for their practice of different health behaviors (vitamin
use, UV exposure) were presented with a menu of information about the
UNDERSTANDING HOW OPTIMISM WORKS

225

benefits and risks of these behaviors, as well as neutral information

(Aspinwall & Brunhart, 1996). We measured reading time for each kind
of information as a function of optimism and participants practice of each
behavior. Two findings were of interest. First, optimists paid more attention
to risk information than to neutral or benefit information. Second, for
vitamin use, optimists attention to risk information (e.g., about vitamin
overdoses) increased as a function of lifetime vitamin use. That is, the
greater the optimists reported use of vitamins, the greater their reading
time for information about its potential risks. These results suggest that
optimism predicts increased attention to risk information, and that it does
so as a function of increasing self-relevance of the negative information
presented. It is of interest that among participants who did not use vitamins,
there was no relation of optimism to reading time for information about
the risks of vitamin use.
These findings about optimists selective processing of health information as a function of how relevant the information is to them have recently
been replicated in two studies by Abele (1999). In Abeles studies, participants were exposed to information about the health-protective benefits of
exercise as well as to neutral information and then asked to recall as much
information as possible in a surprise recall test 1 week later. In her first
study, Abeles participants all were regular exercisers, and optimists were
more likely than pessimists to recall information about the health-protective
benefits of exercise at the follow-up, as compared to neutral information.
In her second study, Abele included students who were not regular exercisers
in addition to those who were. The results show striking evidence for the
moderation of recall for information about exercise as a function of its
relevance to the participants. Among those participants who exercised regularly, the recall advantage for exercise information over neutral information
among optimists was replicated at a 1-week follow-up. I t is of interest that,
among participants who did not exercise regularly, optimism predicted lower
recall for information about exercise. That is, when the information was
relevant to them, optimists remembered more of it than did pessimists, but
when the information was not relevant to them, optimists remembered less
of it than did pessimists. The results of this second study are especially
important because they suggest that there is no main effect relation of
optimism to outcomes, such that optimists alwuys do a particular behavior
(e.g., that they always remember more health-protective information). Instead, these results suggest that optimists direct their attention to whatever
is more relevant to them, given their needs and the situation.
A final piece of evidence consistent with the idea that optimists are
more likely than pessimists to moderate their attention to threatening information based on important features of the situation comes from a study of
attention to information about the risks and benefits of UV exposure among

226

ASPINWALL ET AL.

young adult women who sunbathed or used tanning salons regularly

(Aspinwall & Brunhart, in press). This study was conducted to address two
major limitations of our previous work. First, it is not clear that elevated
reading time among optimists can be interpreted to mean greater acceptance
of the risk information; instead, it is possible that increased reading time
may be in the service of generating counterarguments to refute the information and its negative implications for the self. T o test this possibility, we
conducted stimulated recall interviews with participants following their
exposure to the threatening health information. The interview protocols
were coded for statements consistent with the risk information presented
(elaboration), for statements refuting the risk information (counterarguing),
and for other statements downplaying the risk or its relevance to the self
(downplaying).
A second limitation of our previous work was that the health threats
used (e.g., risks of vitamin use) were not especially serious, nor were participants necessarily engaged in health-compromising behavior. In the present
study, we recruited young women college students who reported moderateto-high levels of weekly UV exposure, either through sunbathing or the use
of tanning booths. To create an especially high level of threat for some
participants, we experimentally manipulated the immediate relevance of
the threatening information by telling half of the participants that the
average melanoma patient was a 25-year-old woman (proximal threat condition). The other half was told that the average patient was a 55-year-old
woman (distal threat condition). In this design, the proximal threat condition presents the greater threat to immediate well-being.
All participants were presented with a computerized menu of information about the medical and cosmetic benefits and risks of UV exposure, as
well as neutral information. After participants had read as much information
as they wished, they were shown a videotape of their on-line session and
asked to describe what they were thinking at each point. An analysis of
the interview protocols revealed that optimists in the proximal threat condition were more likely to elaborate on the risk information and how it related
to them than were pessimists, whereas the reverse was true in the distal
threat condition. For example, one optimist in the proximal threat condition
said, I didnt know that it [skin damage] was cumulative. That kind of
scared me, because I tan a lot. So I got a little more information on that.
Another participant remarked, This is interesting, I thought, because it
talks about when your tan fades, it [the effect or damage] stays there. Because
I do tan a decent amount of time, so it made me think a little bit. You
forget its so risky because I am young now. It is important to note that
there was no evidence in the protocols that optimists were counterarguing or
downplaying the risk information in either condition. These results provide
additional evidence that optimists do indeed elaborate, rather than counterUNDERSTANDING HOW OPTIMISM WORKS

227

argue, threatening information about their own behavior, and that they do
so as a function of whether the threat is said to be proximal or distal.

BENEFITS OF OPTIMISTS MODERATION OF BELIEF, BEHAVIOR,

AND COGNITIVE PROCESSING
Thus far, we have reviewed several lines of research suggesting a similar
conclusion: Optimists seem to be able to moderate their beliefs and behavior
(coping in naturalistic settings, problem solving in experimental settings,
beliefs about themselves and their prospects, and processing of threatening
health information) based on important features of the task, situation, or
information. Note that this moderation of belief, behavior, and processing
effort does not appear to represent a general tendency toward greater variability or creativity in responding, but rather a predictable correspondence of
optimists responses to useful features of the situation, such as whether
negative events are controllable, self-relevant, and proximal.
The ability to moderate ones responses to match important features
of current problems may confer several adaptational benefits. Specifically,
one benefit of recognizing and accepting unchangeable negative realities
rather than trying to change them is that one may save coping resources
and prevent further disappointments. For example, the optimists in our
anagram study who gave up on the first seven (unsolvable) tasks were able
to go on to the solvable ones with sufficient time remaining. A t the same
time, they limited their exposure to unsolvable tasks and to the motivational
and cognitive deficits working on such tasks can create. Disengagement
from unsolvable problems may conserve other resources, such as money and
time. For example, the college sophomore who recognizes after a few failing
grades in organic chemistry that he or she is not meant to be a physician
and changes his or her major has invested fewer resources in an unsuitable
career path than does someone who waits until junior year to reach that
realization. Finally, adaptive disengagement may have the long-term effect
of preserving the favorable beliefs that appear to facilitate such behavior.
Specifically, people may better retain their optimism about future outcomes
by recognizing situations that are not amenable to effort and by allocating
their effort to those that may be. Otherwise, if optimism predisposed people
to nonproductive persistence, it would be difficult to imagine how such
beliefs could be maintained over time.
Armor and Taylors (1998) analysis of situated optimism suggests other
benefits as well. Holding favorable beliefs about oneself and ones prospects
strategically and selectively may reduce the chances that ones beliefs will
lead one to behave in ways that lead to failure, embarrassment, or harm.
For example, the person whose hopes and dreams are pinned on having the

228

ASPINWALL ET AL.

fastest time in the 100 meter race has chosen a desired outcome that is rare
(only one person in the world can have this distinction) and disconfirmable
because it can be obtained only through one particular outcome, whereas
the person who aspires to be the most spirited, inspiring, or original runner
has a lot more leeway in defining success and, therefore, has a much smaller
chance of a crushing public or private disconfirmation of this dream.
Finally, one major way in which optimists moderation of belief and
behavior may prove beneficial is in the flexible and selective allocation of
processing resources, such as attention and elaborative processing, to only
the most useful, self-relevant information in an array. The ability to discern
which kinds of situations and information require attention and which may
be safely ignored allows one to (a) devote ones resources to the most pressing
problems, and (b) avoid the discouragement and frustration that would
result from devoting close attention to all kinds of problems, regardless of
whether they are solvable, relevant to the self, or otherwise worth ones
attention. By being selective, optimists may be able to get the facts, without
letting the facts get to them (Taylor, Collins, Skokan, & Aspinwall, 1989).

THREE EXPLANATIONS OF OPTIMISTS MODERATION OF

BELIEFS AND BEHAVIOR
Our review suggests that optimists are less likely to persist on uncontrollable or unsolvable problems, that they modify their favorable beliefs to be
in line with the situation, and that they selectively direct cognitive processing to useful, self-relevant information, even when that information is negative. We have suggested that many advantages may accrue from such moderations in behavior. A n important next step for research on optimism is to
understand how such moderations are achieved. That is, how do optimists
know the difference between controllable and uncontrollable stressors; how
do they adjust their beliefs and behaviors to match important features of a
problem or task; and how do they selectively process only self-relevant or
otherwise useful information? In this final section, we identify three sets of
potential mechanisms that might underlie these effects.
Parallel Findings in the Study of Positive Mood, Cognition,
and Behavior
We begin our discussion of the mechanisms underlying optimists apparent moderation of beliefs and behavior by noting that there are striking
parallels between the kinds of moderation we have described among optimists
and those found among people in whom a positive mood has been induced
(see Aspinwall, 1998, for a review). It is important to note from the outset
UNDERSTANDING HOW OPTIMISM WORKS

229

that we are not suggesting all positive beliefs and states work in the same
manner or that all the benefits of optimism may be attributable to positive
mood. Instead, we suggest that there may be some common mechanisms
that underlie adaptive patterns of moderation of belief and behavior among
optimists and people who are in positive moods.
First, there is evidence that people in a positive mood are more sensitive
to important features of tasks and situations, such as whether a particular
decision is important, than are people in neutral moods, and that people
in positive moods moderate their behavior accordingly (see Isen, 1993, for
review). For example, studies by Isen and her colleagues show that people
in a positive mood take more risks when a hypothetical or unimportant
task is presented to them, but become more risk averse and conservative
than neutral participants when the stakes are real and high (e.g., when they
are gambling for money or for research participation credits; Isen & Geva,
1987; Isen, Nygren, 6r Ashby, 1988). Under such conditions, positive-mood
participants report more thoughts about losing than do neutral participants
(Isen & Geva, 1987). These findings suggest that when a situation poses
real potential for loss, people in a positive mood carefully consider the
potential negative consequences of their actions and moderate their behavior accordingly.
Second, several studies suggest that people who undergo a positive
mood induction or who experience success on a prior, unrelated task show
greater interest in and greater processing of negative information about the
self. Of particular interest is the fact that these effects, like those found for
optimism, have been found to increase as a function of increasing relevance
of the domain to the self. Specifically, Trope and Pomerantz (1998) found
that people who succeeded on a prior, unrelated task showed greater interest
in reading diagnostic feedback about their personal liabilities as the importance of the problematic domain increased. In contrast, those who had
failed indicated greater interest in reading about their personal strengths as
opposed to their weaknesses. Reed and Aspinwall (1998) found similar
results concerning more rapid orientation to self-relevant health-risk information and lower levels of recall for risk-disconfirming information among
people who had just undergone a positive experience. These findings parallel
those obtained for optimism, in that optimists appear to attend selectively
to risk information that is self-relevant.
Based on our studies of optimism and our review of parallel findings
concerning positive mood and experiences, we have identified three potential explanations of optimists apparent moderation of belief and behavior.
Compared to people low in optimism, optimists may (a) process negative
information more successfully, (b) process information more thoroughly and
flexibly, and (c) differ in the development and refinement of procedural
knowledge and skills in coping and problem solving. It is important to note

230

ASPINWALL ET AL.

that these explanations are not mutually exclusive, either with each other
or with traditional explanations of the benefits of optimism, such as its link
to active coping and persistence. Instead, as we will suggest, these different
processes may work in concert as optimists encounter negative events and
information in ways that enable them to moderate their behavior to match
important features of the situation at hand.

Optimism Facilitates Processing of Negative Information

As we have reviewed, several studies now suggest that optimism and
other positive beliefs and states reliably predict increased, rather than decreased, attention to negative information, if that information is useful or
relevant to the self in some way. Attention to negative information is a
critical first step in coping, problem solving, preventive health behavior, selfimprovement, and a host of other self-regulatory behaviors (see Aspinwall &
Taylor, 1997, for a review). In such domains as education, health, work,
and interpersonal relations, careful processing of information about the
self and the environment, especially negative information, is essential to
obtaining feedback about ones progress toward goals, to formulating appropriate performance standards, and to evaluating behavioral alternatives.
Therefore, any state or belief that contributes to the ability to process such
information is likely to have enormous benefits.
The ability of people in a positive mood or people with positive beliefs
to pay more attention to negative information may account for the more
successful moderation of behavior, belief, and processing effort that we have
reviewed in this chapter. That is, to determine whether a problem is worth
addressing with active coping methods or is something one must accept or
disengage from, one must be able to attend to the problem closely and to
process even unflattering, threatening, or discouraging information in an
unbiased manner. Veridical processing of such information is likely to assist
optimists in developing solutions that correspond well to the objective
features of stressors and problems and to assist them in modifying their
efforts accordingly (Aspinwall & Taylor, 1997).
One explanation for such effects is that positive mood or experiences
may serve as resources that allow people to confront negative information
(Aspinwall, 1998; see also Tesser, Crepaz, Collins, Cornell, & Beach, in
press; Trope & Pomerantz, 1998). That is, the presence of a positive mood
may be a signal that ones current resources are sufficient to deal with the
task at hand. This perceived adequacy of resources may influence a persons
primary motive for responding to negative information in the following
ways: If resources are perceived to be inadequate, one may be motivated to
preserve short-term well-being by denying or counterarguing the information. However, if resources are perceived to be sufficient, one may be able
UNDERSTANDZNG HOW OPTZMlSM WORKS

231

to process such information veridically. In the case of optimism, the belief

that future events are likely to be good may provide a chronically high
estimate of ones affective resources, leading to the belief that one can
weather the costs of attention to negative information in most situations.

Positive Beliefs and States Give Rise to More Thorough

and Flexible Cognitive Processing
There are several other processing differences associated with positive
beliefs and states that may confer advantages in noticing and responding
to information relevant to actual or potential stressors. Isen (1993, in press)
has reviewed several processing advantages conferred by induced positive
mood, including greater creativity and more flexible, thorough, and efficient
decision making and problem solving. In an extension of this work, Isen
and her colleagues have proposed that positive affect leads to increased
flexibility in judgment and behavior because of increased dopamine in two
specific areas of the frontal cortex (Ashby, Isen, & Turken, 1999). Increased
dopamine levels, in turn, are associated with processing differences attributed
to greater flexibility in the executive attention system, such as the ability
to switch set, or change ones understanding of a problem. If positive
mood has similar effects, this dopamine-mediated mechanism might account
for the more flexible and more appropriate problem solving and decision
making as well as the richer view of the task context found among people
in whom a positive mood has been induced. It is possible that similar benefits
accrue to optimists because of their characteristically more favorable mood.
If this is the case, optimists should be better able to adjust their understanding
of situations and problems to reflect new information, a processing difference
that would surely confer advantages in moderating ones behavior to match
the features of the situation.
There are other processing differences that are associated with positive
mood and optimism that may affect the amount and nature of peoples
attention to their surroundings. For example, positive mood seems to promote an external focus of attention (Sedikides, 1992), whereas negative
mood has been reliably associated with increased self-focus and corresponding deficits in self-regulation (see Pyszczynski & Greenberg, 1987, for a
review). A n outward focus may be more likely to promote attention to
useful features of situations. There may also be important differences in how
situations themselves are processed. For example, Basso, Schefft, Ris, and
Dember (1996) found that both positive mood and optimism were positively
related to a global processing bias (processing the overall structure of a
stimulus before that of its component elements), whereas negative states,
such as depression and anxiety, were predictive of a local processing bias
(processing the components before the overall structure). If, for some reason,

232

ASPlNWALL ET AL.

positive beliefs and states are associated with a tendency to process the
whole before the parts, such a tendency could also account for optimists
more successful moderation of belief, behavior, and effort in that they may
be more likely to respond to the features of the entire situation, rather than
to its isolated local elements. It will be important for extensions of this
work to include a neutral comparison group to establish whether positive
states promote a broader processing style (see Fredrickson, 1998), whether
negative states create perceptual narrowing or other deficits (see Derryberry
& Reed, 1994), or both.
Taken together, the increased ability to devote attention to the situation as opposed to oneself, to process the global features of situations, and
to switch set or flexibly change ones understanding of a problem may
confer advantages in formulating responses that correspond well to the
problem at hand. Such processes, working together and in conjunction with
the increased ability to process negative information, may well account for
optimists apparent ability to moderate their beliefs and behavior to match
important features of a situation or task.

Optimism Promotes the Development of Procedural Knowledge About Coping

and Problem Solving
Thus far, our discussion has emphasized differences in attention to
negative information and several aspects of processing style as explanations
for optimists typically more adaptive ways of responding to negative events
and information. We believe that these two processes are functionally related
to a third, namely, the development of procedural knowledge and coping
skills in the course of ones problem-solving efforts. In describing this final
mechanism, we return to our earlier discussion of the relation of optimism
to active coping. There is ample evidence across a wide range of stressors
that optimists are more likely to engage in active coping and less likely to
engage in avoidant coping. One benefit of active attempts to solve problems,
even if such attempts are unsuccessful, is that they are likely to yield information about a problem or situation. For example, if a problem does not respond
to ones efforts, one learns that certain kinds of problem solving may not
be successful for particular kinds of problems, or that some kinds of problems
may not be amenable to active coping of any kind. Optimists, and people
with other beliefs that predispose them to more active forms of coping, such
as beliefs in personal control, may be more likely to try active ways of coping
in a wide range of situations. As a result, they may acquire greater procedural
knowledge about coping strategies and about problems and situations (see
Aspinwall & Taylor, 1997; Skinner, 1997, for discussion). This information
may then allow them to moderate their behavior, based on the features of
the situation.
UNDERSTANDING HOW OPTIMISM WORKS

233

Related formulations, such as Snyders (1994; chap. 5 , this volume)

work on dispositional hope, also note that favorable beliefs about ones
future and a sense that one knows how to meet ones goals go hand in
hand. Specifically, Snyder describes hope as a combination of the will to
act toward ones goals (agency) and a knowledge of the wuys one might
achieve them (pathways). In his framework, the will and the ways are
reciprocally derived through successful problem-solving experience. In support of this idea, there is some evidence that people with high levels of
hope have greater knowledge about specific ways of coping and meeting
their goals. In a role-playing study of coping with cancer, women college
students who were high in hope were more knowledgeable about cancer
and reported greater knowledge of superior coping strategies for dealing
with a potential diagnosis (Irving, Snyder, & Crowson, 1998). What is
interesting here is that none of these women had actually received a diagnosis
of cancer, yet the coping behaviors they reported they would engage in
were those deemed by the researchers as likely to be successful. Findings
like this may reflect greater knowledge about coping and problem solving
that these women were able to apply to the cancer role-playing situation.
Such results are consistent with the present analysis. People who engage
in actions directed toward influencing the environment learn from the
results of those actions a more refined sense of the pathways to meet their
goals and the contingencies that govern whether those actions are likely
to be successful. Over time, then, people who are predisposed to active
coping should acquire a highly refined sense of how to cope with different
situations. Such knowledge should promote problem-solving efforts that
correspond well to important features of particular problems and that should
generalize to new problems. In these terms, the success of optimists may be
seen as a sort of self-fulfilling prophecy, but, it is important to note, not as
one that depends only on the success of optimists active efforts, but also
on the successful application of information and skills gained from the failure
of previous efforts.

How These Potential Mechanisms May Work in Concert

Future research might profitably examine how the process of developing
and refining procedural knowledge and coping skills through active coping
may work in tandem with some of the information-processing differences
associated with optimism. Behaving in ways that elicit information about
ones capabilities, about different ways of coping, and about different kinds
of problems will only be valuable to the extent that one is able to learn
from such experiences. Deriving benefits from ones failed coping attempts
may require quite a bit of attention to negative information, especially
information that is self-relevant. To the extent that optimists are better

234

ASPlNWALL ET

AL.

able to process negative information, they may be more likely to profit

from the information elicited by their attempts at active coping. Similarly,
increased processing flexibility, the ability to switch set, and the tendency
to process global aspects of situations should also increase the informational
gain from ones coping attempts. These processing differences may confer
advantages in detecting and responding to task contingencies, in changing
ones perspective on emerging or changing stressors, and in recognizing
whether tasks or problems are worth continued problem-solving and coping effort.
Conversely, the adaptive value of being able to process negative information and modify ones understanding of it is likely to depend on the
amount of information available. Some unique benefits of optimism may
lie in the information about problems and stressors elicited by active coping
attempts. Specifically, although we have suggested that optimism may share
with positive mood a beneficial relation to the processing of negative information and to several different aspects of processing style, the unique benefit
of optimism may lie in the active coping tendencies promoted by favorable
expectations. Positive mood, in contrast, is associated with few action tendencies (see Fredrickson, 1998, for discussion). In other words, although
positive mood and other positive beliefs may show similar linkages to attention to negative information, optimism has unique relations to more adaptive
coping strategies, such as more active and less avoidant ways of coping,
when concurrent positive mood is statistically controlled (Aspinwall &
Taylor, 1992; see Scheier et al., 1994, for related evidence). Therefore, it
is unlikely that the effects linked to optimism will be solely attributable to
the association of optimism with positive mood. What is much more likely
is that optimism works because optimists enjoy processing advantages that
are linked to positive beliefs and states and because optimists are more likely
to engage in active problem-solving and coping efforts. These sets of processes
working together may underlie optimists more successful moderation of
behavior and, ultimately, better outcomes in a wide range of situations.

CONCLUSION
Several different kinds of evidence reviewed here suggest that the
beliefs and behavior of optimists, unlike those of pessimists, seem to correspond well to the objective features of the situation, especially when the
situation has the potential to pose a threat to health or well-being.
Although such variations in beliefs and behaviors are just beginning
to be understood, they may prove to be an important piece of the puzzle
in understanding the adaptive benefits of optimism. To the extent that
optimists are simultaneously more likely to act in ways that elicit information
UNDERSTANDING H O W OPTIMISM WORKS

235

about the environment and more successful in processing information generated by these attempts, they may be better able to respond appropriately to
important features of tasks or problems, leading to more efficient and successful coping efforts. Continued examination of the mechanisms underlying
optimists and pessimists responses to negative events and information is
likely to provide a more complete account of how optimism works.

REFERENCES
Abele, A. E. (1999). Dispositional optimism, unrealistic optimism and memory for more
or less self-relevant information. Manuscript submitted for publication.
Armor, D. A., & Taylor, S. E. (1998). Situated optimism: Specific outcome expectancies and self-regulation. In M. P. Zanna (Ed.), Advances in experimental
social psychology (Vol. 30, pp. 309-379). New York: Academic.
Ashby, F. G., Isen, A. M., & Turken, A. U. (1999). A neurological theory of positive
affect and its influence on cognition. Psychological Review, 106, 529-550.
Aspinwall, L. G. (1998). Rethinking the role of positive affect in self-regulation.
Motivation and Emotion, 22, 1-32.
Aspinwall, L. G., & Brunhart, S. M. (1996). Distinguishing optimism from denial:
Optimistic beliefs predict attention to health threats. Personality and Socid
Psychology Bulletin, 22, 993-1003.
Aspinwall, L. G., & Brunhart, S. M. (in press). What I do know wont hurt me:
Optimism, attention to negative information, coping, and health. In J. E.
Gillham (Ed.), Dimensions of optimism and hope. Philadelphia, PA: Templeton Foundation.
Aspinwall, L. G., & Richter, L. (1999). Optimism and self-mastery predict more
rapid disengagement from unsolvable tasks in the presence of alternatives.
Motivation and Emotion, 23, 221-245.
Aspinwall, L. G., &Taylor, S. E. (1992). Modeling cognitive adaptation: A longitudinal investigation of the impact of individual differences and coping on
college adjustment and performance.Journal of Personality and Social Psychology,
63, 989-1003.
Aspinwall, L. G., & Taylor, S. E. (1997). A stitch in time: Self-regulation and
proactive coping. Psychological Bulletin, 121, 417-436.
Basso, M. R., Schefft, B. K., Ris, D. M., & Dember, W. N. (1996). Mood and
global-local visual processing. Journal of the International Neuropsychological
Society, 2, 249-255.
Carver, C. S., Pozo, C., Harris, S. D., Noriega, V., Scheier, M. F., Robinson, D. S.,
Ketcham, A. S., Moffat, F. L., Jr., & Clark, K. C. (1993). How coping mediates
the effect of optimism on distress: A study of women with early stage breast
cancer. Journal of Personality and Social Psychology, 65, 375-390.
236

ASPINWALL ET AL

Carver, C. S., & Scheier, M. F. (1990). Principles of self-regulation: Action and

emotion. In E. T. Higgins & R. M. Sorrentino (Eds.), Handbook of motivation
and cognition (Vol. 2, pp. 3-52). New York: Guilford Press.
Chang, E. C. (1998). Dispositional optimism and primary and secondary appraisal
of a stressor: Controlling for confounding influences and relations to coping
and psychological and physical adjustment. Journal of Personality and Social
Psychology, 74, 1109-1 120.
Derryberry, D., & Reed, M. A. (1994). Temperament and attention: Orienting
toward or away from positive and negative signals. Journal of Personality and
Social Psychology, 66, 1128-1139.
Epstein, S., & Meier, P. (1989). Constructive thinking: A broad coping variable with
specific components. Journal of Personality and Social Psychology , 57, 332-350.
Fredrickson, B. L. (1998). What good are positive emotions? Review of General
Psychology, 2, 300-3 19.
Greenberg, M. (1997). High-rise public housing, optimism, and personal and environmental health behaviors. American Journal of Health Behavior, 21,388-398.
Irving, L. M., Snyder, C. R., & Crowson, J. J., Jr. (1998). Hope and coping with
cancer by college women. J o u m l of Personality, 66, 195-2 14.
Isen, A. M. (1993). Positive affect and decision making. In M. Lewis & J. M.
Haviland (Eds.), Handbook of emotions (pp. 261-277). New York: Guilford
Press.
Isen, A. M. (in press). Positive affect and decision making. In M. Lewis & J. M.
Haviland (Eds.), Handbook of emotions (2nd ed.). New York: Guilford Press.
Isen, A. M., & Geva, N. (1987). The influence of positive affect on acceptable
level of risk: The person with a large canoe has a large worry. Organizational
Behavior and Human Decision Processes, 39, 145-154.
Isen, A. M., Nygren, T. E., & Ashby, F. G. (1988). Influence of positive affect on
the subjective utility of gains and losses: It is just not worth the risk. Journal
of Personality and Social Psychobgy, 55, 710-717.
Jerusalem, M. ( 1993). Personal resources, environmental constraints, and adaptational processes: The predictive power of a theoretical stress model. Personality
and Individd Differences, 14, 15-24.
Pyszczynski, T., & Greenberg, 1. ( 1987). Self-regulatory perseveration and the
depressive self-focusing style: A self-awareness theory of reactive depression.
Psychological Bulletin, 102, 122-138.
Reed, M. B., & Aspinwall, L. G. (1998). Self-affirmation reduces biased processing
of health-risk information. Motivation and Emotion, 22, 99-132.
Scheier, M. F., & Carver, C. S. (1985). Optimism, coping and health: Assessment
and implications of generalized outcome expectancies. Health Psychology, 4,
219-247.
Scheier, M. F., Carver, C. S., & Bridges, M. W. (1994). Distinguishing optimism
from neuroticism (and trait anxiety, self-mastery, and self-esteem):A re-evaluaUNDERSTAAJDlNG H O W OPTIMISM WORKS

237

tion of the Life Orientation Test. Journal of Personality and Social Psychology,
67, 1063-1078.
Scheier, M. F., Weintraub, J. K., & Carver, C. S. (1986). Coping with stress:
Divergent strategies of optimists and pessimists. Journal of Personality and Social
Psychology, 51, 1257-1264.
Schwarzer, R., & Jerusalem, M. (1995). Optimistic self-beliefs as a resource factor
in coping with stress. In S. E. Hobfoll & M. W. deVries (Eds.), Extreme stress and
communities: Impact and intervention (pp. 159-177). Boston: Kluwer Academic.
Sedikides, C. (1992). Mood as a determinant of attentional focus. Cognition and
Emotion, 6, 129-148.
Skinner, E. A. (1997). Planning and perceived control. In S. Friedman & E.
Scholnick (Eds.), The developmentalpsychology ofplanning: Why, how, and when
do we plan? (pp. 263-284). Hillsdale, NJ: Erlbaum.
Snyder, C. R. (1994). The psychology of hope: You can get there from here. New York:
Free Press.
Taylor, S. E., & Aspinwall, L. G. (1996). Mediating and moderating processes in
psychosocial stress: Appraisal, coping, resistance and vulnerability. In H. B.
Kaplan (Ed.), Psychosocial stress: Perspectives on structure, theory, life-course,
and methods (pp. 71-110). San Diego: Academic Press.
Taylor, S. E., & Brown, J. D. (1988). Illusion and well-being: A social psychological
perspective on mental health. Psychological Bulletin, 103, 193-2 10.
Taylor, S. E., Collins, R. L., Skokan, L. A., & Aspinwall, L. G. (1989). Maintaining
positive illusions in the face of negative information: Getting the facts without
letting them get to you. Journal of Social and Clinical Psychology, 8, 114-129.
Taylor, S. E., & Gollwitzer, P. M. (1995). The effects of mindset on positive
illusions. Journal of Personality and Social Psychology, 69, 213-226.
Taylor, S. E., Kemeny, M. E., Aspinwall, L. G., Schneider, S. G., Rodriguez, R.,
& Herbert, M. ( 1992). Optimism, coping, psychological distress, and high-risk
sexual behavior among men at risk for Acquired Immunodeficiency Syndrome
(AIDS). Journal of Personality and Social Psychology, 63, 460-473.
Taylor, S. E., Kemeny, M. E., Reed, G. M., Bower, J. E., & Gruenewald, T. L.
(2000). Psychological resources, positive illusions, and health. American Psychologist, 55, 99-109.
Tesser, A., Crepaz, N., Collins, J. C., Cornell, D., & Beach, S. R. H. (in press).
Confluence of self-defense mechanisms: On integrating the self-zoo. Personality
and Social Psychology Bulktin.

Trope, Y., & Pomerantz, E. M. (1998). Resolving conflicts among self-evaluative

motives: Positive experiences as a resource for overcoming defensiveness. Motivation and Emotion, 22, 5372.
Weinstein, N. D. (1987). Unrealistic optimism about susceptibility to health problems: Conclusions from a community-wide sample. Journal of Behavioral Medicine, 10, 481-500.

238

ASPlNWALL ET AL

11
GREAT EXPECTATIONS:
OPTIMISM AND PESSIMISM
IN ACHIEVEMENT SETTINGS
JONATHON D. BROWN AND MARGARET A. MARSHALL

(Youve got to) accentuate the positive;

Eliminate the negative;
Latch on to the affirmative;
And dont mess with mister in between.

-Johnny Mercer, 1944

Johnny Mercer was not a psychologist, but his insights regarding the
role of positive thinking in psychological life are penetrating and insightful.
Mercer, who enjoyed great success as one of the premier lyricists of his day,
suffered from depression. Allegedly, he wrote this song on the way home
from his psychiatrists office in an attempt to remind himself to focus on
his perceived strengths rather than his perceived failings.
The research reported in this article was supported by a Presidential Young Investigator Award from
the National Science Foundation (SBR-8958211).

239

Research during the past 20 years has added empirical weight to Mercers observation. Across a variety of indicators, people who hold positive
beliefs about (a) their personal attributes, (b) their ability to bring about
desired outcomes, and (c) their future, are better off than those who are
more realistic (Aspinwall & Taylor, 1992; Bandura, 1986, 1997; Carver et
al., 1993; Scheier & Carver, 1985, 1987; Scheier et al., 1989; Taylor &
Brown, 1988,1994).These benefits seem to be most apparent in achievement
and health-related settings (Brown, 1998).
Positive self-relevant beliefs are not always advantageous, however.
Too much self-aggrandizement can create interpersonal and intrapersonal
problems (Baumeister, 1989, 1997; Baumeister, Heatherton, & Tice, 1993;
Baumeister, Smart, & Boden, 1996; Colvin & Block, 1994; Colvin, Block,
& Funder, 1995). Indeed, under some conditions, inflated self-views have
been linked to self-destruction and violence toward others (Baumeister
et al., 1996). Understanding when positive self-views have favorable and
unfavorable consequences is, therefore, an important topic for research.
In this chapter, we examine the manner in which expectancies of
success affect task performance and peoples emotional reactions to their
performance outcomes. The issue under consideration is the following: Are
people better served by undertaking achievement-related tasks with high
expectancies of success or are they better off being more modest and realistic?
Before addressing this question, we note that expectancies and optimism are
closely related constructs. The two terms are often used interchangeably,
and high expectancies of success are one component of optimism. As Scheier,
Carver, and Bridges (1994) note, optimists are people who tend to hold
positive expectancies of their future; pessimists are people who tend to hold
more negative expectations for the future (p. 1063). Because of these
parallels, the manner in which expectancies affect task performance is relevant to a volume devoted to the study of optimism and pessimism.

EXPECTANCIES AND PSYCHOLOGICAL RESEARCH

The study of expectancies has a rich legacy in the field of psychology.
The construct figures prominently in self-efficacy theory (Bandura, 1986,
1997); self-regulation theory (Carver & Scheier, 1981);goal-setting theory
(Locke & Latham, 1990);attribution theory (Weiner, 1985,1986); theories
of achievement motivation (Atkinson, 1964; Feather, 1982); and Lewins
work on level of aspiration (Lewin, Dembo, Festinger, & Sears, 1944). In
general, this research indicates that high expectancies of success are beneficial. People who expect to succeed perform better than do those who are
less optimistic. In large part, this is because people with high expectancies
of success work harder and longer and adopt more effective problem-solving

240

BROWN AND MARSHALL

strategies than do those who are pessimistic about their expectancies of

success. These findings fit well with evidence indicating that positive selfrelevant beliefs generally have positive consequences (Brown, 1991, 1998;
Taylor & Brown, 1988, 1994).
Although expectancies have been an active area of research, several
issues remain unresolved. One such issue is the precise manner in which
expectancies affect task performance. A search of the relevant literature
revealed that previous research relating expectancies to performance has
either, (a) examined only two levels of expectancy (high and low) or, (b)
used a correlational approach. These analytic approaches leave unanswered
the question of whether high expectancies facilitate performance, low expectancies undermine performance, or both of these effects occur. It may be,
for instance, that low expectancies are a liability, but that medium and high
expectancies are equally beneficial.
A related question has been raised with respect to the construct of
optimism. Optimism is often viewed as a unitary construct, but it also can
be divided into two correlated dimensions: the presence of an optimistic
outlook and the absence of a pessimistic outlook (G. N. Marshall, Wortman,
Kusulas, Hervig, & Vickers, 1992). This distinction is important because the
presence of a pessimistic outlook seems to be more predictive of psychological
well-being than is the absence of an optimistic outlook (Robinson-Whelen,
Kim, MacCallum, & Kiecolt-Glaser, 1997; Schulz, Bookwala, Knappy,
Scheier, & Williamson, 1996). If expectancies affect task performance in
a similar manner, we should find that low expectancies of success are a
better predictor of performance than are high expectancies of success.
The condition under which expectancies influence performance is
another issue to consider. When examining the manner in which expectancies affect performance, prior researchers have not distinguished between
easy and difficult tasks. This issue is important because expectancies might
not have much of an effect when tasks are easy. Easy tasks, almost by
definition, require little in the way of effort, concentration, and persistence.
Consequently, the benefits high expectancies provide in these areas are
largely irrelevant.
To summarize, previous research has not established whether the relation between expectancies and performance is (a) linear and (b) affected
by task difficulty.

Study 1: Task Performance

We (M. A. Marshall & Brown, 1999) recently conducted an investigation to address the nature of the relation between expectancies and performance. During the first part of the investigation, college students were told
they were about to take a test that measured an intellectual ability called
GREAT EXPECTATIONS

24 1

integrative orientation. Integrative orientation was described as a problemsolving skill involving the ability to find creative and unusual answers to
problems. After attempting to solve several practice problems (of varying
difficulty), the students indicated how many of the 10problems they expected
to solve on the upcoming test. After recording their expectancies, they took
the test. Half of the participants were given a set of easy problems to solve;
the other half were given a set of difficult problems to solve.
To determine the manner in which expectancies affected performance,
we divided the students into 3 groups: those who expected to solve few of
the 10 problems (M = 2.57); those who expected to solve about half of the
problems (M = 5.24); and those who expected to solve most of the problems
(M = 8.40).We then analyzed task performance as a function of expectancies
of success and problem difficulty.
A 3 x 2 (Expectancies x Problem Difficulty) analysis of variance
(ANOVA) on these data revealed a significant two-way interaction.' Inspection of the data shown in Figure 11.1 reveals two effects that are particularly
noteworthy. First, expectancies had very little effect in the easy problem condition, but a sizable effect in the difficult problem condition. Second, in the
difficult problem condition, low expectancies were associated with poor performance, but medium and high expectancies were equally beneficial.
These findings clarify several issues regarding the nature of the relationship between expectancies and performance. As noted earlier, correlational
approaches or two-group studies (high vs. low expectancies) have obscured
the manner in which expectancies affect performance. Our data show (a) that
expectancies do not affect performance when problems are easy and (b) that
low expectancies are associated with poor performance when problems are
difficult, but that medium and high expectancies are equally effective.
Study 2: Emotional Reactions to Performance Outcomes

One conclusion to be drawn from these findings is that low expectancies

are a liability and that high expectancies are not. This conclusion is premature, however, because we have yet to examine the manner in which expectancies influence emotional reactions to performance outcomes. Conceivably, low expectancies might soften the emotional impact of a negative
performance.
William James (1890) appears to be the first psychologist to have
considered this issue. In an oft-quoted passage, James argued that
our self-feeling in this world depends entirely on what we back ourselves
to be and do. It is determined by the ratio of our actualities to our
IPrior to analyzing the data, we performed a log transformation to stabilize the variances among the
experimental conditions.

242

BROWN A N D MARSHALL

2-25

2.00

1.75

-0

0,

Low

Medium

High

Expectancy
figure 11.1. Task performance as a function of problem difficulty and expectancies
of success: Study 1.

supposed potentialities; a fraction of which our pretensions are the

denominator and the numerator our successes; thus, Self-esteem = Success/Pretensions. (p. 310)

Although Jamess use of the term pretensions was not entirely consistent (see
Brown, 1998),he often used the term to refer to a persons aspiration level:
So we have the paradox of a man shamed to death because he is only
the second pugilist or the second oarsman in the world. That he is able
to beat the whole population of the globe minus one is nothing; he
has pitted himself to beat that one; and as long as he doesnt do that
nothing else counts. Yonder puny fellow, however, whom every one
can beat, suffers no chagrin about it, for he has long ago abandoned
the attempt to carry that line as the merchants say, of self at all.
(James, 1890, pp. 310-311)

Jamess point here is that how a person feels about an attained outcome is
not simply a function of the outcome itself-it depends on the standards
the person uses for gauging success and failure. Expectancies of success are
one determinant of aspiration level. All else being equal, a student who
GREAT EXPECTATIONS

243

expects to get an A in a course should be less satisfied with a B than

a student who expects to get a C.
In more general terms, it would seem that James believed expectancies
and positive emotional reactions to performance outcomes are inversely
related: The higher our expectancies, the less satisfied we are with a good
performance and the more dissatisfied we are with a poor performance. We
arent aware of any research that has directly tested this assumption, but it
does have important implications. It holds that high expectancies are a
liability: They lessen positive emotional reactions to success and deepen
negative emotional reactions to failure.
The opposite would seem to hold true for low performance expectancies.
According to James, setting our sights low should magnify our positive emotional reactions to success and minimize our negative emotional reactions to
failure. This seems to be what James (1890) had in mind when he wrote:
[Self-esteem] may be increased as well by diminishing the denominator
as by increasing the numerator. To give up pretensions is as blessed a
relief as to get them gratified.. . . Everything added to the Self is a
burden as well as a pride . . . our self-feeling is in our power. As Carlyle
says: Make thy claim of wages a zero, then hast thou the world under
thy feet. (p. 311)

We conducted a second study to examine the manner in which expectancies affect peoples emotional reactions to success and failure (M. A.
Marshall & Brown, 1999). The procedures were similar to those used in
Study 1, with two exceptions. After taking the test, the participants received
feedback regarding their performance (i.e., they were told how many problems they solved). The participants then indicated how proud, pleased with
themselves, humiliated, and ashamed of themselves they felt (1 = not at all;
7 = very much). These emotions were selected for study because they are
explicitly concerned with how people feel about themselves when they succeed
or fail (see Brown & Dutton, 1995; Brown 6r Marshall, 1998; Dutton &
Brown, 1997). After reversing the scoring for the negative emotions, we
averaged the four items to derive a single emotion score.
As in our earlier study, the number of problems each participant solved
was analyzed using a 3 x 2 (Expectancy x Problem Difficulty) ANOVA.
Replicating the results of our earlier study, the two-way interaction was
significant. Figure 11.2 reveals that, once again, expectancies had virtually
no effect in the easy problem condition, but a substantial effect in the
difficult problem condition. Moreover, in the difficult problem condition,
low expectancies were linked with poor performance, but medium and high
expectancies were equally effective.
To determine how expectancies affected participants emotional reactions to success and failure, we performed a similar ANOVA on the emotion

244

BROWN AND MARSHALL

Figure 11.2. Task performance as a function of problem difficulty and expectancies

of success: Study 2.

ratings. Analysis of these data produced a single main effect of task difficulty.
Inspection of Table 11.1 shows that participants who received the easy set
of problems felt better about themselves (M = 5.52) than did those who
received the difficult set of problems (M = 4.50), and this effect did not
depend on how many problems the participants expected to solve. Substantively, this finding means that expectancies did not influence emotional
responses to success and failure, as James (1890) had hypothesized.

Table 1 1 .l.
Emotion as a Function of Expectancies of Success and Problem Difficulty:
Study 2
Expectancies
Problem difficulty
Hard
Easy

Low

4.17
5.60

Medium

High

4.60
5.47

4.74
5.49

Note. Values could range from 1-7; higher numbers = more positive emotional reactions.

GREAT EXPECTATIONS

245

Table 11.2.
Emotion as a Function of Expectancies of Success and Task
Performance: Studv 2
Expectancies
Task performance
Low
Medium
High

Low

3.84
4.80
6.00

Medium

4.04
5.00
5.93

High

4.80
4.72
5.83

Note. Values could range from 1-7; higher numbers = more positive emotional reactions.

It might be argued that the data in Table 11.1 do not provide a fair
test of Jamess (1890) hypothesis. Ideally, one would want to examine the
match between the number of problems participants expected to solve and
the number of problems they actually did solve. Table 11.2 presents data
of this nature. If James is right, we should find that, at every level of
task performance, those with low expectancies of success experienced more
positive emotions than did those with high performance expectancies. The
data in Table 11.2 provide very little support for this hypothesis. A 3 x 3
(Task Performance x Expectancy) ANOVA produced only a single main
effect of task performance: Independent of their performance expectancies,
participants found that the more problems they solved, the better they felt
about themselves.
The means in the medium performance condition are especially informative. These participants solved between four to seven problems. Yet,
among these participants, those who expected to solve more than seven
problems and fell short of their expectations (high expectancy group) felt
just as proud of themselves as did those who expected to solve less than
four problems and exceeded their expectations (low expectancy group).
In summary, the results from the two preceding studies suggest that low
performance expectancies are not beneficial. They undermine performance
when difficult problems are encountered, and they do not soften the pain
of failure when performances are poor. It appears, therefore, that making
thy claim of wages a zero does not, as James opined, diminish the disappointment of failure.
Study 3: An Experimental Approach to Task Performance

At this point, the above-mentioned conclusion should be regarded as

tentative. Expectancies were not experimentally manipulated in the two
previous studies. People who expect to succeed are quite different from
those who expect to fail, and these differences are associated with peoples
emotional lives. The effect is such that people who hold higher expectancies

246

BROWN AND MARSHALL

of success generally feel better about themselves than do those who are more
pessimistic (Smith, Pope, Rhodewalt, & Poulton, 1989). These preexisting
differences could obscure the manner in which expectancies influence peoples emotional reactions to success and failure.
We conducted a third study to examine the influences of these preexisting differences (M. A. Marshall & Brown, 1999). In this study, we experimentally manipulated expectancies of success. This was accomplished by giving
participants a pretest before they performed the target test. Using random
assignment to conditions, one third of the participants were given a set of
very difficult problems; one third were given a set of moderately difficult
problems; and one third were given a set of easy problems. These variations
were used to manipulate expectancies on the target test.
The target test was the same as the one used in our previous studies.
And, as before, approximately half of the participants received an easy set
of problems and half received a difficult set of problems.
Figure 11.3 shows the number of problems participants solved in each
of the six experimental conditions (after using a log transformation). The
data pattern is highly similar to the one observed in Studies 1 and 2.

Figure 11.3. Task performance as a function of problem difficulty and expectancies

of success: Study 3.

GREAT EXPECTATIONS

247

Table 11.3.
Emotion as a Function of Pretest Performance and Problem Difficulty:
Study 3
Pretest performance
Problem difficulty
Hard
Easv

Poor

3.02
4.05

Medium

2.78
4.14

Good

2.63
4.23

Note. Values could range from 1-5; higher numbers = more positive emotion.

Expectancies had little effect in the easy problem condition, but a substantial
effect in the difficult problem condition. Further analyses within the difficult
problem revealed that participants with low expectancies solved fewer problems than did those with medium and high expectancies, and that the
medium and high expectancy groups did not differ. These findings mirror
the results from our two earlier studies.
Table 11.3 shows participants emotional reactions to their performance on the second test. It is clear that expectancies again had little effect
on these emotional reactions. Instead, participants emotional reactions were
entirely a function of how they performed on the target task. Participants
who succeeded felt better than those who failed, and this effect did not
depend on how well the participants expected to do. These results are
essentially the same as those in Study 2.
In summary, the results from Study 3 replicate and extend the results
of Study 2. For easy problems, expectancies had virtually no effect; for
difficult problems, low expectancies were a liability and moderate and high
expectancies were equally advantageous. Furthermore, low expectancies did
not cushion the pain of failure, and high (and moderate) expectancies did
not diminish the pleasure of success.

Study 4: Application of These Effects in the Real World

Thus far, the data we have collected suggest that low expectancies of
success are of little benefit. They are linked with poor performance at difficult
tasks, and they do not lighten the blow of failure or brighten the glow of
success. It is important to note, however, that the studies we have reported
here have been conducted in laboratory settings. Furthermore, participants
were performing a novel task that measured a fictitious ability, and their
expectancies were apt to be held with only limited certainty. Each of these
aspects calls into question the external validity of our findings. Accordingly,

248

BROWN AND MARSHALL

we conducted a final investigation to see whether our findings extend to

real-world situations.
The participants were 132 University of Washington undergraduates
who were enrolled in an upper division psychology course. Immediately
prior to taking a midterm exam, students indicated how many of the 35
exam problems they expected to solve. Four days later, immediately after
receiving their exam scores, the students rated their current emotional states.
Prior to analyzing the data, we divided the students into groups on
the basis of their expectancies of success and exam performance. Students
who expected to solve 29 or fewer problems were classified as having low
expectancies (M = 26.22; N = 32); those who expected to solve 30 or 31
problems were classified as having medium expectancies (M = 30.22; N =
60), and those who expected to solve 32 or more problems were classified
as having high expectancies (M = 33.10; N = 40). A similar procedure was
used to further divide students into groups on the basis of their actual exam
performance. Students who solved 29 or fewer problems were placed into
the low performance condition (M = 26.65; N = 51); those who solved 30
or 31 problems were placed into the medium performance condition (M =
30.44; N = 32), and those who solved 32 or more problems were placed
into the high performance condition (M = 33.35; N = 49).
A 3 x 3 (Expectancy x Task Performance) ANOVA was performed
on students emotional reactions to their task performance. The ANOVA
revealed a single main effect of task performance. Inspection of Table 11.4
shows that students who did very well on the test felt better about themselves
(M = 4.22) than did those whose performance was moderate (M = 3.59),
and students who performed at a moderate level felt better about themselves

Table 11.4.
Emotion as a Function of Expectancies of Success and Exam
Performance: Study 4
Expectancies
Exam performance

Low
Medium
High

Medium

Low

2.86
3.88
3.94

~~~~

2.76
3.47
4.38

High
-

2.50
3.46
4.35

Note. Values could range from 1-5; higher numbers = more positive emotion.

Because the difficulty of the exam problems was not randomly varied in this study, we did not
examine the manner in which expectancies affected task performance (as we have done in the
previous studies reported in this chapter).

GkEAT EXPECTATIONS

249

than did those whose performance was rather poor (M

expectancies did not affect these emotional reactions.

2.71). Students

DISCUSSION
When people undertake an achievement-related task, they usually
have an idea how they will perform. Some people are optimistic. They have
confidence in their ability and expect to do well. Others are less sanguine.
They doubt their ability and fear the worst. Although pessimism can have
positive consequences under some conditions (Cantor & Norem, 1989;
Norem & Cantor, 1986), optimism seems to be more beneficial. People
who hold positive beliefs about their ability generally fare better than do
those who are pessimistic.
In this chapter, we have explored the manner in which expectancies
of success (a construct quite similar to optimism) affect task performance and
peoples emotional reactions to their task performance. Several assumptions
guided our inquiry into these issues.
Expectancies and Task Performance

First, we predicted that expectancies are important only when people

confront difficult tasks. This prediction derives from the fact that expectancies affect performance by influencing how hard (vigor) and how long
(persistence) people try. An easy task requires little in the way of vigor and
persistence, so it makes sense to assume that expectancies are not important
in this context.
W e also predicted that although low expectancies of success are a
liability when difficult tasks are confronted, moderate or high expectancies
are equally effective. Our thinking here was more tentative, as previous
research did not provide a firm basis for making this prediction. Research has,
however, suggested that pessimism might be a more important psychological
variable than optimism (G. N. Marshall et al., 1992; Robinson-Whelen et
al., 1997; but see also Scheier et al., 1994). These findings suggested to us
that expecting to fail is detrimental when difficult tasks are encountered,
but it isnt necessary to think one is definitely going to succeed. More modest
or moderate expectancies can also be effective.
The research reported in this chapter supported both of these predictions. Expectancies did not affect performance at easy tasks, but they did affect
performance at difficult tasks. Moreover, when difficult problems were encountered, low expectancies were a liability, but moderate and high expectancies were equally advantageous. T o cast these findings in more colloquial
terms, it seems that when the going gets tough, the pessimistic stop going.

250

BROWN A N D MARSHALL

Expectancies and Emotional Reactions to Task Performance

The present research also examined peoples emotional reactions to
their task performances. William James ( 1890) suggested that high expectancies are a burden. They lead people to be less satisfied with success and
more dissatisfied with failure. We found no evidence to support this claim.
People who expected to succeed were not more devastated when they failed
than were people who expected to fail, and people who expected to fail
were not more pleased with themselves when they succeeded than were
those who expected to do well. Instead, expectancies had very little effect
on peoples emotional reactions to their task performances.
The attributions people make for performance outcomes may explain
why we failed to find support for Jamess (1890) claim. Weiner and his
associates (see Weiner, 1985, 1986 for reviews) have shown that peoples
emotional reactions to performance outcomes depend not only on their
level of accomplishment, but also on the attributions they make for their
performance. People who attribute success to high ability feel greater pride
than do those who attribute success to good luck or help from others.
Similarly, people who attribute failure to low ability feel greater shame and
humiliation than do those who blame failure on bad luck or hindrance from
others (Brown & Weiner, 1984).
Attributions are also thought to influence expectancies. Peoples attributions to ability are thought to be more likely when their performance
matches expectancies (Miller & Ross, 1975). If so, people with high expectancies should be more apt to attribute success to high ability than those
with low expectancies, and people with low expectancies should be more
likely to attribute failure to low ability than would those with high expectancies.
Figure 11.4 outlines the nature of these presumed relations. The constellation of factors shown suggests that, under some circumstances, pessimistic expectancies of success can have negative, not positive, emotional consequences. They are associated with increased feelings of shame following
failure and diminished feelings of pride following success. These associations
oppose Jamess claim that making thy claim of wages a zero has positive
emotional consequences, and provide a fruitful topic for future research.

CONCLUSION
In addition to clarifying the manner in which expectancies operate
in achievement-related settings, our findings also bear on Taylor and Browns
(1988) work on positive illusions. After surveying a great deal of literature,
Taylor and Brown concluded that positive self-relevant beliefs are generally
GREAT EXPECTATIONS

25 1

Exwctancies Outcome
Success
High
Failure

Success

Failure

Attribution

Emotion

High ability

Pride

Factors othethan low ability

Factors othethan high ability

Little shame and

humiliation

Little pride

Low ability ____* Shame and

humiliation

Figure 7 7.4. Schematic representation of the manner in which expectancies and

attributions interact to affect emotional reactions to success and failure.

beneficial. Our findings offer a refinement of this position. Although negative

thinking clearly offers no benefits in the situations we studied, a person may
not need to be excessively optimistic either. In the data we collected,
moderate and high expectancies of success were equally advantageous.
This point can, perhaps, best be conveyed by reconsidering Johnny
Mercers advice to accentuate the positive, eliminate the negative, and
dont mess with mister in between. In support of Mercers claim, one should
clearly eliminate the negative when difficult tasks are encountered. Under
these conditions, negative thinking (a) leads to poor performance and (b)
does not cushion the blow of failure. Mercers recommendation to accentuate
the positive also seems sound. Optimistic expectancies are linked with
superior performance on difficult tasks, and do not diminish the luster of
success or exacerbate the pain of failure. Whether one needs to avoid
messing with mister in between is less clear, however. Peoples moderate
performance expectancies functioned very similarly to high performance
expectancies. These moderate expectancies were linked with superior performance at difficult problems, and they did not reduce positive emotional
reactions to success. In this sense, then, they were just as effective as were

252

BROWN A N D MARSHALL

positive expectancies. It appears, therefore, that those people who wish to

mess with mister in between may do so with impunity.

REFERENCES
Aspinwall, L. G., &Taylor, S. E. (1992). Modeling cognitive adaptation: A longitudinal investigation of the impact of individual differences and coping on
college adjustment and performance.Journal of Personality and Social Psychology,
63,989-1003.
Atkinson, J. W. (1964). An introduction to motivation. Princeton, NJ: Van Nostrand.
Bandura, A. (1986). Social foundations of thought and action. Englewood Cliffs, NJ:
Prentice-Hall.
Bandura, A. (1997). Self-efficacy: The exercise of control. New York: W. H. Freeman.
Baumeister, R. F. (1989). The optimal margin of illusion. Journal of Social and
Clinical Psychology, 8, 176-189.
Baumeister, R. F. (1997). Evil: Inside h u m n violence and cruelty. New York:
W. H. Freeman.
Baumeister, R. F., Heatherton, T. F., & Tice, D. M. (1993). When ego threats lead
to self-regulation failure: Negative consequences of high self-esteem.Journal of
Personality and Social Psychology, 64, 141-156.
Baumeister,R. F., Smart, L., & Boden, J. M. (1996). Relation of threatened egotism
to violence and aggression: The dark side of high self-esteem. Psychological
Review, 103, 5-33.
Brown, J. D. (1991). Accuracy and bias in self-knowledge. In C. R. Snyder & D. F.
Forsyth (Eds.), Handbook of social and clinical psychology: The health perspective
(pp. 158-178). New York: Pergamon Press.
Brown, J. D. (1998). The self. New York: McGraw-Hill.
Brown, J. D., & Dutton, K. A. (1995). The thrill of victory, the complexity of
defeat: Self-esteem and peoples emotional reactions to success and failure.
Journal of Personality and Social Psychology, 68, 712-722.
Brown, J. D., & Marshall, M. A. (1998). Self-esteem and emotion: Some thoughts
about feelings. Manuscript submitted for publication.
Brown, J. D., & Weiner, B. (1984). Affective consequences of ability versus effort
attributions: Controversies, resolutions, and quandaries. Journal of Educational
P S Y C ~ O 76,
~O~
146-158.
Y,
Cantor, N., & Norem, J. K. (1989). Defensive pessimism and stress and coping.
Social Cognition, 7, 92-1 12.
Carver, C. S., Pozo, C., Harris, S. D., Noriega, V., Scheier, M. F., Robinson, D. S.,
Ketcham, A. S., Moffat, F. L., Jr., & Clark, K. C. (1993). How coping mediates
the effects of optimism on distress: A study of women with early stage breast
cancer. Journal of Personality and Social Psychology, 65, 375-390.
GREAT EXPECTATIONS

253

Carver, C. S., & Scheier, M. F. (1981). Attention and self-regulation: A control-theory

approach to human behavior. New York: Springer-Verlag.
Colvin, C. R., & Block, J. (1994). Do positive illusions foster mental health?
An examination of the Taylor and Brown formulation. Psychological Bulletin,
116, 3-20.
Colvin, C. R., Block, J., & Funder, D. C. (1995). Overly positive self-evaluations
and personality: Negative implications for mental health. Journal of Personality
and Social Psychology, 68, 1152-1162.
Dutton, K. A., & Brown, J. D. (1997). Global self-esteem and specific self-views
as determinants of peoples reactions to success and failure. Journal of Personality
and Social Psychology, 73, 139-148.
Feather, N. T. (Ed.). (1982). Expectations and action: Expectancy-value models in
psychology. Hillsdale, NJ: Erlbaum.
James, W. (1890). The principles of psychology (Vol. 1). New York: Holt.
Lewin, K., Dembo, T., Festinger, L., & Sears, P. S. (1944). Level of aspiration. In
J. M. Hunt (Ed.), Personality and the behavioral disorders (pp. 333-378). New
York: Holt.
Locke, E. A., & Latham, G. P. (1990). A theory of goal setting and task performance.
Englewood Cliffs, NJ: Prentice-Hall.
Marshall, G. N., Wortman, C. B., Kusulas, J. W., Hervig, L. K., & Vickers, R. R.,
Jr. (1992). Distinguishing optimism from pessimism: Relations to fundamental
dimensions of mood and personality. Journal of Personality and Social Psychology,
62, 1067-1074.
Marshall, M. A., & Brown, J. D. (1999). Expectations and realizutions: The role of
expectancies in achievement situations. Manuscript in preparation.
Miller, D. T., & Ross, M. (1975). Self-serving biases in the attribution of causality:
Fact or fiction? Psychological Bulletin, 82, 213-235.
Norem, J. K., & Cantor, N. (1986). Anticipatory and post hoc cushioning strategies:
Optimism and defensive pessimism in risky situations. Cognitive Therapy and
Research, 10, 347-362.
Robinson-Whelen, S., Kim, C., MacCallum, R. C., & Kiecolt-Glaser, J. K. (1997).
Distinguishing optimism from pessimism in older adults: Is it more important
to be optimistic or not to be pessimistic? Journal of Personality and Social
Psychology, 73, 1345-1353.
Scheier, M. F., & Carver, C. S. (1985). Optimism, coping, and health: Assessment
and implications of generalized outcome expectancies. Health Psychology, 4,
219-247.
Scheier, M. F., & Carver, C. S. (1987). Dispositional optimism and physical wellbeing: The influence of generalized outcome expectancies on health. Journal
of Personality, 55, 169-210.
Scheier, M. F., Carver, C. S., & Bridges, M. W. (1994). Distinguishing optimism
from neuroticism (and trait anxiety, self-mastery, and self-esteem): A reevalua-

254

BROWN AND MARSHALL

tion of the life orientation test. Journal of Personality and S o c d Psychology,

67, 1063-1078.
Scheier, M. F., Matthews, K. A., Owens, J. F., Magovem, G. J., Sr., Lefebvre, R. C.,
Abbott, R. A., & Carver, C. S. (1989). Dispositional optimism and recovery
from coronary artery bypass surgery: The beneficial effects on physical and
psychological well-being. Journal of Personality, 57, 1024-1040.
Schulz, R., Bookwala, J., Knappy, J. E., Scheier, M., & Williamson, G. M. (1996).
Pessimism, age, and cancer mortality. Psychology and Agmg, 11, 304-309.
Smith, T. W., Pope, M. K., Rhodewalt, F., & Poulton, J. L. (1989). Optimism,
neuroticism, coping, and symptom reports: An alternative interpretation of the
Life Orientation Test. Journal of Personality and Social Psychology, 56, 640-648.
Taylor, S. E., & Brown, J. D. (1988). Illusion and well-being: A social psychological
perspective on mental health. Psychological Bulletin, 103, 193-210.
Taylor, S. E., & Brown, J. D. (1994). Positive illusions and well-being revisited:
Separating fact from fiction. Psychological Bulletin, 116, 21-27.
Weiner, B. (1985).An attributional theory of achievement motivation and emotion.
Psychological Review, 92, 548-573.
Weiner, B. (1986). An attributional theory of motivation and emotion. New York:
Springer.

GREAT EXPECTATIONS

255

12
CULTURAL INFLUENCES O N
OPTIMISM AND PESSIMISM:
DIFFERENCES IN WESTERN AND
EASTERN CONSTRUALS OF THE SELF
EDWARD C. CHANG

To be uncertain is to be uncomfortable, but to be certain is to be ridiculous.

-Chinese Proverb

In recent years, psychologists have joined anthropologists,sociologists,

and linguists in recognizing the role of culture in shaping human behavior
(Brislin, 1993; Hofstede, 1980; Jahoda, 1982; Marsella, DeVos, & Hsu,
1985). As a consequence, psychologists are quickly learning that what were
once universally held notions about human nature, such as the need for
positive self-regard, are more valid for understanding the behaviors of Westerners, but are often less valid for understanding the behaviors of nonI would like to acknowledge Chang Suk-Choon and Tae Myung-Sook for their encouragement and
support throughout this project.

257

Westerners, particularly, of Asians (Heine, Lehman, Markus, & Kitayama,

1999). Accordingly, some psychologists have begun to urge researchers to
appreciate the different cultural contexts in which human behavior is studied, distinguishing between those contexts that are individualistic and those
that are collectivistic (U. Kim, Triandis, Kagitcibasi, Choi, & Yoon, 1994;
Triandis, 1995). In general, Western cultures (e.g., those of Europe and
North America) are considered individualistic, given their prevailing emphasis on attending to the needs of the self over others (Greenwald, 1980;
Waterman, 1984). Thus, for most Westerners, the attainment of personal
happiness, rather than of group happiness is highly regarded and sought
after (Heine et al., 1999). Consequently, the self that emerges from such
cultures is one that perceives itself as independent of others. As embodied
in Descartes famous notion of cogito ergo sum, the thinking organism (res
cogitans) is detached from all external objects, including other organisms
(res extensa).
In contrast, the traditional focus in Eastern cultures (e.g., that of many
Asian countries) has been to foster a view of the self as fundamentally
interrelated with others (Doi, 1973; U. Kim et al., 1994; Markus & Kitayama,
1991). Hence, attending to others, experiencing harmonious interdependence with them, and fitting in not only are valued, but also are often
strongly expected among members of such cultures (Weisz, Rothbaum, &
Blackburn, 1984; Yee, 1992). For example, consistent with the Confucian
notion of filial piety, order, hierarchy, and duty are cultural elements that
are often highly valued and reinforced in Asian families. Thus, many Asians
and Asian Americans are raised to show unconditional respect and duty to
their parents while they are living and continued vigilance even after they
have long passed away. For another example, consider one of the Four Noble
Truths of Buddhism, namely, that all existence entails suffering. To progress
to a state of enlightenment, one must not only eliminate individual desires,
but also become attuned to the immanent harmony that exists between all
living things. Hence, in contrast to the Western self, the exemplary self
that is projected here is one that views itself as interdependent with others
(Markus & Kitayama, 1991; Roland, 1988). These noted differences between
Eastern and Western cultures are believed to not only promote two relatively
distinct construals of the self (viz., independent vs. interdependent, respectively), but also to materialize in the differences in the way individuals from
such cultures respectively think, feel, and act (Markus & Kitayama, 1991).
Accordingly, this chapter focuses on the influence of Eastern and
Western cultures on optimism and pessimism, and is organized in two sections. The first looks at some of the notable research findings associated
with the examination of optimism and pessimism between Easterners (Asian
people) and Westerners (White people). Specifically, this section reviews
findings from Heine and Lehmans (1995) study on unrealistic optimism,

258

EDWARD C. CHANG

and Lee and Seligmans ( 1997) study on explanatory style, and then examines
in greater detail findings from Changs (1996a) study on dispositional optimism. The second section includes some implications of these and other
notable findings for assessment and intervention. In particular, Becks ( 1976)
cognitive model of psychological adjustment is used to consider culturespecific interventions for working with distressed Asian and White Americans, and some pitfalls of viewing pessimism as a key marker of depression
in others are discussed in light of recent findings. Finally, this chapter
concludes with a brief section on challenges and future directions for
research.

STUDIES OF OPTIMISM AND PESSIMISM IN

EASTERNERS VERSUS WESTERNERS
From a scientific standpoint, the question of whether differences between Eastern and Western cultures have a significant influence on optimism
and pessimism is one that must be addressed empirically. In that regard,
one of the first major investigations focusing on cultural influences was
conducted by Heine and Lehman (1995), who looked at potential differences
in the expression of unrealistic optimism. Weinstein ( 1980) defined unrealistic optimism as a general tendency to expect that negative events are more
likely to occur to others than to oneself, and conversely, that positive events
are more likely to occur to oneself than to others. Based on a series of
studies, Weinstein ( 1980, 1984) found that college students consistently
judged their own susceptibility to a variety of threatening events and problems (e.g., alcoholism, suicide, auto accident injury) to be less than average
compared with someone like them, thus implying that such events were
more likely to happen to others.
In their first study, Heine and Lehman (1995) examined the tendency
to express unrealistic optimism in 196 Japanese college students and in 90
Canadian college students of European descent. Among the different measures administered, participants were asked to rate the likelihood that they
would experience a number of positive and negative future life events.
Consistent with previous findings and with the view that Westerners tend
to be unrealistically optimistic, these investigators found that Canadian
students expressed a greater tendency to believe that positive events (e.g.,
You will enjoy your career) were more likely to happen to them, whereas
negative events (e.g., Sometime in the future you will have a nervous
breakdown) were more likely to happen to others. In contrast, the opposite
pattern was found for Japanese students. Specifically, Japanese students
indicated that positive events were more likely to happen to others, whereas
negative events were more likely to happen to them. Hence, the Japanese
CULTURAL INFLUENCES

259

students exhibited what Heine and Lehman (1995) referred to as greater

unrealistic pessimism for themselves, rather than unrealistic optimism. Noteworthy is the fact that these cultural differences were found to be significant
both within each cultural group and between the two cultural groups. That
is, Japanese students were significantly more unrealistic in their pessimism
compared with Canadian students.
Nonetheless, as Heine and Lehman (1995) noted, their findings may
have been unintentionally confounded with other factors. Specifically, most
of the event items that were used focused on consequences that only involved
the individual (i.e., had an individualistic focus). Given that Japanese students come from a culture that promotes an interdependent construal of
the self, such items may not have been considered important. If event items
that focused on important consequences to others had been used (i.e., had
an interdependent focus), then Japanese students may have displayed a level
of unrealistic optimism similar to that of Canadian students. To address
this concern, these investigators conducted a second study.
In their second study, Heine and Lehman (1995) examined 105 Japanese college students and 110 Canadian college students of European descent. Similar to their first study, items that asked about the likelihood that
certain events would occur compared with the likelihood they would occur
to others. Unlike the first study, however, the list of events now included
those that were independent (e.g., After growing old, you will find out
that you never realized your most important dreams) and interdependent
(e.g., In the future you will not be able to provide a decent standard of
living for your family). If the previous findings were due to the lack of
interdependent event items on the questionnaire, the current findings should
show Japanese students expressing as much unrealistic optimism for interdependent events as Canadian students would for independent events. However, Heine and Lehman (1995) did not find this to be the case. Rather,
they found again significant cultural differences both within and between
the two cultural groups, similar to findings from their first study. That is,
Canadian students expressed a greater tendency to believe that negative
(independent and interdependent) events were more likely to happen to
others than to them, whereas the opposite pattern was found for Japanese
students. Thus, the type of event made little difference in these results.
Given that Japanese students compared with Canadian students consistently
failed to express unrealistic optimism for themselves, Heine and Lehman
(1995) argued that this cultural difference suggests that the normality of
self-enhancing biases might be specific to Western culture (p. 605).
Although Easterners may not exhibit strong self-enhancing tendencies,
Heine and Lehmans (1995) findings raise the notion that Asians may
express strong self-effacing tendencies (i.e., unrealistic pessimism) or the
tendency to see themselves as more average than others (Mizuno & Yama-

260

EDWARD C. CHANG

guchi, 1997). However, given the bipolar nature of their conceptualization

of optimism and pessimism, Heine and Lehmans findings are ambiguous:
It is impossible to tell if Asians are actually scoring low on self-enhancing
tendencies (optimism), high on self-effacingtendencies (pessimism), or both
(low optimism, high pessimism). T o help address this issue, we now move
to a study that assessed for optimism and pessimism separately.

OPTIMISTIC AND PESSIMISTIC EXPLANATORY STYLE

IN ASIANS AND NON-ASIANS: IS HEIGHTENED PESSIMISM
A KEY SENSIBILITY AMONG ASIANS?
Using an explanatory style framework (see chap. 3, this volume), Lee
and Seligman (1997) looked at differences in expressions of optimism and
pessimism between Easterners and Westerners. Specifically, these investigators studied the responses of 3 12 mainland Chinese, 44 Chinese Americans,
and 257 White Americans on the Attributional Style Questionnaire (ASQ;
Peterson et al., 1982). As noted earlier in the introduction to this volume,
the ASQ provides separate scores for a pessimistic explanatory style (for
negative events) and for an optimistic explanatory style (for positive events).
By including separate measures of optimism and pessimism, findings from
Lee and Seligmans (1997) study can help us discern whether Easterners
are simply less optimistic, more pessimistic, or both compared with Westerners (see Heine & Lehman, 1995).
Beginning with results for specific optimistic explanatory style dimensions associated with attributions for positive events, Lee and Seligman
(1997) found no significant difference between Chinese and White American students on internality. However, mainland Chinese students were found
to have significantly lower internality scores than those of Chinese American
and White American students. Chinese American students were not significantly different on stability compared with White American and mainland
Chinese students. O n the other hand, mainland Chinese were found to
have significantly lower stability scores than did White American students.
For globality, a similar pattern emerged. That is, Chinese American students
were not found to be significantly different on globality compared with White
American and mainland Chinese students. However, mainland Chinese
students were found to have significantly lower scores than did White
American students on this attributional dimension for positive events.
For specific pessimistic explanatory style dimensions associated with
attributions for negative events, no significant difference between Chinese
and White American students was found on internality. However, mainland
Chinese students were found to have significantly lower scores on internality
than did Chinese and White American students. For stability, again, no
CULTURAL INFLUENCES

26 I

significant difference between Chinese American and mainland Chinese

students was found. In contrast, White American students were found to
have significantly lower scores on stability for negative events compared with
scores of Chinese American and mainland Chinese students. For globality,
Chinese American students were not found to be significantly different
from mainland Chinese and White American students. However, mainland
Chinese students were found to have significantly higher scores on globality
than did White American students. Taken together, these findings provide
a very interesting but also complex picture of how mainland Chinese,
Chinese American, and White American students differ on specific attribution dimensions for positive and negative events.
Yet, as has increasingly become the convention in studies of explanatory style, another way to consider differences in optimism and pessimism
between the three groups is to compare their general explanatory styles by
using composite scores (sum of internal, stable, and global scores). There
are at least two advantages for doing so. Namely, findings based on composite
scores are likely to be more reliable, and findings based on composite scores
are also easier to interpret than findings based on specific attribution dimensions. Therefore, we now consider Lee and Seligmans (1997) comparative
findings for optimism and pessimism when composite explanatory style scores
were used. For optimistic explanatory style, Lee and Seligman (1997) found
no significant difference between Chinese and White American students.
However, mainland Chinese students were found to have a significantly
lower optimistic explanatory style than did Chinese and White American
students. That is, Chinese American students appeared to be just as optimistic as White American students, whereas mainland Chinese students were
less optimistic than both their Chinese and White American counterparts.
It is noteworthy that, for pessimistic explanatory style, these investigators
found no significant difference between mainland Chinese and Chinese
American students. However, the pessimistic explanatory styles of mainland
Chinese and Chinese American students were found to be significantly
greater than those of White American students. Thus, Chinese American
students appeared to be just as pessimistic as mainland Chinese students,
whereas White American students were less pessimistic than both Asian
groups. These noted findings, especially for general explanatory styles, offer
a more useful picture of the differences in optimism and pessimism between
Easterners and Westerners than previously presented (see Heine & Lehman,
1995). For example, Lee and Seligmans (1997) findings indicate that some
Asian groups (viz., Asian Americans) are just as self-enhancing or optimistic
as Westerners (e.g., White Americans).
Thus far, I have presented findings from recent studies that indicate
group differences in optimism and pessimism between Easterners and Westerners. However, group differences tell us little about relational differences.

262

EDWARD C. CHANG

For example, even if elevations on optimism are comparable between Asian

and White Americans as found in Lee and Seligmans (1997) study, this
does not necessarily mean that optimism is related to other variables in the
same manner for the two ethnic groups. Accordingly, we now move to a
study that looked at the relations between optimism, pessimism, and other
important variables (viz., coping and adjustment) between Asian and
White Americans.

DISPOSITIONAL OPTIMISM AND PESSIMISM AMONG ASIAN

AND WHITE PEOPLE: THE POWER OF PESSIMISTIC THINKING?
According to Scheier and Carvers dispositional model ( 1985),optimism
and pessimism, defined as generalized positive and negative outcome expectancies, represent individual differences variables that promote or abate
psychological and physical adjustment (see chaps. 2 & 9, this volume, for
more detailed discussions of this dispositional model). Specifically, these
researchers have argued that optimism is associated with adaptive coping
efforts (e.g., problem solving, seeking social support) and leads to securing
positive outcomes, whereas pessimism is associated with maladaptive coping
efforts (e.g., problem avoidance, social withdrawal) and leads to incurring
negative outcomes. Consistent with this view, numerous empirical studies
have shown that optimists are psychologically and physically better adjusted
than their more pessimistic counterparts (see chaps. 6 & 9, this volume).
The consistency and robustness of such findings over the past 2 decades
have resulted in an almost incontrovertible acceptance of the view that
optimism is always good, and pessimism is always bad.
However, according to Chang and his associates (Chang, 1996a;
Chang, DZurilla, & Maydeu-Olivares, 1994; Chang, Maydeu-Olivares, &
DZurilla, 1997), at least two issues remain unresolved in Scheier and Carvers ( 1985) dispositional model of optimism and pessimism. First, there
has been considerable debate and controversy over the dimensionality of
optimism and pessimism. Scheier and Carvers ( 1985) conceptualization of
optimism and pessimism as measured by their Life Orientation Test (LOT)
represents these constructs as polar opposites on a unidimensional continuum. However, as noted in the introduction, a number of investigators have
seriously challenged this unidimensional model, arguing that dispositional
optimism and pessimism are better conceived as representing two partially
independent dimensions.
A second, related issue involves the generalizability of previous findings
about the study of dispositional optimism and pessimism for different cultural
groups. As Chang (1996a) noted, because most of the published studies in
psychology have been based on largely White samples, we know very little
CULTURAL INFLUENCES

263

about optimism and pessimism among different racial or ethnic groups and
about how they and related variables such as the use of different coping
strategies are associated with psychological and physical health outcome
(p. 113). T o address these concerns, Chang (1996a) conducted a study
that looked at cultural differences between Asian Americans and White
Americans on optimism, pessimism, coping, and adjustment.
Changs (1996a) study involved 111 self-identified Asian Americans
who were matched on age and sex with a random sample of 111 selfidentified White Americans. It is noteworthy that the responses provided
by mainland Asian (or international) students were not included in this
study (cf. Lee & Seligman, 1997). All participants completed a set of
measures at two points in time. At Time 1, participants completed the
Extended Life Orientation Test (ELOT; Chang et al., 1997) which provides
separate measures of optimism and pessimism. As noted in the introduction
to this volume, the ELOT is an extended measure of the original LOT, and
is based on the same conceptual definition proposed by Scheier and Carver
(1985). However, unlike the LOT, the ELOT is considered a bidimensional
measure that provides separate scores for optimism and pessimism (Chang,
199813; Chang et al., 1997).
At Time 2, approximately 6 weeks later, participants completed measures assessing psychological and physical adjustment. Specifically, psychological adjustment was assessed by the Beck Depression Inventory (BDI;
Beck, Ward, Mendelson, Mock, & Erbaugh, 1961) and by the Symptoms
Checklist-90-Revised (SCL-90-R; Derogatis, 1983). The BDI assesses for
symptoms of depression (e.g., I am so sad or unhappy that I cant stand
it), whereas the SCL-90-R assesses for general psychological symptoms
(e.g., anxiety, hostility, obsession, and compulsion). Because optimism and
pessimism have been found to be related to physical adjustment, the Pennebaker Inventory of Limbic Languidness (PILL; Pennebaker, 1982) also was
included. The PILL assesses for physical symptoms (e.g., coughing, insomnia,
upset stomach, and headaches).
In addition, because coping also has been linked to adjustment (Lazarus
& Folkman, 1984), a measure of coping strategies also was included in
Changs (1996a) study. Specifically, coping was assessed with the Coping
Strategies Inventory (CSI; Tobin, Holroyd, Reynolds, & Wigal, 1989).
Consistent with other measures of coping strategies such as the Ways of
Coping Checklist (Folkman & Lazarus, 1980), this measure provides an
assessment of several engaged coping strategies (viz., problem solving, cognitive restructuring, expressing emotion, and seeking social support) and several disengaged coping strategies (viz., problem avoidance, wishful thinking,
self-criticism, and social withdrawal). The CSI was administered with the
ELOT at Time 1.

264

EDWARD C. CHANG

Several important questions were addressed in Changs (1996a) study.

First, would Asian Americans differ on optimism or pessimism from White
Americans? Second, would optimism or pessimism predict psychological and
physical adjustment among Asian Americans as among White Americans?
If so, would optimism and pessimism account for a similar amount of the
variance in adjustment across the two cultural groups? And third, would
optimism and pessimism have the same relational pattern with coping and
adjustment among Asian Americans as among White Americans?

Do Asian and White Americans Differ on Levels of Optimism

and Pessimism?
Table 12.1 presents the results o f t tests reported in Chang (1996a),
which compared differences in optimism, pessimism, coping strategies, and
psychological and physical adjustment between Asian Americans and White
Table 12.1.
Ethnic Group Differences in Optimism, Pessimism, Coping Strategy,
and Health Outcome
Ethnic group
Asian
Criterion
Outcome expectancy
Optimism
Pessimism
Coping strategies
Problem solving
Cognitive
restructuring
Express emotions
Social support
Problem avoidance
Wishful thinking
Self-criticism
Social withdrawal
Health outcome
Depressive
symptoms
Psychological
symptoms
Physical symptoms

White
SD

SD

t(1, 220)

21.33
25.88

3.62
6.36

22.20
21.14

26.28

6.11

3.41
6.58

-1.83
5.45

26.01

6.88

0.31

ns

26.48
24.41
28.94
21.42
26.30
22.53
21.55

6.44
6.49
7.33
5.33
7.14
8.71
6.63

26.51
23.59
26.57
17.19
24.36
18.66
18.45

7.46
8.29
8.97
5.29
8.05
9.49
5.90

-0.04
0.81
2.16
5.94
1.90
3.17
3.68

ns
ns
ns

10.13

8.02

7.62

7.41

2.42

76.83 53.74 53.41 45.49

97.78 26.09 102.57 25.01

3.51
0.16

ns

.000026

.0000026

.0013

ns
ns

ns
c

.0013
ns

Note. For both ethnic groups, N = 111. All criteria were assessed at Time 1, except for health outcome,
which was assessed at Time 2. From Cultural Differences in Optimism, Pessimism, and Coping: Predictors of Subsequent Adjustment in Asian American and Caucasian American College Students, by E. C.
Chang, 1996, Journal of Counseling Psychology, 43, p. 118. Copyright 1996 by the American Psychological Association. Reprinted with permission.

CULTURAL INFLUENCES

265

Americans. As this table shows, Asian Americans were not found to be

significantly lower in their levels of optimism. However, Asian Americans
were significantly more pessimistic than White Americans. Note, these
findings are consistent with Lee and Seligmans (1997) findings for optimistic
and pessimistic explanatory style (based on use of composite scores) between
Chinese and White Americans as mentioned previously.
As Table 12.1 also shows, Asian Americans used more problem avoidance and social withdrawal strategies to deal with stressful situations than
did White Americans. On the other hand, no significant cultural differences
were found in the use of the other coping strategies. In addition, compared
with White Americans, Asian Americans reported significantly more psychological symptoms as measured by the SCL-90-R. It is of interest that
no significant differences were found between the two cultural groups on
depressive and physical symptom reports as measured by the BDI and the
PILL, respectively. Thus, Asian Americans reported more general psychological problems, but not more depressive or physical problems, than did White
Americans. Taken together, these findings suggest that there are important
differences and similarities in the levels of optimism, pessimism, coping,
and adjustment between Asian and White Americans.

Are Optimism and Pessimism Important Predictors of Adjustment in

Asian and White Americans?

To address this question, Chang (1996a) conducted a series of regression analyses that sought to identify significant predictors of each adjustment
measure for each ethnic group. However, in addition to optimism and
pessimism scores, scores on each of the coping strategies (as well as on age
and gender) were included to determine if coping strategies would be found
to be more important predictors of adjustment than optimism and pessimism.
Results of these analyses as reported in Chang (1996a) for Asian Americans
and White Americans are presented in Tables 12.2 and 12.3, respectively.
As Table 12.2 shows, for Asian Americans, optimism, problem solving,
and pessimism were the only significant predictors found for depressive
symptoms. Lack of optimism was found to account for the largest amount
of the variance in depressive symptoms (13%), followed by problem solving,
and pessimism, respectively. The final regression model accounted for 20%
of the variance in depressive symptoms. For general psychological symptomatology, social withdrawal and optimism were the only significant predictors
found. Although social withdrawal was found to account for the largest
amount of the variance in psychological symptoms, lack of optimism was
found to account for almost as much additional variance (9%). The final
model accounted for 20% of the variance in psychological symptoms. Last, for

266

EDWARD C. CHANG

Table 12.2.
Stepwise Regression Analyses Showing Amount of Variance
Accounted for by Significant Predictors of Each Outcome Measure for
Asian Americans
Outcome Measure

BDI
Optimism
Problem solving
Pessimism
SCL-904
Social withdrawal
Optimism
PILL
Optimism
Problem solving

R2

F(1, 109)

-.23**
-.23**
.22*

.36
.41
.45

.13
.04
.03

15.94***
4.85*
5.07*

.30***
-.29***

.34
.45

.I 1

.09

14.04***
12.24***

-.32***
-.17*

.34
.39

.12
.03

14.53***
4.13*

Note. N = 111. BDI = Beck Depression Inventory; SCL-90-R = Symptom Checklist-90-Revised;

PILL = Pennebaker Inventory of Limbic Languidness. From "Cultural Differences in Optimism, Pessimism,
and Coping: Predictors of Subsequent Adjustment in Asian American and Caucasian American College
Students," by E. C. Chang, 1996, Journal of Counseling Psychology, 43,p. 118. Copyright 1996 by the
American Psychological Association. Reprinted with permission.
'p < .05. "p < .01. *"p < ,001.

physical symptoms, optimism and problem solving were the only significant
predictors found. Lack of optimism was found to account for the largest
amount of the variance in physical symptoms (12%), followed by lack of
problem solving. The final model accounted for 15% of the variance in
physical symptoms for Asian Americans.
For White Americans, the only significant predictors of depressive
symptoms found were pessimism, wishful thinking, optimism, and selfcriticism (see Table 12.3). Pessimism was found to account for the largest
amount of the variance in depressive symptoms (17%), followed by the
other three predictors in the order just listed. The final regression model
accounted for 3 1% of the variance in depressive symptoms. The only significant predictors of general psychological symptomatology found were pessimism, self-criticism, gender, and social withdrawal. Again, pessimism was
found to account for the largest amount of the variance in psychological
symptoms (20%), followed by the other three predictors. The final model
accounted for 35% of the variance in psychological symptoms. Last, for
physical symptoms, gender was the only significant predictor found. The
final model accounted for 9% of the variance in physical symptoms for
White Americans. Thus, lack of optimism appeared to be a strong predictor
of psychological and physical adjustment among Asian Americans, whereas
pessimism appeared to be a strong predictor of psychological adjustment
among White Americans. I will discuss the implications of these particular
findings for working with Asians who are distressed shortly.

CULTURAL INFLUENCES

267

Table 12.3
Stepwise Regression Analyses Showing Amount of Variance Accounted
for by Significant Predictors of Each Outcome Measure for
White Americans

.20*
.21*
-.20*
.17*

.42
50
53
.55

Outcome Measure
BDI
Pessimism
Wishful thinking
Optimism
Self-criticism
SCL-90-R
Pessimism
Self-criticism
Gender
Social withdrawal
PILL
Gender

.I7
.08
.03
.03

.30***
.22**
.20+
.17*

.45
.53
56
-58

.20
.08
.04
.03

.29*

.30

.09

22.73*+*
11.88***
4.04*
4.02*
27.37***
11.98***
5.51
4.2P
+

10.90**

Note. N = 111. BDI = Beck Depression Inventory; SCL-90-R = Symptom Checklist-90-Revised:PILL =

Pennebaker Inventory of Limbic Languidness. From Cultural Differences in Optimism, Pessimism, and
Coping: Predictors of Subsequent Adjustment in Asian American and Caucasian American College Students, by E. C. Chang. 1996, Journal of Counseling Psychology, 43, p. 119. Copyright 1996 by the American Psychological Association. Reprinted with permission.
p c .05.p c .01. p c ,001.

Are There Important Differences in the Pattern of Associations

Among the Study Variables Between Asian and White Americans?
It is noteworthy that Chang (1996a) found that the pattern of correlations between the various study measures for Asian Americans was different
from the one found for White Americans (see Chang, 1996a, Table 1).
Most striking was the difference in the direction of the associations between
pessimism and two of the engagement coping strategies (viz., problem solving
and expressing emotions) for Asian Americans compared with White Americans. Specifically, pessimism was found to be negatively associated with use
of problem solving and expressing emotions for White Americans. In sharp
contrast, pessimism was found to be associated positively with use of these
coping strategies for Asian Americans.
Given that pessimism and two of the engaged coping strategies were
found to be associated positively for Asian Americans, but not for White
Americans, Chang (1996a) conducted 2 x 2 (high vs. low pessimism x Asian
vs. White American) ANOVAs with each of the eight coping strategies from
the CSI as dependent variables to assess for significant interactions. Results
of these analyses indicated only one significant interaction after controlling
for number of comparisons that involved problem solving, F( 1, 218) =
13.33; p < .0013 (Figure 12.1). As the figure shows, for problem-solving
coping, highly pessimistic White Americans were found to use this coping
268

EDWARD C. CHANG

27

26
Problem

Solving

25

24

22

I
High

21Low

Pessimism

Figure 72.7. Problem solving as a function of pessimism and ethnicity for Asian
Americans and White Americans. From CulturalDifferences in Optimism, Pessimism,
and Coping: Predictors of Subsequent Adjustment in Asian American and Caucasian
American College Students,by E. C. Chang, 1996, Journal of Consulting Psychology,
43, p. 119. Copyright 1996 by the American Psychological Association. Reprinted
with permission.

strategy less frequently compared with less pessimistic White Americans.

In contrast, highly pessimistic Asian Americans were found to use problemsolving coping more frequently than did less pessimistic Asian Americans.
This finding is worth further consideration in relation to the heightened
pessimism of Asians as noted earlier in Lee and Seligmans (1997) findings.
Are Asians defensive pessimists?Although a general self-effacing tendency among Easterners may help account for their heightened pessimism
compared with Westerners (Heine & Lehman, 1995), this tendency does
CULTURAL 1NFLUENCES

269

not help us answer why heightened pessimism would be maintained among

Asians. One possibility may be that pessimism helps Asians incur positive
consequences. Specifically, by anticipating the worst, Asians may paradoxically gain an immediate sense of certainty or control in their uncertainty
about future outcomes (Chang, 1996b). This can be particularly advantageous for individuals who come from Eastern cultures and traditions that
ascribe great value and importance to maintaining interpersonal connectedness and harmony. For example, by anticipating important negative interpersonal outcomes (e.g., shaming ones family, offending ones superior), Asians
may be able to consider a greater range of preventive actions they can
engage in to ensure more positive or acceptable outcomes.
Yet, considering greater options is not likely to provide a sufficient
answer if changes in behavior do not occur. In that regard, Norem and
Cantors (1986) findings about defensive pessimism appear to be relevant.
These researchers found that some high-achieving individuals use a defensive pessimism strategy that involves negative outcome expectancies, coupled with the use of active and engaged coping behaviors. According to
Norem and Cantor (1986; see chap. 4,this volume), pessimism is not simply
what people have because they are pessimists (similar to the notion of
dispositional optimism or pessimism), but rather a reflection of what people
do (e.g., using a particular way of thinking about certain things) in relation
to a specific goal. Hence, consistent with Norem and Cantors findings,
Changs (1996a) findings suggest that Asian Americans also may use their
pessimism as a strategy to think about potential negative consequences and
as a means to motivate themselves toward proactive behavior (e.g., problem
solving). Thus, insofar as the link between pessimism and problem solving
results in more positive rather than negative consequences, this can help
explain why elevated pessimism scores are maintained among Asians.

USING BECKS (1976) COGNITIVE MODEL OF

PSYCHOLOGICAL DISTURBANCE TO WORK WITH
ASIAN AND WHITE AMERICANS
The aforementioned findings, particularly those from Changs ( 1996a)
study, can have important implications for considering culture-specific interventions in working with distressed Asian compared with White Americans.
Recall, Chang (1996a) found that (lack of) optimism was found to be one
of the best, if not the best, predictors of depressive symptoms, psychological
symptoms, and physical symptoms for Asian Americans, whereas pessimism
was found to be a strong predictor of depressive symptoms and psychological
symptoms for White Americans.
270

EDWARD C . CHANG

According to Becks ( 1976) cognitive model of psychological disturbance, pessimism is presumed to play an important role in the development
of depression. Changs (1996a) findings for White Americans are consistent
with this view. A key intervention used in changing how pessimistic thoughts
lead to depressive symptoms is to refute or challenge evidence supporting
such negative thoughts (Beck, Rush, Shaw, & Emery, 1979). For instance,
after a clients negative beliefs (e.g., I am a failure) have been translated
to testable hypotheses (e.g., I cant do anything right), the practitioner
can then help the distressed client focus on disconfirming these maladaptive
beliefs by assisting in obtaining or introducing sufficient counterevidence
(e.g., observing that the client successfully showed up for the appointment;
Beck et al., 1979).
However, given that Chang (1996a) found lack of optimism, but not
pessimism, predicted subsequent depressive symptoms in Asian Americans,
a culture-specific intervention consistent with Becks cognitive framework
may focus distinctly on increasing optimistic thoughts in distressed Asian
American clients rather than on decreasing their pessimistic thoughts. As
noted by Chang (1996a), by decreasing pessimism in Asian Americans,
one could conceivably take away a major source of motivation that is related
to adaptive and engaged coping behaviors such as problem solving (p. 121).
For instance, in working with Asian American clients who are dysphoric,
the practitioner can help the client first identify untapped positive beliefs
by referring to or obtaining supportive evidence (e.g., noting that the client
has always shown up on time for his or her appointment). In sum, Changs
( 1996a) findings suggest that in working with distressed Asian Americans,
it may be particularly valuable to use interventions that focus on increasing their level of optimism, whereas for White Americans, it may be
more valuable to use interventions that focus on decreasing their level
of pessimism.

ASSESSING PSYCHOLOGICAL DISTURBANCE IN ASIAN

AMERICANS: DISTINGUISHING PESSIMISM FROM DEPRESSION
Moreover, findings from Changs ( 1996a) study also indicate a potential
need for examining the clinical significance of various combinations of high
and low optimism and pessimism within individual clients. That is, because
Asians and Asian Americans may normally express elevated levels of
pessimism (Lee & Seligman, 1997), as found in Changs (1996a) study to
be positively correlated with expressions of depressive, psychological, and
physical symptoms, it would be important to distinguish between subgroups
who do and do not express a concomitant level of optimism. For example,
Asian Americans who express high levels of pessimism relative to White
CULTURAL INFLUENCES

271

Americans, but little optimism, may be particularly vulnerable to psychological disturbance and physical illness. In contrast, it may be that some Asians,
despite their robust pessimism, express a balance of negative and positive
thoughts (yin and yang) that functions adaptively for them. Therefore, in
assessing for pessimism, it may be critical to also assess for expressions of
optimism. In fact, by neglecting to assess for expressions of optimism, mental
health professionals (especially those with little experience in working with
Asians) may inaccurately construe Asian Americans expression of pessimism
as a sign of a depressive disorder.
Because pessimism is an important indicator of clinical depression
(i.e., in expressions of hopelessness and helplessness; American Psychiatric
Association, Diagnostic and Statistical Manual of Mental Disorders; DSM-IV,
1994), clinicians and counselors may need to exercise great caution when
assessing Asian Americans to differentiate those who have depressive symptoms from those with nondepressive, but pessimistic thoughts. Findings from
a recent set of studies conducted by Helweg-Larsen, Sadeghian, and Webb
(1999) that looked at the stigma of being pessimistic strongly underscore
this important and practical point.
In their first study, Helweg-Larsen et al. (1999) presented students
with information describing an optimistic target (i.e., a hypothetical student
who expressed that his health risks were lower than those of other students);
a pessimistic target (a hypothetical student who expressed that his health
risks were greater than those of other students); and a neutral target (a
hypothetical student who expressed that his health risks were about the
same as those of other students). After this presentation, students were then
asked questions regarding how sad or happy, depressed or not depressed,
and hopeless or not hopeless each of the targets appeared. In addition,
students were also asked a set of questions that tapped their tendency to
socially accept or reject the target (e.g., fun to hang out with the person).
It is of interest to note that results from this study show that students were
more likely to perceive the pessimistic target as being more sad, depressed,
and hopeless, and less likely to be socially accepted. A second study was
conducted to determine if these initial findings may have been due to
peoples lay perception that pessimists take more health risks, and thus
account for pessimists greater social rejection by others. It was surprising
to note that results of the second study indicated the exact opposite. That
is, the optimistic target student was perceived as more of a risk taker than
was the pessimistic target student. Nevertheless, consistent with findings
from the first study, social rejection was strongest for the pessimistic target
student, who was again perceived as more sad, hopeless, and depressed.
In their final study, Helweg-Larsen et al. (1999) again presented students with pessimistic, optimistic, and neutral targets, but this time, students

272

EDWARD C. CHANG

heard a tape recording of a hypothetical student who responded pessimistically, optimistically, or in a neutral manner to an interviewer asking about
his chances of various health risks. However, before students had a chance
to listen to these recordings, information regarding the mental state of the
target student was manipulated by the researchers. Specifically, students
were either given no information about the mental health of the target
student, or given the information that the mental health of the target
student was good (e.g., person was happy and neither depressed nor hopeless). Hence, for example, prior to hearing the target student express his
pessimism about health risks, a student was told that the target student was
mentally healthy. What these investigators found confirmed their earlier
findings that people tend to associate pessimism with depression. Specifically,
when no information was given about the target student, student-participants
again perceived the pessimistic target as more sad, hopeless, and depressed,
and in turn, were again more likely to reject him than to reject the optimistic
target student. However, when information was given indicating that the
target student was not dysphoric, students showed no difference in their
ratings between optimistic and pessimistic targets on sadness, hopelessness,
and depression. Moreover, students also showed no difference in their ratings
between these target students on measures of social rejection.
In sum, Helweg-Larsen et al.s research not only identified important
social costs of being pessimistic, but also identified a critical source of peoples
prejudice and social rejection of pessimists, namely, the presumption that
pessimists are depressed individuals. Therefore, given that Asian Americans
are normally more pessimistic than White Americans (Chang, 1996a; Lee
& Seligman, 1997), Asians may be at greater risk of being diagnosed for
depressive disorders. That is, practitioners lacking experience or cultural
competency in working with Asians may be quick to judge an Asian Americans expression of heightened pessimism as a telltale sign of an underlying
depression or enduring dysthymia. However, by assessing for expressions
of both optimism and pessimism in evaluating Asians, as noted earlier,
practitioners may be able to effectively disentangle Asian Americans normal expression of pessimism from signs reflecting clinical depression or
dysthymia (e.g., where expressions of optimism may be absent). In addition,
given the subjective nature of social evaluations, Helweg-Larsen et al.s
(1999) findings also reinforce the value of using more objective (i.e., behavioral) techniques in the assessment of depression or dysthymia, which may
be particularly useful when interacting with pessimistic Asians. No doubt,
given that the studies conducted by Helweg-Larsen et al. (1999) mostly
involved White participants and no Asians (M. Helweg-Larsen, personal
communication, December 26, 1999), it would be interesting to see how
Asians perceive expressions of optimism and pessimism in others.

CULTURAL INFLUENCES

273

CHALLENGES AND FUTURE DIRECTIONS FOR RESEARCH

Asian Americans as More Than a Single Ethnic Category
Just as it would be incorrect to say that all White Americans look or
act the same, it is incorrect for researchers and mental health professionals
to assume that Asian Americans are a homogenous population. Although
genetically influenced features such as skin color may indicate that Asian
Americans are more racially similar to each other compared with other
racial groups, this should not be taken to suggest that they are all the same.
In fact, when one considers factors such as language and cultural practices,
Asian Americans begin to look and sound quite different (e.g., Sue &
Morishima, 1982; Uba, 1994). They represent a very diverse population.
According to Uba (1994), one can enumerate as many as 25 distinct Asian
American ethnic groups living in the United States. Each of these ethnic
groups differs across a number of important dimensions. For example, the
history of different Asian American groups living in the United States varies
considerably. Whereas Chinese and Japanese Americans are likely to have
been members of American society for several decades (Daniels, 1988),
Korean and Vietnamese Americans are likely to present as more recent
immigrants (Takaki, 1989).
Nonetheless, although they share similar immigration histories, Chinese and Japanese Americans are not alike. As a tragic historical example,
consider the treatments of Chinese and Japanese Americans in the United
States during WWII. Unlike Chinese Americans, about 120,000 Japanese
Americans, many of whom were US citizens, were transported to internment
camps. This prejudicial treatment experienced by many Nisei (second generation) Japanese Americans has profoundly influenced Sansei (third generation) Japanese Americans. According to Nagata ( 1991), third-generation
Japanese Americans not only struggle with the same problems their parents
did, but also they struggle to believe and trust in American society. For
such reasons, some researchers have argued that the term Asian American
itself can be misleading and potentially harmful if such important differences
between and within the different Asian ethnic groups are overlooked (Abe
& Zane, 1990; Phinney, 1996). Therefore, it would be important to consider
how optimism and pessimism may differ and relate to other important
variables across the different Asian American ethnic groups.
Optimism and Pessimism in Predicting Adjustment in Asian Americans
Although optimism and pessimism have been found to be important
predictors of adjustment, such findings are often based on studies that rarely
compare the predictive power of these variables to other robust predictors

274

EDWARD C. CHANG

of adjustment. For example, studies have shown that positive and negative
affectivity are also strongly associated with psychological adjustment (e.g.,
Watson, Clark, & Carey, 1988; Watson, Clark, & Tellegen, 1988). Furthermore, Marshall, Wortman, Kusulas, Hervig, and Vickers (1992) have found
that positive and negative affectivity map significantly onto optimism and
pessimism, respectively. Therefore, it is not clear how important optimism
and pessimism are in predicting adjustment beyond positive and negative
affectivity, and in predicting adjustment between Asians and White people.
T o address some of these issues, Chang (1999) examined the relative
contributions of optimism and pessimism above those made by positive and
negative affectivity in predicting scores on several commonly used measures
of psychological disturbance in 92 Asian and 252 White Americans. After
controlling for the variance accounted for by positive and negative affectivity
in each of the outcomes (viz., measures assessing for trait anxiety, depression,
hopelessness, psychological distress, and psychological symptoms), overall
for White Americans, optimism and pessimism continued to account for a
significant amount of additional variance in each outcome measure. In
contrast, for Asian Americans, optimism and pessimism accounted for a
significant amount of additional variance in only two of the outcomes (viz.,
trait anxiety and hopelessness) beyond what was accounted for by positive
and negative affectivity. This is not to say that optimism and pessimism
are not important variables associated with adjustment for Asians, but rather,
that the influence of these variables on adjustment for Asians may be
mediated by other important variables, such as mood.
For still another important predictor of psychological outcomes, consider perfectionism (Chang, 2000; Chang & Rand, 2000). Similar to findings
for pessimism, findings for expression of perfectionism are greater for Asian
Americans than for White Americans (Chang, 1998a; Kawamura, Frost, 6r
Harmatz, 1999). Yet, we know little about how optimism and pessimism
compare with perfectionism in predicting adjustment for Asians or for White
people. Similarly, insofar as coping has been conceptualized and measured
in many different ways (Chang, in press), there is still no definitive study
of optimism, pessimism, and coping as predictors of adjustment in Asians
and White people. No doubt, more studies are needed to compare optimism
and pessimism to other robust predictors of adjustment.
Outcome Measures for Studying Asian People
Western researchers are becoming increasingly aware that when studying the culturally different, they must not only be careful to not impose
their cultural values and biases in understanding non-Westerners, but also
that they must appreciate those processes and outcomes that are most meaningful to such groups. It is noteworthy that some culture-bound disturbances
CULTURAL INFLUENCES

275

reflecting symptoms of both anxiety and depression are particular to Asians.

For example, shenjing shuairuo (or neurasthenia) refers to a condition observed among Asian people, including Asian Americans, where the expressed
symptoms can be very similar to those typically presented in mood and
anxiety disorders (DSM-IV, 1994). However, there has been little study of
neurasthenia in Asian Americans (e.g., Lin, 1989;Yamamoto, 1992). Hence,
we know very little about how optimism and pessimism predict culturally
relevant outcomes such as neurasthenia among Asian Americans.
Also, because researchers have commonly focused on maladaptive
outcomes, we know little about how variables such as optimism and pessimism relate to positive outcomes. Yet, what are the positive outcomes that
should be considered? For example, Ryff (1989; Ryff & Keyes, 1995) has
offered a useful conceptualization of subjective well-being composed of six
relatively distinct dimensions. These dimensions are autonomy, environmental mastery, personal growth, positive relations with others, purpose in
life, and self-acceptance. However, some of these dimensions tend to reflect
attributes that are likely to be highly regarded by most Westerners, but not
by most Easterners. Therefore, the consideration of culturally appropriate
and meaningful outcome measures (both positive and negative) should be
an important priority for researchers and practitioners who study or work
with Asian populations (Okazaki, 1998).

CONCLUSION
Although the mechanisms by which optimism and pessimism develop
and are maintained in different cultural groups have yet to be clearly understood, the present chapter makes it clear that any model of optimism and
pessimism that ignores the influence of culture is likely to be incomplete.
For example, E.-Y. Kim (1993) has noted that some Asian Americans (viz.,
Korean Americans) express effort optimism, a perception that with direct
effort, anything is possible. Hence, despite findings indicating comparable
elevations on expressions of optimism between Asian and White Americans
(Chang, 1996a; Lee &Seligman, 1997), important differences may be uncovered when specific aspects of optimism are considered (e.g., optimism for
engaging in successful behavior). Clearly, more studies are needed to address
the many questions that remain to be considered in studying optimism and
pessimism in groups of Easterners and Westerners. As I hope this chapter
has impressed on the reader, the study of cultural influences on variables
such as optimism and pessimism can be both fascinating and arduous. It
will require researchers, scholars, and practitioners alike to take on the
challenge of considering optimism and pessimism as much a function of the
individual as of the persons social and cultural environment.

276

EDWARD C. CHANG

REFERENCES
Abe, J. S., & Zane, N. W. S. (1990). Psychological maladjustment among Asian
and White American college students: Controlling for confounds. Journal of
Counseling Psychology, 37, 437-444.
American Psychiatric Association. ( 1994). Diagnostic and statistical manual of mental
disorders (4th ed.). Washington, DC:Author.
Beck, A. T. (1976). Cognitive therapy and the emotional disorders. New York: Intemational Universities Press.
Beck, A. T., Rush, A. J., Shaw, B. F., & Emery, G. (1979). Cognitive therapy of
depression: A treatment manual. New York: Guilford Press.
Beck, A. T., Ward, C. H., Mendelson, M., Mock, L., & Erbaugh, J. (1961). An
inventory for measuring depression. Archives of General Psychiatry, 4 , 561-5 7 1.
Brislin, R. (1993). Understanding cultures influence on behvior. Fort Worth, TX:
Harcourt Brace and Jovanovich.
Chang, E. C. (1996a). Cultural differences in optimism, pessimism, and coping:
Predictors of subsequent adjustment in Asian American and Caucasian American college students. Journal of Counseling Psychology, 43, 113-123.
Chang, E. C. (199613). Evidence for the cultural specificity of pessimism in Asians
versus Caucasians: A test of a general negativity hypothesis. Personality and
Individual Differences, 21, 819-822.
Chang, E. C. (1998a). Cultural differences, perfectionism, and suicidal risk in a
college population: Does social problem solving still matter? Cognitive Therapy
and Research, 2 2 , 237-254.
Chang, E. C. (1998b). Distinguishing between optimism and pessimism: A second
look at the optimism neuroticism hypothesis. In R. R. Hoffman, M. F.
Sherrik, and J. S. Warm (Eds.), Viewing psychology as a whole: The integrative
science of William N. Dernber (pp. 415-432). Washington, DC: American
Psychological Association.
Chang, E. C. ( 1999).Cultural differencesin psychological distress in Asian and Caucasian
American college students: Examining the role of cognitive and affective concomitants.
Unpublished manuscript, University of Michigan, Ann Arbor.
Chang, E. C. (2000).Perfectionism as a predictor of positive and negative psycholog
ical outc,omes: Examining a mediational model in younger and older adults.
Journal of Counseling Psychology, 47, 18-26.
Chang, E. C. (in press). A look at the coping strategies and styles of Asian Americans: Similar and different? In C. R. Snyder (Ed.), Coping and copers: Aduptive
processes and people. New York: Oxford University Press.
Chang, E. C., DZurilla, T. J., & Maydeu-Olivares, A. (1994). Assessing the dimensionality of optimism and pessimism using a multimeasure approach. Cognitive
Therapy and Research, 18, 143-160.
Chang, E. C., Maydeu-Olivares, A., & DZurilla, T. J. (1997). Optimism and
pessimism as partially independent constructs: Relations to positive and negaCULTURAL INFLUENCES

277

tive affectivity and psychological well-being. Personality and Individual Differ-

ences, 23, 433-440.

Chang, E. C., & Rand, K. L. (2000). Perfectionism as a predictor of subsequent

adjustment: Evidence for a specific diathesis-stress mechanism among college
students. Journal of Counseling Psychology, 47, 129-137.
Daniels, R. (1988). Asian Americans: Chinese and Japanese in the United States since
1850. Seattle: University of Washington Press.
Derogatis, L. R. (1983). The SCL-90 R: Administration, scoring, andprocedures munual
II. Baltimore, MD: Clinical Psychometric Research.
Doi, T. (1973). The anatomy of dependence. Tokyo: Kodansha.
Folkman, S., & Lazarus, R. S. (1980). An analysis of coping in a middle-aged
community sample. Journal of Health and Social Behavior, 21, 219-239.
Greenwald, A. G. (1980). The totalitarian ego: Fabrication and revision of personal
history. American Psychologist, 35, 603-618.
Heine, S. J., & Lehman, D. R. (1995). Cultural variation in unrealistic optimism:
Does the West feel more invulnerable than the East?Journal of Personality and
Social Psychobgy, 68, 595-607.
Heine, S. J., Lehman, D. R., Markus, H. R., & Kitayama, S. (1999). Is there a
universal need for positive self-regard?Psychological Review, 106, 766-794.
Helweg-Larsen, M., Sadeghian, P., & Webb, M. S. (1999). The stigma of being
pessimistically biased. Unpublished manuscript, Transylvania University, Lexington, KY.
Hofstede, G. ( 1980). Cultures consequences. Beverly Hills, CA: Sage Publications.
Jahoda, G. (1982). Psychology and anthropology. London: Academic Press.
Kawamura, K. Y., Frost, R. O., & Harmatz, M. G. (1999). The relationship between
parental harshness and perfectionism: A study of Asian-American and Caucasian
college students. Unpublished manuscript, University of Massachusetts at Amherst.
Kim, E.-Y. ( 1993). Career choice among second-generation Korean-Americans:
Reflections of a cultural model of success. Anthropology and Education Quarterly,
24, 224-248.
Kim, U., Triandis, H. C., Kagitcibasi, C., Choi, S. C., & Yoon, G. (Eds.). (1994).
Individualism and collectivism: Theory, method, and applications. Newbury Park,
CA: Sage.
Lazarus, R. S., & Folkman, S. (1984). Stress, appraisal, and coping. New York:
Springer.
Lee, Y.-T., & Seligman, M. E. P. (1997). Are Americans more optimistic than the
Chinese? Personality and Social Psychology Bulletin, 23, 32-40.
Lin, T.-Y. (1989). Neurasthenia revisited: Its place in modern psychiatry. Culture,
Medicine and Psychiatry, 13, 105-129.
Markus, H. R., & Kitayama, S. (1991). Culture and the self: Implications for
cognition, emotion, and motivation. Psychological Review, 98, 224-253.

278

EDWARD C . CHANG

Marsella, A., DeVos, G., & Hsu, F. L. K. (1985). Culture and self. London: Tavistock.
Marshall, G. N., Wortman, C. B., Kusulas, J. W., Hervig, L. K., & Vickers, R. R.,
Jr. ( 1992). Distinguishing optimism from pessimism: Relations to fundamental
dimensions of mood and personality.Journal of Personality and Social Psychology,
62, 1067-1074.
Mizuno, M., & Yamaguchi, S. (1997, August). Reframing Japanese pessimism: I'm
more typical h n the average person. Paper presented at the 2nd annual meeting
of the Asian Association of Social Psychology, Kyoto, Japan.
Nagata, D. ( 1991). Transgenerational impact of the Japanese American internment:
Clinical issues in working with children of former internees. Psychotherapy,
28, 121-128.
Norem, J. K., & Cantor, N. (1986). Defensive pessimism: Harnessing anxiety as
motivation. Journal of Personality and Social Psychology, 51, 1208-1217.
Okazaki, S. ( 1998). Psychological assessment of Asian Americans: Research agenda
for cultural competency. Journal of Personality Assessment, 70, 54-70.
Pennebaker, J. W. (1982). 7% psychology of physical symptoms. New York:
Springer-Verlag.
Peterson, C., Semmel, A., von Baeyer, C., Abramson, L. Y., Metalsky, G. I., &
Seligman, M. E. P. (1982). The Attributional Style Questionnaire. Cognitive
Therapy and Research, 6, 287-299.
Phinney, J. (1996). When we talk about American ethnic groups, what do we
mean? American Psychologist, 51, 918-927.
Roland, A. (1988). In search of self in India and Japan: Toward a cross-cultural
psychology. Princeton, NJ: Princeton University Press.
Ryff, C. D. (1989). Happiness is everything, or is it? Explorations on the meaning
of psychological well-being. Journal of Personality and Social Psychology, 57,
1069-1081.
Ryff, C. D., & Keyes, C. L. M. (1995). The structure of psychological well-being
revisited. Journal of Personality and Social Psychology, 69, 719-727.
Scheier, M. F., & Carver, C. S. (1985). Optimism, coping, and health: Assessment
and implications of generalized outcome expectancies. Health Psychology, 4 ,
2 19-247.
Sue, S., & Morishima, J. (1982). The mental health ofAsian Americans. San Francisco:
Jossey-Bass.
Takaki, R. (1989). Strangers from a different shore: A history of Asian Americans.
Boston: Little Brown.
Tobin, L. D., Holroyd, K. A., Reynolds, R. V., & Wigal, J. K. (1989). The hierarchical factor structure of the Coping Strategies Inventory. Cognitive Therapy and
Research, 13, 343-361.
Triandis, H. C. (1995). Individualism and colkctivism. Boulder, CO: Westview Press.
Uba, L. (1994). Asian Americans: Personality patterns, identity, and mental health.
New York: Guilford Press.
CULTURAL INFLUENCES

279

Waterman, A. S. (1984). The psychology of individualism. New York: Praeger.

Watson, D., Clark, L. A., & Carey, G. (1988). Positive and negative affectivity
and their relation to anxiety and depressive disorders. Journal of Abnormal
Psychology, 97, 346-353.
Watson, D., Clark, L. A., & Tellegen, A. (1988). Development and validation of
brief measures of positive and negative affect: The PANAS scales. Journal of
Personality and Social Psychology, 54, 1063-1070.
Weinstein, N. D. (1980). Unrealistic optimism about future life events. Journal of
Personality and Social Psychology, 39, 806-820.
Weinstein, N. D. (1984). Why it wont happen to me: Perceptions of risk factors
and susceptibility. Health Psychology, 3 , 43 1-457.
Weisz, J. R., Rothbaum, R. M., & Blackburn, T. C. (1984). Standing out and
standing in: The psychology of control in America and Japan. American Psychologist, 39, 955-969.
Yamamoto, J. ( 1992). Psychiatric diagnoses and neurasthenia. Psychiatric Annals,
22, 171-172.
Yee, A. H. (1992). Asians as stereotypes and students: Misperceptions that persist.
Educational Psychology Review, 4, 95-132.

280

EDWARD C . CHANG

13
THE OPTIMISM-PESSIMISM
INSTRUMENT: PERSONAL AND
SOCIAL CORRELATES
WILLIAM N.DEMBER

My interest in the constructs of optimism and pessimism originated

in a search for correlates of individual differences in susceptibility to the
Pollyanna pnncipk, the broad tendency for people to give priority to the
positive in perception, memory, and language usage (see Dember, Martin,
Hummer, Howe, & Melton, 1989; Dember & Penwell, 1980; Matlin &
Stang, 1978). That tendency, manifested in group means on a variety of
simple tasks, although widespread, is not universal. Some people conform
strongly to the Pollyanna tendency, while a few show what might be called
an anti-Pollyanna bias. For example, in a classroom demonstration, I often
ask undergraduate students to write a list of 12 pairs of antonyms that might
be used to describe a person, physically or psychologically (e.g., slim-fat;
happy-sad). They then are asked to circle the member of each pair of
adjectives that is the more complimentary or positive. The Pollyanna tendency is revealed in the greater than chance number of times that the first
28 1

member of each pair (the one on the left, if written, or the first adjective
uttered, if spoken) is assessed as the more positive, as would be the case
with the two examples I have just given. In such demonstrations, the mean
frequency comes out to be about 9, where chance is 6. The first time I ran
that demonstration, I asked the students to score their own protocols and
then announce the total, which could range from 0 to 12. I put the frequency
distribution on the board as the numbers were called out. Sure enough,
the scores were biased toward the high end (conformity to the Pollyanna
principle), but there were some scores below 6, and one such low scoring
participant asked what that might mean. I mumbled something to the effect
that whatever score any particular individual achieved probably had no
significance; it was the group tendency that mattered. I am not convinced
that the student was satisfied with that brush-off, and I certainly was not
ready to dismiss the notion that scoring low on a Pollyanna task might
say something about the person. My initial thought was that low scoring
participants might have strongly pessimistic personalities, just as those with
very high scores might be unrealistically optimistic.
My desire to test that hypothesis led me on a search for a good measure
of optimism and pessimism, assuming along with most everyone else at the
time that optimism and pessimism lay at opposite poles of a single dimension.
These events occurred in the late 1970s, and either out of a lack of diligence,
or because there were no satisfactory measures yet in the literature, my
search was unsuccessful. So I decided to create a questionnaire, the optimism-pessimism (OP) instrument for measuring optimism and pessimism.
This chapter offers a progress report on that effort. It is not intended as a
general review of the optimism-pessimism literature. Of the results that
research with the OP instrument yielded, the most important, I believe, is
that optimism and pessimism are partially independent constructs: They are
only moderately negatively correlated (usually in the low-to-mid -.50s,
although on occasion in the low -.60s) and in many, though by no means
all, instances are differentially related to other variables.

CREATION OF THE OP INSTRUMENT

In developing the OP instrument, I followed the familiar strategy of
creating items in the form of brief statements that the respondent could
either endorse or reject by using a four-alternative, Likert-type format
(strongly agree, agree, disagree, strongly disagree). To balance for response
biases, such as yea-saying or nay-saying, I decided to have an equal number
of items worded such that endorsement implied either an optimistic or a
pessimistic orientation. It was my initial intent to combine the responses
on the two subsets of items, reverse-scoring one set and adding the item
282

WILLIAM N.DEMBER

values to get a total score. In addition to the relevant items, which had
been prescreened for face validity by a panel of advanced graduate students,
there were also to be a number of filler statements, hopefully to mask the
intent of the instrument. I began the task of writing items with the aid of an
undergraduate student. Part way through, a graduate student, Mary Newman
(later Mary Hummer) took up the project for her masters thesis, where the
focus was to be on basic psychometric issues such as item-test correlations
and internal reliability. Of secondary concern at that point was the issue
of validity.
One outcome of the initial analyses of the Newman (1983) dataobtained from 2 16 undergraduate students participating to fulfill a course
requirement-was the deletion of 2 optimistic and 2 pessimistic items from
initial sets of 20. As a result 18 items of each type were left along with the
original 20 filler items.
With regard to the second, more important outcome, although it had
been our intent to create a single OP scale, the data were nevertheless
analyzed separately by item type, and it became apparent that we had two
scales, that is, an 18-item 0 (optimism) scale and an 18-item I (pessimism)
scale, which were only moderately correlated with one another ( r = -.52
in this first study). Moreover, the value of Cronbachs alpha for each of the
two scales was higher (a= .84 for optimism and .86 for pessimism) than
the canonical correlation (.74) between the two scales (Cronbach, 1951).
That general finding has been replicated in numerous follow-up studies,
with the value of r typically running in the low -SOs to the low -.60s.
Even more dramatic is the finding that 0 and P scores were uncorrelated
in a sample of Ethiopian international students, whereas that correlation
for a sample of comparable noneEthiopian Africans was -.52 (Bantirgu,
1995).Test-retest correlations over a 2-week interval were found to be .75
for optimism and .84 for pessimism (Dember & Brooks, 1989). Neither in
those, nor in follow-up studies have we found any differences between men
and women on level of 0 or P, on the magnitude of the correlation between
them, or in correlations with other variables.
It is primarily on the basis of those modest correlations between optimism and pessimism that we routinely compute both an 0 score and a I
score for each participant in our research and typically do not combine
those scores into a composite optimism-pessimism value. However, in a
recently completed experiment on the role of optimism and pessimism in
vigilance performance (Helton, Dember, & Warm, 1999), it proved useful
to create two extreme groups, High 0-Low P and Low 0-High P, with
High and Low defined as above or below the group median. When so
categorized, pessimistic participants showed a steeper decline over time in
performance efficiency than did optimistic participants. Fischer & Leitenberg
(1986) have also reported the non-bipolarity of optimism and pessimism
THE OPTIMISM-PESSIMISM INSTRUMENT

283

based on an instrument they developed for use with children, suggesting

that the surprisingly moderate correlation between 0 and P is not peculiar
to the OP instrument.
Note that the content of the 0 and P items is quite varied. It was our
intent not to restrict the substance of the statements to a single category
(e.g., expectations about ones own future prospects or assessments of the
present or future state of the world in general), but rather to cover a wide
variety of domains. In that way, we hoped to have an instrument which
would be related to a broad range of other measures, even at the cost of
not being highly predictive of any single measure. For example, some items
in the OP instrument are personally relevant, and some refer to universal,
timeless truths (e.g., April showers bring May flowers). That heterogeneity
of content shows up in factor analyses (e.g., Chang, DZurilla, & MaydeuOlivares, 1994) that identify more than the intended two factors, optimism
and pessimism, but it does not seem seriously to vitiate internal reliability.

CORRELATIONS WITH PERSONALITY MEASURES

As a first pass at examining construct validity of the 0 and P scales,
Newman did obtain data from her participants on several other variables,
including a rating on a 7-point scale in response to the question: In general,
how optimistic or pessimistic would you say you are? A rating of 1 was
associated with extreme optimism, and a rating of 7 with extreme pessimism.
(In hindsight, we should have asked two questions: HOWoptimistic would
you say you are and how pessimistic would you say you are?). Those selfratings were correlated with 0 at r = -.55 and with P at .53, about the
same level as the correlation between 0 and P. Clearly, the scales and the
self-ratings were tapping a common construct, but the self-ratings could not
simply substitute for the scales. Those results lend support to the construct
validity of the 0 and P scales, and at the same time indicate that they
have potential use beyond that of the self-ratings. Of course, as single-item
measures, the self-ratings are likely to have less than optimal reliability; it
would be interesting to check out their test-retest reliability, as well as to
ask the two questions alluded to above rather than just the one question
we did ask, which assumes a unidimensional continuum running from highly
optimistic at one end to highly pessimistic at the other. Asking both questions would also permit calculation of the correlation between responses to
the optimism and the pessimism questions, just as we correlate the values
on the 0 and P scales.
Data on additional single-item questions asked of Newmans participants have been reported in Dember et al. (1989). Especially interesting
are responses to two questions designed to tap the extent to which a partici-

284

WILLIAM N.DEMBER

pant was committed to a religious or a political ideology. The expectation

was that such commitment would be associated with high optimism or low
pessimism on the grounds that the solace afforded by an ideological system
fosters an upbeat attitude by helping the individual find meaning and direction in a complex, confusing, even chaotic world (see Dember, 1974,1989).
Moreover, political ideologies, such as communism and socialism, often
promise a rosy future on this earth once those who oppress the downtrodden
are removed, forcibly if necessary, from positions of power. Similarly, religious
ideologies typically point to an afterlife, when, for true believers, the woes
of our corporeal existence will be replaced by heavenly bliss. The data
showed that for both politics and religion, those participants who answered
yes to the question about commitment had significantly higher 0 scores
than those who said no, but the groups did not differ on P scores, an
instance in which the 0 and P scales are differentially related to a criterion
measure. Those findings with regard to religion were replicated by Dember
and Brooks (1989), who also reported impressive correlations (high SOs,
low .60s) between both 0 and P and happiness. Similar findings with
regard to religious fundamentalism have also been reported by Sethi and
Seligman (1993).
Because the results come from a correlational design, it is possible, of
course, that optimistic people are attracted to an ideological system, rather
than that adherents derive their elevated optimism from their commitment
to the ideology. My preference at this point is for the latter interpretation.
Indeed, I would expect that those people who are already highly optimistic
would not need to seek comfort in a ready-made ideology; if that were
indeed so, the difference in optimism scores that Newman found is all the
more impressive, because the noncommitted group would contain a subset
of people high in optimism independent of their political or religious affiliations. Responses to another single-item question about the imminent likelihood of nuclear war revealed a similar pattern, except in this instance only
pessimism was predictive. That is, those people who believed that a nuclear
war was likely by the year 2000 had significantly higher P scores than those
who did not, but the two groups did not differ on 0 scores.
Participants were also asked to predict their grade-point average (GPA)
for the coming quarter. In this case, both 0 and P were predictive, in the
expected direction, with values of r at .21 and -.28, respectively. incidentally,
those low values of r, statistically significant by virtue of the large sample
size, are fairly typical of the magnitude of correlations with 0 and P, as
well as with other measures of optimism and pessimism, and with many of
the variables investigated thus far. Moreover, as suggested by the last result,
in both the Newman study and in subsequent research, the P scale tends
to be more strongly related to other variables than the 0 scale, perhaps
because P scores typically have a greater variance than 0 scores. At this
THE OPTIMISM-PESSIMISM INSTRUMENT

285

point, there are neither sufficient empirical data nor sufficiently well-articulated theories to enable researchers to anticipate the kinds of variables
(dispositional or behavioral) that are better predicted from one scale or the
other, or to anticipate when the two will be equivalent in the strength of
their relations with other variables.
In the remainder of this section, we summarize much of the evidence
we do have on various personality correlates of 0 and P, including their
relations with two other popular instruments for measuring optimism and
pessimism, the Life Orientation Test (LOT; Scheier & Carver, 1985) and
the Attributional Style Questionnaire (ASQ and its variant, the Expanded
Attributional Style Questionnaire or EASQ; Peterson et al., 1982; Peterson
& Villanova, 1988). (See Introduction, this volume for a brief discussion
of the LOT, the ASQ, and other related instruments.) We also present
some data on the relative predictive ability of the three instruments (OP,
LOT, & EASQ) with regard to coping style and depression. These comparative data offer some encouragement that the OP instrument provides value
added in this increasingly crowded domain. The article by Dember et al.
(1989) alluded to above reports the results of both the Newman thesis and
a subsequent thesis by Stephanie Martin (1986). The latter was concerned
primarily with personality correlates of the 0 and P scales, in particular
anxiety and defense (see Dember et al., 1989, for complete details). Here
1 want first to note that Martins data, based on 228 participants and drawn,
as were Newmans, from the Introductory Psychology pool of participants
at the University of Cincinnati, very closely replicate the values Newman
obtained for the level of optimism and pessimism, the correlation between
the two scales, and their internal reliability. The consistency between the
statistical properties of the two data sets is especially noteworthy given that
the studies were run about 3 years apart.
Second, in addition to the OP, participants in the Martin study responded to the Repression-Sensitization Scale (RS; Byrne, Barry, & Nelson,
1963) and to the Defense Mechanisms Inventory (DMI; Ihilevich & Gleser,
1986). The latter is an objectively scorable instrument for assessing the
extent to which the respondent uses each of five categories of defense
mechanism, turning against self (TAS, an intrapunitive defense); reversal
(REV, a repressive defense); principalization (PRN, an intellectualizing
defense); turning against other (TAO, an aggressive defense); and projection
(PRO, perceiving ones own unconscious motives and conflicts in others).
The RS scale has embedded within it a measure of anxiety, the Welsh A
(Welsh, 1956), and the correlation between RS scores and anxiety scores
is very high (see, e.g., Watson & Clark, 1984). Hence, there is good reason
to consider the RS a measure of anxiety rather than a measure, as originally
conceived, of repressive tendencies or of their opposite, sensitizing tendencies.
286

WlLLlAM N.DEMBER

With regard to RS scores and Welsh A scores, as expected, 0 was

negatively correlated with both (r = -.42 and -.34, respectively), while P
was correlated positively, and more strongly with both (r = .65 and -60,
respectively).' It was further expected, and found, that 0 scores would
correlate positively with REV (r = .28) and with PRN (r = .19). In addition,
0 scores were negatively related to TAS (r = -.27) and with PRO (r =
-.19). P scores, as expected, correlated positively with TAS (r = .29) and
with PRO (r = .21) and negatively with PRN and REV (in both cases, r =
-.32). In three instances, the magnitude of the correlation with P was
significantly greater than that with 0, that is, with regard to PRN, RS, and
A, although the latter two scales, as noted, are themselves so highly correlated (in this study, r = .94) as to constitute a single measure. In general,
there was no instance in which a scale correlated more strongly with 0
than with P. Leaving aside the theoretical rationale for the expectations
indicated above, it is clear that people's optimistic and pessimistic tendencies, as assessed by the OP, do relate, sometimes at respectable levels, to
their levels of anxiety and to their preferences among the defense mechanisms tapped by the DMI. Correlations with measures of coping strategy,
an alternative conception to that of defense mechanisms, follow the same
general pattern, as summarized immediately below.
In a complex and highly ambitious unpublished doctoral dissertation,
Raquel Natali-Alemany ( 1991) examined, among other things, the relation
between 0 and P and the coping strategies tapped by the Constructive
Thinking Inventory (CTI; Epstein & Meier, 1989). The latter is comprised
of 10 scales, including a general factor called global constructive thinking
(GCT). Although the individual scales yielded interesting results, for the
sake of simplicity and brevity only data related to the GCT coping strategy
will be reported here.
The participants were students enrolled in advanced undergraduate
courses who volunteered for strictly altruistic reasons; all 115 filled out the
OP and CTI questionnaires, but of those, only 81 also agreed to keep a
daily log of their mood states over a 2-week period by way of the Positive
and Negative Affect Schedule (PANAS; Watson, Clark, & Tellegen, 1988)
and the Wessman & Ricks (1966) Elation-Depression scale (ED). In addition, they responded daily to a measure of control or lack of control (Harrigan, Kues, Ricks, & Smith, 1984) and a life events inventory (Zautra,
Guarnaccia, & Dohrenwend, 1986). Again, for the sake of brevity, of those
measures, only the data from the PANAS and the ED Scale will be presented
here. The results reported are from the 81 participants who completed all

'Throughout, when reference is made to variables as correlated, significantly is implied when not
explicitly stated; and for significantly, read p < .05.

THE OPTIMISM-PESSIMISM INSTRUMENT

287

of the measures. Note that the PANAS yields two independent scores, one
for positive affect (PA) and one for negative affect (NA), whereas the
Wessman and Ricks scale yields a score on a unidimensional scale, where
high scores reflect high levels of positive mood.
There were significant, relatively high correlations between the GCT
and both 0 ( r = .51) and P ( r = -.67). These correlations indicate that
optimistic individuals as well as those who are low in pessimism tend to
favor use of coping strategies that can be considered adaptive. Because, as
in prior studies, 0 and P scores were significantly correlated ( r = -.37; a
seemingly low value, but not significantly lower than the values reported
in Dember et al., 1989),it was deemed instructive to examine the correlations
between 0 and P and GCT after partialling out the contribution of the
other measure in order to determine the unique contribution of each to the
zero-order correlations. Thus, when 0 with the P tendency statistically
partialled out was related to GCT, the resulting partial correlation was
reduced to .38, which nevertheless remained significant; similarly,the partial
correlation between P and GCT was reduced to -.60, again still statistically
significant. Thus, with regard to GCT, both 0 and P made unique contributions, with P seemingly the better predictor.
The participants PANAS and E (Elation) scores were averaged over
the 14 days during which they kept daily records. In general, these participants reported being high on PA and low on NA; similarly, their average
E score (6.2 out of a maximum of 10) corresponded closely to the statement
Feeling pretty good; OK,which was associated with the scale value of 6.0.
0 scores were correlated significantly with mean PA ( r = .32) and
with mean E scores ( r = .40). P scores were significantly related only to E
scores ( r = -.28). When 0 and P scores were partialled out from one another,
the partial correlation between 0 and PA was necessarily unchanged, treating the nonsignificant correlation between P and PA as though it had a
value of zero. The partial correlation between 0 and E scores was reduced
from .40 to .33, again still significant, and that between P and E was reduced
from -.28 to -.16, not significant. Thus, in this instance, only 0 scores
were uniquely predictive of mood, as assessed by way of both the PA and
the E scales. Recall that it was P scores that made the greater unique
contribution to GCT. Hence, in this study, it was clearly valuable to have
both 0 and P measures because they operated differentially with regard to
coping strategy and mood.
The Natali-Alemany dissertation was followed up in part and also
extended in a project reported by Geers, Reilley, DeRonde, & Dember
(1996). Reilley, Geers, Lindsay, Dember, and Schafer (1998) also provide
data by way of self-reported physical health, which is best predicted by the
0 and P scales as compared with LOT and attribution style measures (ASQ
and EASQ). More directly pertinent to issues of personality per se, three

288

WlLLlAM N.DEMBER

main concerns were addressed in the Geers et al. (1996) study: (a) The
replicability of the findings by Natali-Alemany that 0 and P scores are
related to the GCT scale on the CTI, a general measure of coping strategy;
(b) the interrelations among three measures of optimism and pessimism,
the OP, Scheier & Carver's (1985) LOT, and the ASQ (see the 48-item
version in Seligman, 1990); and (c) the relative ability of those three
measures to predict scores on both the GCT and the Beck Depression
Inventory (BDI; Beck, Ward, Mendelson, Mock, & Erbaugh, 1961).
Participants were drawn from the Introductory Psychology participant
pool at the University of Cincinnati. The number entering into the correlations fluctuated around 200 (for some scales, there were missing data because
of failure of a few participants to complete all the instruments).
With regard to the GCT, the correlations between it and the 0 and
P scales were both significant and highly similar in level and pattern to
those reported by Natali-Alemany ( r = .54 and -.74). Exactly as in the
original Newman thesis, the correlation between 0 and P scores was -.52.
Hence, partial rs were computed, yielding the following reduced values: For
0 with GCT, the partial T = .35, still statistically significant; for P, the
partial r = -.64, again reduced, but still, of course, significant. Those partial
r values are remarkably similar to those found by Natali-Alemany between
GCT and 0 and P (.38 and .60, respectively). As before, it is P that seems
to make the greater unique contribution to GCT.
Intercorrelations among 0, P, ASQ, and the LOT are given in Table
13.1. With regard to the three instruments for measuring optimism or
pessimism, the highest correlations are those between both 0 and P and
the LOT. 0 and P also correlate with the ASQ, but more modestly and
quite symmetrically. The correlation between the LOT and the ASQ is also
modest. Clearly, and not unexpectedly, the instruments that share a common
format, the OP and the LOT, yield scores that are the most highly related.
The table also shows that 0, P, and LOT are more highly correlated
with BDI than is the ASQ, and that 0, P, and LOT correlate to about the

Table 13.1.
Intercorrelations Among Measures of Optimism and Pessimism
1
2
3
4

P
LOT
ASQ

-.52
.57
.39

-.66

-.39

.29

Note: N = 204; all values of rare significant at p < .05. 0 = optimism: P = pessimism; LOT = Life Orientation Test; ASQ = Attributional Style Questionnaire. Data from "Optimism, Pessimism, Depression, and Coping," poster presented at the annual meeting of the American Psychological Society, San Francisco, CA,
1996. Adapted with permission.

THE OPTIMISM-PESSIMISM INSTRUMENT

289

same degree with BDI. Although not presented here, when the ASQ is
decomposed into scores based only on goodand bad scenarios, attribution
scores from the bad scenarios are the more highly related to BDI scores.
Relations between 0 and P and GCT scores are presented in detail
above. LOT and GCT are correlated at r = .60, a value intermediate between
that for the 0 scale (r = .54) and that for the P scale (r = -.74). The
correlation between ASQ and GCT scores is considerably lower (T = .34).
Two multiple regression analyses were performed, with 0, P, LOT,
and ASQ as predictor variables, and GCT and BDI as criterion measures.
For GCT, the multiple R was .77; P was by far the strongest predictor (p =
-.56), followed by 0 (p = .17) and LOT (p =.13), with ASQ scores making
no unique contribution (p = -.02). For BDI, the multiple R was .54; again,
P was the strongest predictor (p = .29), followed by 0 (p = -.15) and LOT
(p = -.12, ns), with ASQ scores again making no unique contribution to
the prediction of BDI scores (p = .08, ns). Given those outcomes, one might
question the reliability or even the accuracy of scoring of the ASQ protocols,
but note that ASQ scores do relate meaningfully, although not substantially,
to all of the other measures. Finally, here as in prior studies, P seems to be
the most useful of the several scales. 0 does make a unique, although weak,
contribution to the prediction of both GCT and BDI, while the LOT
contributes uniquely, and even more weakly, but only to the prediction of
GCT scores.

SOCIAL INTERACTIONS
The above section focuses on progress to date in elucidating the relations of optimistic and pessimistic tendencies to various personality characteristics. Efforts to examine the impact of these tendencies on social interactions have only recently begun. I use the term social interactions somewhat
loosely to cover disparate issues ranging from the friendship patterns of
optimistic and pessimistic individuals to the optimistic or pessimistic tendencies of special populations, such as business leaders and Peace Corps volunteers.
Anyone at all familiar with the general literature is aware that an
important venue for the application of measures of optimism and pessimism
is the area of physical and mental health. Indeed, the earliest and most
dramatic findings were those that emanated from the Seligman group. Perhaps the best known of those was a longitudinal study by Peterson, Seligman,
and Vaillant (1988) on the physical health of a group of former Harvard
undergraduates, elite students whose health had been followed over several
decades. Through the use of a content analysis system (CAVE; content
analysis of verbatim explanations) applied to essays these students had

290

WILLIAM N.DEMBER

written while at Harvard about a prior wartime experience, it was possible

to assign to each participant a number reflecting that persons explanatory
style. The CAVE technique, in short, served as a surrogate for the ASQ.
Physical health at the time the essays were written was controlled, and
given this there was still a significant correlation between CAVE scores
and a measure of physical health: The more pessimistic the attribution
style in college, the poorer ones health in middle age. Various explanations
for that relation have been offered, including one that directly implicated
impaired immune function. For the present purpose, an especially intriguing
hypothesis was offered by Peterson and Bossio (1991), who argued that
pessimistic individuals have inadequate social support, which has been shown
to put one at risk for poor health. Why should people with pessimistic
attitudes have inadequate social support? Peterson and Bossio further argue
that by virtue of their pessimism, such people simply do not make good
company. Hence, pessimistic individuals have trouble maintaining friendships.
One implication of that line of thought is that pessimistic people will
by default eventually end up with close friends who are equally pessimistic,
that is, the best each member of a friendship dyad can do is to pair up with
people who are no less pessimistic. Of course, in the real world, one does
not have an infinite number of potential friends to choose from or be chosen
by, nor is there time enough for the process to yield friendship pairs exactly
equal in pessimism. In addition, there are other bases for friendship maintenance besides ones pessimistic orientation. Nevertheless, it might be expected that the pessimism scores of close friends would be significantly correlated.
We (Dember, Geers, Reilley, & Raisman, 1995) attempted to test that
expectation by asking students in the Introductory Psychology participant
pool at the University of Cincinnati to come to a room on campus along
with a person they considered to be a close friend. The degree of friendship
was assessed by way of a questionnaire given to both members of each dyad;
based on a stringent criterion, the data from a few dyads were not used for
failure to meet that criterion. One initial surprise was the large number of
male-female friendship pairs. Because such mixed-gender pairs might have
a very different basis for a close friendship than same-gender pairs, we
decided to analyze the data separately for male-male, female-female, and
male-female dyads. In addition to having the participants respond to the
OF instrument, we also asked them to respond to the EASQ, and for another
purpose entirely, to fill out a health questionnaire. Only the correlational
data pertinent to the hypothesized relation between pessimism scores of
close friends will be reported here.
In essence, no significant correlations were found for the male-female
(N = 20) or female-female (N = 40) pairs on 0,P, or EASQ scores. Indeed,
THE OPTIMISM-PESSIMISM INSTRUMENT

29 I

the only significant correlations involving EASQ were with 0 and P (both
in the low .20s; N = 184, p < .01). However, the 0 scores of the 20
male-male pairs were highly correlated ( 7 = .65, p < .05). So, from these
results, what seems to matter in friendship maintenance among male college
students is their level of optimism, not their level of pessimism. In an effort
to pin that finding down and to see whether there might be other personality
variables on which friendships are founded (perhaps it is nothing more than
birds of a feather flocking together), Sean Reilley conducted a second
study (Reilley & Dember, 1996; see also Geers, Reilley, & Dember, 1998)
following the same general logic, only this time omitting the EASQ and
adding three other brief measures, the Self-Expression (Assertiveness) Scale
(SES; Galassi, Delo, Galassi, & Bastien, 1974); the NYU Loneliness Scale
(Rubenstein & Shaver, 1992); and the Sensation Seeking Scale I1 (SSS;
Zuckerman, 1979).Again, dyads were categorized as male-female (N= 31),
female-female (N = 39), and male-male (N = 22). As before, only the
correlation between 0 scores was significant and only for the male-male
pairs (again, T = .65, p < .05). With regard to the other measures, the only
significant between-friends correlation was that for male-female dyads on
the SSS (T = .37, p <.05). One can offer a reasonable interpretation of the
last finding, but it is not pertinent to the theme of the present chapter.
What does seem clear from these two studies is that although optimism
does seem important as basis for friendship among male friends, there is no
evidence of a role for pessimism. Why optimism works, and why only for
male-male dyads is not clear.
It is possible that the results represent an artifact of the way that the
students friends were solicited for the study. We have speculated that
optimistic and pessimistic individuals have several friends covering a range
of optimism and pessimism values. Indeed, our data from the friendship
questionnaire show that optimistic and pessimistic individuals have equal
numbers of friends, although the length of their typical friendships differs,
with optimists reporting friendships of longer duration. We assume that
optimistic people will in general be more inclined to assent to participate
in a study of this sort (with nothing in it for them except being able to
help out a friend in need) than will pessimistic individuals. So, when an
optimistic student turns, perhaps wittingly, to one of his or her optimistic
friends with a request to participate, that overture is likely to be accepted;
similar requests to pessimistic friends, should they occur, will generally
be rejected. So optimists come to the study with friends who are also
predominantly optimistic. By contrast, when pessimistic individuals solicit
participation from among their friends, who may indeed be predominantly
pessimistic, they may also turn initially to whatever optimistic friends they
have, expecting them to accede. But, since their friends are predominantly
pessimistic, some of whom may agree to participate, pessimistic individuals
292

WILLIAM N.DEMBER

may come to the study with a mixture of optimistic and pessimistic friends.
Such a scenario might explain why optimism worked in this study, but
pessimism did not, although it is silent on why only for male-male pairs.
In any case, it would seem worthwhile to find an alternative method of
determining the optimism and pessimism levels of peoples close friends to
check the generality of the present results. In the meantime, the Peterson
and Bossio speculation about pessimists inadequate support networks has
led us to some intriguing and puzzling findings as well as to further evidence
for the non-bipolarity of optimism and pessimism.
The final two studies described below are a bit hard to categorize as
involving either personality correlates of optimism and pessimism or social
interactions. Perhaps it is best to think of them as concerned with social
settings in which optimism and pessimism are implicated, either because
the settings induce such orientations or because such orientations serve as
screens for entry into those settings.
In the first study, one question, among others, was whether Peace
Corps volunteers who had returned from service would show higher than
usual levels of optimism, unusually low levels of pessimism, or perhaps both.
Based on some pilot interviews, the critical variable seemed to be pessimism.
And that is how it turned out. Peace Corps volunteers who had returned
from their assignments filled out the OP along with other measures and
participated in a lengthy interview (Gerety, 1996). Of interest here is the
finding that this group (N = 30), although having a mean 0 score (54.90)
not significantly different from that of a norm group (53.75) of 1,033 undergraduates who had served in four prior studies (Dember et al., 1989, Studies
1 and 2; Hummer, Dember, Melton & Schefft, 1992; Terezis, 1990), did
have a significantly and substantially lower mean P score (norm group M =
37.1; Gerety M = 32.4; t = 3.74, p < .001). I need to note here that the
norm group mean 0 score was incorrectly reported as 55.75 in a prior study
(Lewis, Dember, Schefft, & Radenhausen, 1995) and used by Gerety as well
as in the study below by Wunderley, Reddy, and Dember (1998). Use of
that inflated norm group value had no impact on the Gerety et al. results,
but it did lead to an erroneous conclusion by Wunderley et al. regarding
the relative optimism level of the participants in their study.
As suggested above, it is not clear whether their unusually low level
of pessimism is what made it easy for the Peace Corps volunteers to undertake
this difficult assignment and, perhaps, made them attractive to the selection
officers, or whether their success in the field lowered their initial level of
pessimism. Of course, all of the above factors may have been implicated.
It would be interesting to assess applicants for Peace Corps service for their
0 and P levels, but to keep those data from entering into the selection
decision. In that way, one could examine both the 0 and P levels of
applicants in general as well as the scores of those who had been selected.
THE OPTIMISM-PESSIMISM INSTRUMENT

293

It seems plausible that P scores would be lower than typical among the
applicants and then would progressively decline through the selection and
postservice phases.
A somewhat different outcome was found in the second study (Wunderley et al., 1998) on special groups, in this case 48 male business leaders
and 197 (49 women, 148 men) of their direct reports or constituents.
These participants worked for mid- to large-size manufacturing firms in the
greater Cincinnati area. Leaders as well as their constituents filled out the
OP instrument along with two other measures: (a) The Kirton AdaptionInnovation Inventory (KAI; Kirton, 1976), which assesses ones tendencies
to respond to challenge either simply by adapting to it or by initiating
change through innovative strategies; and (b) the Leadership Practices
Inventory (LPI; Kouzes & Posner, 1988). The latter instrument can be
taken either from the perspective of ones own behavior (LPI-Self) or to
characterize the leadership practices of someone else (LPI-Observer). In this
study, the measure of interest was the LPI-Observer scores of constituents
judging their particular leader, that is, a measure, on five factors (scales),
of how the constituents viewed their leader. For purposes of feedback, the
leaders filled out the LPI-Self, but those data were not analyzed, because
the focus of the study was on leaders per se. For the sake of economy, the
constituents did not fill out either the KAI or the LPI-Self. The five LPI
scales are called Challenging the Process, Inspiring a Shared Vision, Enabling
Others to Act, Modeling the Way, and Encouraging the Heart.
Both the leader group and the constituents had significantly and substantially lower mean P scores (leaders: M = 31.35, t = 7.61; constituents:
M = 32.57, t = 11.16; in both cases, p < .001) than the norm group. Both
groups also had significantly, although moderately, higher 0 scores than
the norm group (leaders, M = 56.38, t = 3.41; constituents: M = 55.65, t =
2.46; in both cases, p < .01). Again, it seems that what differentiates these
successful business leaders and their constituents, who are on a track toward
leadership positions, from the norm group is their dramatically low P scores,
although their 0 scores were also somewhat elevated. Indeed, it seems
reasonable that excessively optimistic business leaders would seem naive
and lack credibility, whereas very low levels of pessimism would be a virtue.
In the Peace Corps study and in this one, the participants were older than
members of the norm group, which might account for their low P scores.
However, correlational analyses indicated that age per se probably does not
account for the results obtained.
For the measure of innovativeness, the KAI, a significant, but very
modest correlation was obtained with P scores ( r = -.27, p < .05). That is,
pessimistic leaders tend not to employ innovative problem-solving strategies.
For the LPI-Observer measures, which assess the constituents impressions
of their leaders, the data were examined by first obtaining a mean value on
294

w u r m N.DEMBER

each of the five scales of the ratings given by all of a given leadersconstituents
(typically, 5 or 6, but ranging from 3 to 9). As expected, all five scales were
positively correlated with 0 and negatively with P, but of those correlations,
only two were significant: Inspiring a Shared Vision and Encouraging the
Heart, both of which correlated with 0, again modestly, at r = .27 (p <
.05). Thus, with regard to level, it was P that differentiated these participants
from the norm group, and P that was predictive of problem-solving style,
whereas 0 was predictive of the constituents assessments of their leaders
ability to inspire and encourage. Once more, the results indicate that there
is virtue in measuring 0 and P separately and not combining them into a
single measure. Note also that the LPI-Observer scores are the first to
correlate with an OP measure that does not emanate from self-report, that
is, to tap the impact that ones optimism or pessimism has on close associates,
lending what might be called ecological validity to the OP instrument.

CONCLUSION
What began as a search for correlates of individual differences in
susceptibility to the Pollyanna principle has led to the development of an
instrument for measuring optimism and pessimism as separate constructs.
The OP instrument has thus far proven to have reasonable psychometric
properties v i s - h i s internal consistency and test-retest reliability. It yields
means that are quite consistent from study to study, except when special
populations are investigated, such as returned Peace Corps volunteers and
business leaders. Both of the latter two groups departed from normative
values on pessimism, markedly so for the business leaders. The latter were
also somewhat more optimistic than the normative population, although
the Peace Corps volunteers were not.
Having a separate measure of each construct turns out to be not
only psychometrically indicated, because 0 and P scores are usually only
moderately correlated, but also to offer additional predictive use, because
the two measures often correlate differentially with other variables.
0 and P have been shown to correlate sensibly with both measures
of defense mechanisms and of coping style as well as with a variety of
personality traits and relevant single-item questionnaire responses. Although
such correlations typically are only in the modest-to-moderate range, correlation coefficients in the .60s and .70s have been found between P and measures of anxiety and coping. The OP instrument yields reasonable correlations with other purported measures of optimism and pessimism, that is the
LOT and the ASQ, but also contributes unique variance to the prediction
of such variables as depression and coping style. Indeed, the P measure
THE OPTIMISM-PESSIMISM INSTRUMENT

295

generally is the strongest predictor of those variables compared with 0,

LOT, and ASQ.
There is some evidence that close male friends have highly correlated
0, but not P scores, a finding that probably reflects more than a simple
like-seeks-like tendency. W h y no between-friend correlations on 0 are
found when a female is part of the dyad or when P is the measure being
examined remain questions for further research. The developmental-genetic
origins of differences in optimistic and pessimistic tendencies need to be
further explored, as well as the extent to which 0 and P are subject to longterm change through success or failure in meeting formidable challenges. It
would be instructive to examine optimism and pessimism in people for whom
pessimism might be adaptive, such as commercial airline pilots. Finally,
there is need for further clarification of the meaning of the two constructs,
optimism and pessimism, some of which hopefully will emerge from the
several chapters in this volume and some, undoubtedly, from the future
work that the volume inspires. Most important, more highly sophisticated
studies on the issue of the purported non-bipolarity of optimism and pessimism should be conducted. Additional research is also warranted on the
moderating role of optimism and pessimism in the performance of both
laboratory tasks and so-called real-world work assignments and especially on
the impact of both negative and positive feedback on mood and subsequent
performance of optimistic and pessimistic individuals.

REFERENCES
Bantirgu, T. (1995). Psychological effects of prolonged stress on Ethiopian expatriates
in the United States. Unpublished doctoral dissertation, University of Cincinnati, Ohio.
Beck, A., Ward, C., Mendelson, M., Mock, J., & Erbaugh, J. ( 1961). An inventory
for measuring depression. Archives of General Psychiatry, 4, 53-63.
Byrne, D., Barry, J., & Nelson, D. (1963). Relation of the revised repressionsensitization scale to measuresof self-description.Psychological Reports, 13,323334.
Chang, E. C., DZurilla, T. J., & Maydeu-Olivares,A. (1994). Assessing the dimensionality of optimism and pessimism using a multi-measure approach. Cognitive
Therapy and Research, 18, 143-161.
Cronbach, L. J. (1951). Coefficient alpha and the internal structure of tests. Psychomrrika, 16, 297-335.
Dember, W. N. (1974). Motivation and the cognitive revolution. American Psychologist, 29, 161-168.

296

WILLIAM N . DEMBER

Dember, W. N. (1989). Cognition, motivation, and emotion: Ideology revisited.

In R. R. Hoffman & D. S. Palermo (Eds.), Cognition and the symbolic processes:
Applied and ecological perspectives (pp. 153-162). Hillsdale, NJ: Erlbaum.
Dember, W. N., & Brooks, J. (1989). A new instrument for measuring optimism
and pessimism: Test-retest reliability and relations with happiness and religious
commitment. Bulktin of the Psychonomic Society, 27, 365-366.
Dember, W. N., Geers, A. L., Reilley, S. P., & Raisman, S. (1995, May). Optimism,
pessimism, fiendship, and health. Poster presented at the annual meeting of the
American Psychological Society, New York.
Dember, W. N., Martin, S. H., Hummer, M. K., Howe, S. R., & Melton, R. S.
(1989). The measurement of optimism and pessimism. Current Psychology:
Research and Reviews, 8, 109-119.
Dember, W. N., & Penwell, L. (1980). Happiness, depression, and the Pollyanna
principle. Bulletin of the Psychonomic Society, 15, 321-323.
Epstein, S., & Meier, P. (1989). Constructive thinking: A broad coping variable with
specific components. Journal of Personality and Social Psychology, 57, 332-350.
Fischer, M., & Leitenberg, H. (1986). Optimism and pessimism in school-aged
children. Child Development, 57, 241-248.
Galassi, J. P., Delo, J. S., Galassi, M. D., & Bastien, S. (1974). The college selfexpression scale: A measure of assertiveness. Behavior Therapy, 5 , 165-1 71.
Geers, A. L., Reilley, S. P., & Dember, W. N. (1998). Optimism, pessimism, and
friendship. Current Psychology: Research and Review, 17, 3-19.
Geers, A. L., Reilley, S. P., DeRonde, L., & Dember, W. N. (1996, July). Optimism,
pessimism, depression, and coping. Poster presented at the annual meeting of
the American Psychological Society, San Francisco, CA.
Gerety, C. A. (1996). Admired figures, values and attitudes of returned Perace Corps
volunteers. Masters thesis, University of Cincinnati, Ohio.
Kues, J. R., Ricks, D. F., & Smith, R. (1984). Moods that predict
Harrigan, J. ,4.,
coming migraine headaches. Pain, 20, 385-396.
Helton, W. S., Dember, W. N., & Warm, J. S. (1999). Optimism-pessimism and
false failure feedback: Effects on vigilance performance and stress. Current
Psychology: Research and Review, 18, 31 1-325.
Hummer, M. K., Dember, W. N., Melton, R. S., & Schefft, B. K. (1992). On the
partial independence of optimism and pessimism. Current Psychobgy : Research
and Reviews, 11, 37-50.
Ihilevich, D., & Gleser, G. C. (1986). Defense mechanisms: Their classification,
correlates and meusurement with the Defense Mechanisms Inventory. Owosso, MI:
DMI Associates.
Kirton, M. ( 1976). Adaptors and innovators. Journal of Applied Psychology, 61,622629.
Kouzes, J. M., & Posner, B. 2. (1988). The leadership practices inventory. San Diego,
CA: Pfeiffer.
THE OPTIMISM-PESSIMISM INSTRUMENT

297

Lewis, L. M., Dember, W. N., Schefft, B. K., & Radenhausen, R. A. (1995). Can
experimentally induced mood affect optimism and pessimism scores? Current
Psychology: Developmental, Learning, Personality, Social, 14, 29-41.
Martin, S. H. (1986). Optimism, pessimism, and psychological defense. Masters thesis,
University of Cincinnati, Ohio.
Matlin, M. W., & Stang, D. J. (1978). The Pollyanna principle. Cambridge, MA:
Schenkman.
Natali-Alemany, R. (1991). Moods, coping, and perception of daily life ewents: Are
these factors related to optimism and pessimism? Doctoral dissertation, University
of Cincinnati, Ohio.
Newman, M. K. (1983). The measurement of optimism/pessimism. Masters thesis,
University of Cincinnati, Ohio.
Peterson, C., & Bossio, L. M. (1991). Health and optimism. New York: Free Press.
Peterson, C., Seligman, M. E. P., & Vaillant, G. E. (1988). Pessimistic explanatory
style is a risk factor for physical illness: A thirty-five year longitudinal study.
Journal of Personality and Social Psychology, 55, 23-27.
Peterson, C., Semmel, A., von Bayer, C., Abramson, L. Y., Metalsky, G. I., &
Seligman, M. E. P. (1982). The Attributional Style Questionnaire. Cognitive
Therapy and Research, 6, 287-299.
Peterson, C., & Villanova, P. (1988). An expanded Attributional Style Questionnaire. Journal of Abnormal Psychology, 97, 87-89.
Reilley, S. P., & Dember, W. N. (1996, July). Correlations among fnends scores
on optimism, pessimism, loneliness, assertiveness, and sensation seeking. Poster
presented at the annual meeting of the American Psychological Society, San
Francisco, CA.
Reilley, S. P., Geers, A. L., Lindsay, D. L., Dember, W. N., & Schafer, J. C. (1998,
May). Disentangling the importance of different optimism-pessimism instruments:
Research issues and implications. Poster presented at the annual meeting of the
American Psychological Society, Washington, DC.
Rubenstein, C. M., & Shaver, P. (1992). The experience of loneliness. In L. A.
Peplau & D. Perlman (Eds.), Loneliness: A source book of cuwent theory, research,
and therapy (pp. 206-223). New York: Wiley.
Scheier, M. F., & Carver, C. S. (1985). Optimism, coping, and health: Assessment
and implications of generalized outcome expectancies. Health Psychology, 4,
2 19-247.
Seligman, M. E. P. (1990). Learned optimism. New York: Knopf.
Sethi, S., & Seligman, M. E. P. (1993). Optimism and fundamentalism. Psychological
Science, 4, 256-259.
Terezis, H. C. (1990). Optimism and pessimism: Manipulation via musical mood induction. Masters thesis, University of Cincinnati, Ohio.
Watson, D., & Clark, L. A. (1984). Negative affectivity: The disposition to experience aversive emotional states. Psychological Bulktin, 96, 465-490.

298

WILLIAM N.DEMBER

Watson, D., Clark, L. A., & Tellegen, A. (1988). Development and validation of
brief measures of positive and negative affect: The PANAS.]ournal of Personality and Social Psychology, 54, 1063-1070.
Welsh, G. S. (1956). Factor dimensions A and R. In G. S. Welsh & W. G.
Dahlstrom (Eds.), Basic readings on the MMPI in psychology and medicine (pp.
264-281). Minneapolis, MN: University of Minnesota Press.
Wessman, A. E., & Ricks, D. F. (1966). Mood and personality. New York: Holt.
Wunderley, L. J., Reddy, W. B., & Dember, W. N. (1998). Optimism and pessimism
in business leaders. Journal of Applied Social Psychology, 28, 751-760.
Zautra, A. J., Guarnaccia, C. A., & Dohrenwend, B. P. (1986). Measuring small
life events effects. American Iournal of Community Psychology, 14, 629-655.
Zuckerman, M. (1979). Sensationseeking: Beyond the optimal level ofarousal. Hillsdale,
NJ: Erlbaum.

THE OPTIMISM-PESSIMISM INSTRUMENT

299

BUILDING OPTIMISM AND

PREVENTING DEPRESSIVE SYMPTOMS
IN CHILDREN
JANE E. GILLHAM, KAREN J. REIVICH, AND ANDREW J. SHATTE

Although adult depression is common (estimates are that as many as one

in five of us will experience clinical depression in our lifetime), until relatively
recently, it was believed that childhood depression was rare. Epidemiological
studies indicate that between 2% and 5% of children and adolescents are
clinically depressed (Anderson, Williams, McGee, 6r Silva, 1987; Kovacs,
1989) and as many as 30% of adolescents experience depressive symptoms at
any given time (Angold, 1988; Rutter, 1986). Depression in young people
seems to be on the rise. Recent findings indicate that as many as 20% of adolescents may experience an episode of clinical depression by the end of high
school (Lewinsohn, Hops, Roberts, 6r Seeley, 1993). Many depressed youth
experience a pervasive sense of hopelessness, which in the worst cases can
Lead to suicide. Each year, 8% of high school students attempt suicide and
approximately 13 of every 100,000adolescents in this country take their own
lives (Centers for Disease Control, 1991). Clearly, it is crucial for researchers

301

to increase our understanding of the origins of optimism and depression, and

to develop interventions to treat and prevent depressive disorders in children.
Cognitive models of adult depression have been widely researched. In
particular, as we discussed earlier, a large body of research indicates that a
pessimistic explanatory style and dispositional pessimism are linked to depression in adults (see chap. 3, this volume; see also Beck, 1976; Robins & Hayes,
1995; Scheier & Carver, 1992). Cognitive-behavioral therapy, which targets
these cognitive vulnerabilities, is effective in boosting explanatory style, increasing optimism, and treating depression in adults (Beck, Hollon, Young,
Bedrosian, & Budenz, 1985;DeRubeis & Hollon, 1995; Ilardi, Craighead, &
Evans, 1997; Seligman et al., 1988). In recent research, psychologists have
begun to apply cognitive models and interventions to childhood and adolescent depression. In this chapter we review this research in three major sections.
First, we provide a brief overview of the cognitive model and of cognitivebehavioral therapy for depression in adulthood. This overview focuses on the
relationship of explanatory style and dispositional optimism to depression.
Second, we review research on the development of optimism, explanatory
style, and depression in children. Developmental and environmental risk factors are discussed, as well as research on the link between optimism and explanatory style and depression in children. In the final section, we describe interventions that may foster optimism and build resilience to depression in
children and adolescents. As will be shown, research has not yet provided
much knowledge about the origins of explanatory style or optimism, and we
have a lot to learn about how to promote resiliency in children. Thus, this
chapter outlines an area that is important and exciting for future research.

COGNITIVE MODELS OF DEPRESSION

Cognitive models of depression have in common the emphasis they
place on the role of beliefs or interpretations in determining our emotional
and behavioral reactions to events. Two types of beliefs that have been
widely researched are explanatory style and optimism (for greater discussion
of research on explanatory style and depression, see chap. 3, this volume).
Briefly, a pessimistic explanatory style and pessimistic expectations are linked
to depressive symptoms and depressive disorders in adults. There is some
evidence that these variables increase the risk for subsequent depressive
symptoms and episodes (Beck, 1976; Robins & Hayes, 1995; Scheier &
Carver, 1992).According to the hopelessness theory of depression, explanations and expectations are closely connected (Abramson, Metalsky, & Alloy,
1989). Individuals who attribute setbacks to stable and global factors are
more likely to experience pessimism and helplessness following negative life
events. Such individuals will believe there is little they can do to change
302

GlLLHAM ET AL.

their circumstances and will expect problems to persist over time. According
to the hopelessness theory, these individuals will be more vulnerable to
developing depression than those with an optimistic explanatory style, particularly to developing a subtype of depression characterized by profound
hopelessness and suicidal ideation.
Cognitive-behavioral therapy (CBT) targets the negative interpretations depressed clients often make for events in their lives. Typically, CBT
involves two overlapping components: a cognitive-restructuring component
and a skill-acquisition component. In the cognitive-restructuring component, clients learn to identify negative interpretations for events and to
evaluate these interpretations by considering alternatives and evidence for
them. The cognitive component helps individuals to see that setbacks they
view as catastrophic are, in fact, often caused by factors that are temporary
or changeable. The skill-acquisition component equips individuals with
behavioral skills that allow them to more successfully accomplish their goals.
Clients learn a variety of skills for coping with difficult emotions and dayto-day problems. Skills typically include relaxation, assertiveness, and procedures for breaking large tasks into manageable components. As clients apply
these skills and are able to achieve desired outcomes, they may become
increasingly confident and optimistic in their interpretations of events. Thus,
both the cognitive restructuring and the behavioral skills appear to be
important for boosting explanatory style and optimism.
CBT is effective at treating adult depression (Beck et al., 1985; DeRubeis & Hollon, 1995; Ilardi et al., 1997; Seligman et al., 1988), and has
been found to be more effective than pharmacotherapy in preventing relapse
(e.g., Evans, Hollon, DeRubeis, & Piasecki, 1992). This is important because
recent research indicates that, for many people, depression is a recurrent
disorder (Cicchetti & Toth, 1998). Long-term improvement following CBT
has been attributed to CBT's focus on providing clients with cognitive and
behavioral skills that can be used when problems are encountered in the
future, long after therapy has ended. Improvement in explanatory style
during cognitive therapy for depression is linked to successful outcome and
maintenance of gains (DeRubeis & Hollon, 1995; Ilardi et al., 1997; Seligman et al., 1988).A recent study indicates that CBT can improve explanatory style and prevent depressive symptoms and anxiety in college students
at risk for depression (Seligman, Schulman, DeRubeis, & Hollon, 1999).

DEVELOPMENT OF DEPRESSION, OPTIMISM,

AND EXPLANATORY STYLE IN CHILDREN
Are cognitive models applicable to depression in children and adolescents?Over the past 15 years, researchers have begun to explore this question.
BUILDING OPTIMISM AND PREVENTING DEPRESSIVE SYMPTOMS

303

Before we can fully answer it, however, we need to know the age at which
children can experience depression. Not surprisingly, there has been a great
deal of debate on this topic (see Bemporad, 1994, for a review). Many
psychodynamic theorists postulated that clinical depression required a fully
developed superego and sense of self. Thus, depression could not be experienced until adolescence. However, research applying diagnostic criteria to
children has revealed that children do experience depression before they
reach adolescence. As many as 7-9% of children may experience an episode
of depression before the age of 14 (Garrison, Schluchter, Shoenbach, &
Kaplan, 1989). Depression in children under 10 appears to be quite rare
(Rutter, 1986).
According to Bemporad ( 1994),depressive disorders rarely occur before
middle-to-late childhood. Very young children may experience transient
depressive reactions to stressful events, but these reactions subside after these
events have ended and endure only when stressors are chronic. Beginning
in middle childhood, depressive reactions become increasingly stable and
independent of stressful life events. Research indicates that depression increases during the middle school years and is quite common by adolescence
(Hankin, Abramson, Moffitt, Silva, & McGee, 1998; Kashani, Rosenberg,
& Reid, 1989;Rutter, 1986; but see Nolen-Hoeksema, Girgus, & Seligman,
1992). Cognitive symptoms of depression are absent or rare in preschoolers,
but become increasingly common during the elementary school years (Bemporad, 1994; Cicchetti & Toth, 1998). For example, McConville and colleagues found that childrens depressive reactions were not characterized by
low self-esteem until age 8. Hopelessness did not emerge as a symptom in
children until closer to age 10 (McConville, Boag, & Purohit, 1973).
Cognitive Development and Explanatory Style and Optimism
Developmental research suggests that cognitive models of depression
become increasingly applicable to children as they develop (Garber & Flynn,
1998). The transition to formal operations, and changes in self-concepts,
causal thinking, and thinking about the future may make explanatory style
and dispositional optimism especially relevant to depression in children by
the beginning of the middle school years.
As childrens thinking moves from the preoperational to the concrete
operational, and finally, to the formal operational stage, it becomes increasingly governed by abstract schemas or rules (Bemporad, 1994; Inhelder &
Piaget, 1958). Although there is considerable debate and inconsistency in
research on the age at which formal operations is achieved, interpretations
of events may be increasingly driven by cognitive styles such as explanatory
style or dispositional optimism as children approach adolescence.

304

GILLHAM ET AL.

Self-Concepts
Childrens self-concepts undergo a dramatic change with development.
In preschool and early elementary school, their self-descriptions involve
observable characteristics and behaviors (e.g., I play the piano). By late
childhood, these descriptions are more stable and more likely to involve
personality traits and social comparison information (e.g., Im a sensitive
person, Im more generous than most people; Cicchetti & Schneider-Rosen,
1986; Damon & Hart, 1988),characteristics that are less tied to observable
behaviors and, therefore, may be more prone to biases (Harter, 1990).
This shift toward more traitlike descriptions is not likely to be a problem
for children with positive self-concepts. Unfortunately, however, there is
some evidence that children evaluate their abilities more negatively as they
move from early to late childhood (e.g., Frey & Ruble, 1987, 1990; Pintrich
& Blumenfeld, 1985). For example, Ruble and colleagues (Ruble, Boggiano,
Feldman, & Loebl, 1980) found that kindergartners ratings of their ability
were higher than the ratings of second and fourth graders following identical
feedback. Pintrich and Blumenfeld (1985) found that second graders rated
their academic ability and their behavioral conduct more positively than
did sixth graders.
Childrens concepts of ability may also change in ways that make their
self-concepts more vulnerable. According to Dweck and Leggett ( 1988),
young children tend to believe that ability is a malleable quality. Thus,
practice and learning can make one smarter. Older children, in contrast,
are more likely to believe ability is a stable quality that cannot change.
Intelligence is perceived to be a fixed entity that is revealed through academic
performance and achievement. Similarly, Langer and Park ( 1990) propose
that young children are taught to attribute failure to precompetence (I
havent learned this yet), and older children are taught to attribute failure
to incompetence (Im not able to do this).
The research on childrens self-concepts and conceptions of ability
has important implications for the development of explanatory style and
optimism. It suggests that older children may be more likely than younger
children to attribute problems to personality flaws that are stable and global.
If this is true, positive outcomes will seem less likely and more difficult
to attain.

Causal Attributions
Older children may be more prone than younger children to making
causal attributions. Piaget (1926, 1959) reported that children under the
age of 6 or 7 were rarely concerned with the causes of events, and frequently
had difficulty expressing causal relationships. For example, in a 1926 study,

BUILDING OPTIMISM AND PREVENTING DEPRESSIVE SYMPTOMS

305

he asked 6- 10-year-old children to finish incomplete sentences that stopped

at the word because (e.g., The man fell off his bicycle because-). Young
children frequently had difficulty completing these sentences and tended
to give responses involving expectations or predictions (e.g., he broke his
arm; Piaget, 1926).
We do not yet know the ages at which explanatory style becomes
stable or first can be measured. Children can complete paper-and-pencil
measures of explanatory style beginning at about 8 years of age (NolenHoeksema et al., 1992). Little research exists on explanatory style in younger
children. Nolen-Hoeksema ( 1986) attempted to assess explanatory style in
preschool children using interviews. Similar to Piagets findings, the findings
in her study revealed that preschoolers rarely gave causal explanations for
events. Instead, they talked about solutions and expectations. It is interesting
to note that a focus on solutions, rather than on causes, may protect young
children from hopelessness and helplessness. Dweck and Leggett ( 1988)
found that elementary school children who are prone to helplessness (who
give up following failure) attribute their failures to stable and global causes
(e.g., Im stupid). In contrast, children who display persistence when
challenged focus on solutions (e.g., I need to try harder). Thus, persistent
children are less likely to describe causes for failure; although it can be
argued that their solutions imply unstable and specific causes (e.g., I didnt
try hard enough).
Thinking About the Future
Thinking about the future becomes increasingly sophisticated as children approach adolescence. They are more able to think about long-term
goals and to consider alternative possibilities for the future (Bemporad,
1994; Keating, 1980). Yet, Bemporad argues that adolescents dont have
the life experience necessary to put negative events into context. Thus,
they may frequently catastrophize, believing that consequences of negative
events will endure through time. For example, a poor grade on a geometry
exam may be taken as evidence that pursuing a career in math or science
is hopeless. If a childs explanatory style becomes increasingly pessimistic
with development, we should expect that his or her dispositional pessimism
will also increase, because our predictions about the future in part stem
from the causal attributions we make (Abramson et al., 1989). Consistent
with this hypothesis, hopelessness has been found to be a common symptom
of depression in adolescence (McConville et al., 1973).
Link Between Pessimistic Interpretive Styks and Depression
Recent research indicates that cognitive models are applicable to depression in children at least by late childhood. Children with depressive
306

GILLHAM ET AL

symptoms report lower self-esteem, greater hopelessness, and more pessimistic explanatory styles than do children who do not suffer from these symptoms
(for reviews, see Garber, Quiggle, & Shanley, 1990; Gladstone & Kaslow,
1995; Joiner & Wagner, 1995a; Kaslow, Brown, & Mee, 1994). It is not
yet clear whether pessimism and a pessimistic explanatory style are causes
or simply covariates of depression. Nolen-Hoeksema and colleagues have
found that explanatory style predicts future symptoms of depression (NolenHoeksema, Girgus, & Seligman, 1986, 1992). However, Hammen and colleagues found that explanatory style did not predict depressive symptoms
in children (Hammen, Adrian, & Hiroto, 1988; for reviews see Gladstone
& Kaslow, 1995; Joiner & Wagner, 1995a). The relationship between explanatory style and depressive symptoms may change over time. In one
longitudinal study of children as they moved from third through eighth
grade, young children's depressive symptoms were predicted by life events
alone (and not by explanatory style). Once children reached sixth grade,
explanatory style began to predict depressive symptoms (Nolen-Hoeksema
et al., 1992).
McCauley and colleagues studied explanatory style in children from
ages 7-17 (McCauley, Mitchell, Burke, & Moss, 1988). Consistent with
the predictions above, the study showed that explanatory style became
increasingly pessimistic over this time period. In our research, we found
that explanatory style grew increasingly pessimistic in a sample of children
followed for .3 '/2 years from the beginning of middle school to the beginning
of high school (Gillham & Reivich, 1999a). Curiously, depressive symptoms
did not increase significantly over this time period, although they have been
found to increase in other studies (Kashani et al., 1989, Rutter, 1986). In
contrast to our findings and to those reported by McCauley and colleagues,
Garber and colleagues found no significant change in pessimistic interpretive styles from seventh through twelfth grade (Garber, Weiss, & Shanley,
1993).
Taken together, these findings suggest cognitive factors implicated in
adult depression become increasingly relevant in children, especially as
they approach adolescence. Early adolescence may be an important window
during which to target explanatory style and optimism.
Environmental Risk Factors

Of course, as Cicchetti and Toth (1998) point out, depression cannot

be explained by developmental changes alone. Although a substantial minority of adolescents will experience a depressive episode, most are resilient.
A variety of environmental risk factors for depression have been proposed,
including parental depression and psychopathology, marital conflict, poverty, and other life stressors (for reviews, see Cicchetti & Toth. 1998; Downy
BUILDING OPTlMlSM AND PREVENTlNG DEPRESSIVE SYMPTOMS

307

& Coyne, 1990). All these factors are likely to affect childrens explanatory
style, levels of optimism, and, in turn, their depressive symptoms.
Little is known about the origins of optimism or explanatory style (see
Garber & Flynn, 1998 for a recent review of literature on the origins of
cognitive styles). Psychologists have suggested that children learn interpretive styles from the important adults in their lives. Parents may instill
optimistic or pessimistic styles through modeling or through the way they
habitually explain events in their childrens lives (Seligman, Reivich, Jaycox,
& Gillham, 1995). They may reinforce certain types of interpretations made
by their children. Only a handful of studies have investigated the relationship
between parents and childrens styles for interpreting events. These have
yielded conflicting findings. In some studies, parent and child explanatory
styles converge. In others, no correlation is found (for a review, see Joiner
& Wagner, 1995b). Dweck and colleagues have proposed that children may
internalize the attributions made by their teachers regarding their failures
and successes (Dweck, Davidson, Nelson, & Enna, 1978).These researchers
found that girls gave more pessimistic explanations for their failures than
did boys, and that this difference was consistent with a discrepancy in the
attributions teachers gave for failure in male versus female students (Dweck
et al., 1978).
Negative life events or stressors may also predispose children to a
pessimistic explanatory style. Early helplessness experiments demonstrated
that exposure to uncontrollable aversive events leads to passivity and depressive symptoms in humans and other animals. Conversely, control and mastery
experiences protect against helplessness (e.g., Seligman, 1975). Children
who grow up in poverty, who are exposed repeatedly to violence or intense
marital conflict, or who are exposed to chronic parental mental illness, and
who repeatedly experience failure may be more prone to developing a
pessimistic explanatory style. Consistent with this hypothesis, pessimistic
interpretive styles have been linked to previous experience of life stressors,
including parental conflict (Nolen-Hoeksema, Girgus, & Seligman, 1986),
maternal psychopathology (Garber & Robinson, 1997), and a history of
child abuse (Gold, 1986; Kaufman, 1991).

PROMOTING OPTIMISTIC INTERPRETIVE STYLES

AND PREVENTING DEPRESSIVE SYMPTOMS IN CHILDREN
Can interventions that promote optimism and reduce depression in
adults be applied effectively to children? Are cognitive-behavioral therapies
effective with children? In the remainder of this chapter, we explore these
questions. We begin with a discussion of cognitive-behavioral interventions

308

GILLHAM ET AL.

that have been used to treat helplessness and depression in children, including reattribution training and CBT. We then discuss the Penn Optimism
Program, designed to prevent depressive symptoms in middle school children.
Finally, we conclude with some speculations about other methods through
which optimism and an optimistic explanatory style can be promoted, especially in younger children.

Attribution Retraining
Attribution retraining is a technique that has been used primarily to
improve students persistence following failure on academic tasks (Craske,
1985; Dweck, 1975; Fowler & Peterson, 1981; for reviews see Cecil &
Medway, 1986; Forsterling, 1985). In the typical intervention, children are
taught to attribute failure to lack of effort (an unstable and specific-or
more optimistic-cause) rather than to lack of ability (a stable and globalor more pessimistic-cause). In some studies, adults model the new explanations, and in others, adults correct the pessimistic explanations and reward
the optimistic explanations that children make. Attribution retraining has
been found to increase childrens persistence on math problems (Craske,
1988; Okolo, 1992) and on reading assignments (Fowler & Peterson, 1981).
Attribution retraining has also been extended to other types of difficulties.
For example, Aydin (1988) used attribution retraining to improve social
competence in children. Children in the intervention group were taught
to attribute social failures to a lack of effort, and members of this intervention
group improved significantly in explanatory style and acceptance by peers.
In contrast, there was no change in explanatory style or peer acceptance
in members of a control group. Attribution retraining has also been used
to improve athletic performance in adolescents (e.g., Miserandino, 1998).
Most attribution retraining interventions are much less cognitively
demanding than the cognitive-restructuring component of CBT. Studies
using this technique typically teach children to substitute one attribution
(I didnt try hard enough) for another (I dont have the ability). Children
are not explicitly taught to rigorously evaluate the accuracy of their interpretations of events. One advantage of attribution retraining is that it may be
accessible to younger children. It has been used with 8- and 9-year-olds
(e.g., Dweck, 1975; Fowler & Peterson, 1981) and with children with mental
retardation (Turner, Dofney, & Dutka, 1997). The technique may be fragile,
however, because new beliefs are not necessarily more realistic than the
ones they replace. Optimistic attributions may be hard to maintain if children
are repeatedly confronted with evidence indicating these attributions are
not accurate. Little is known about the long-term effects of attribution
retraining programs or their effects on depressive symptoms.

BUILDING OPTIMISM AND PREVENTING DEPRESSIVE SYMPTOMS

309

Cognitive-Behavioral Therapy
The cognitive restructuring component of CBT is quite demanding
and requires metacognition (thinking about thinking), consideration of
alternative possibilities, and hypothesis testing-abilities that characterize
formal operations (adolescent and adult thought; Inhelder & Piaget, 1958;
Keating, 1980). It is perhaps not surprising then that CBT has been found
to be effective in treating and preventing adolescent depression (e.g., Clarke
et al., 1995; Clarke, Rohde, Lewinsohn, Hops, & Seeley, 1999; for reviews
see Gillham, ShattC, & Freres, in press; Kaslow & Thompson, 1998). For
example, Clarke and colleagues found that an intervention based on CBT
significantly prevented depressive symptoms and depressive episodes in a
sample of adolescents at risk for depression. It is impressive that the prevention effects lasted through 12 months of follow-up (Clarke et al., 1995).
Recent studies suggest that CBT also may be effective in reducing
depressive symptoms in middle school age children (see Kaslow & Thompson, 1998). It is important to note that interventions used with younger
children frequently place less emphasis on cognitive-restructuring techniques. Thus, positive effects of some interventions may be due to the
acquisition of behavioral, rather than of cognitive, skills. Researchers need
to evaluate the effectiveness of the individual components of these programs.
Penn Optimism Program
Given the evidence that depressive symptoms increase over the middle
school years and then rise dramatically in high school (see Hankin et al.,
1998; Rutter, 1986), we were interested in investigating whether CBT can
prevent depressive symptoms before children reach high school. With this
goal in mind, we developed the Penn Optimism Program (POP), a schoolbased program based on adult CBT. POP is a 12-session, manualized intervention. Children participate in POP in groups of approximately 8-12
students, either after school or during the school day. Groups are led by
school teachers, guidance counselors, and advanced graduate students in
clinical psychology. The program is described briefly below (for more detailed
descriptions, see Seligman et al., 1995; ShattC, Gillham, & Reivich, in press).

Cognitive Restructuring

As in adult CBT, POP involves both cognitive restructuring and skill

acquisition. Middle school students are (at least in theory) at the cusp of
formal operations and are not as skilled as adolescents at tasks that require
hypothesis testing and evaluation of their own thinking. Therefore, a variety
of techniques are used to make the cognitive-restructuring techniques acces310

GILLHAM ET AL.

sible and engaging. Skits, stories, role-plays and cartoons are used to introduce and emphasize concepts.
The cognitive-restructuring component begins with the cognitive
model of emotion and behavior (Beck, 1976; Ellis, 1962). Students learn
that feelings and behaviors are due in large part to the way they think about
events that happen to them. The link between thoughts and feelings is
introduced using cartoons. For example, in one set of worksheets, students
are asked to generate the beliefs that would lead to a given behavioral or
emotional consequence (see Figure 14.1). For homework, they write about
situations in their own lives, using the cartoon format to illustrate their
beliefs and the resulting feelings and behaviors.
POP then teaches children about optimistic versus pessimistic styles
of interpreting events. Skits are used to illustrate the types of thoughts that
characterize pessimistic and optimistic styles, and the consequences of these
styles. For example, in one set of skits two characters, Gloomy Greg and
Hopeful Holly, attend a school dance. Both children ask peers to dance
with them, and both are rejected. However, they respond quite differently
to this rejection. Greg attributes the event to stable, and global causes (Im
a loser, No one likes me) and makes pessimistic predictions (e.g., No
one will ever dance with me, Ill never have any fun at dances). Consequently, he feels dejected and hopeless. He sits down on the bench for the
rest of the evening and has a miserable time. Holly, in contrast, attributes
the event to causes that are more unstable and specific (e.g., Maybe I

Figure 14.1. The link between thoughts and feelings.

BUILDING OPTIMISM AND PREVENTING DEPRESSIVE SYMPTOMS

31 1

wasnt polite enough, Maybe he promised someone else this dance,

Maybe that boy doesnt like me). The predictions she makes are more
hopeful. She asks other peers to dance and, eventually, finds a partner and
has fun at the party. In POP, children learn that pessimistic styles can make
them feel sad and give up on goals, even when these goals are, in fact,
achievable. Cartoon worksheets are also used to give children practice with
the concept of interpretive styles. Children are given cartoons that depict
pessimistic interpretations and are instructed to rewrite the cartoons with
more optimistic attributions and predictions.
Although we focus on explanatory style and optimism, POPSgoal is
not simply to teach positive thinking. The goal is realistic thinking. We
have found that many children are reflexive pessimists; they habitually
interpret events pessimistically and often this pessimism seems to exceed
what is warranted given the situation. The third step in the cognitiverestructuring component, therefore, is to teach children a variety of techniques for thinking accurately about problems. Children learn to evaluate
their beliefs by considering evidence for and against them. They learn to
generate alternative interpretations, to evaluate these alternatives, and to
identify those interpretations that are most accurate. We use a detective
analogy to illustrate this process. Problems are compared to crimes. A good
detective, like Sherlock Holmes, solves a crime by listing a variety of suspects
(initial belief plus alternatives) and looking for clues (evaluates evidence)
for each one. Bad detectives, in contrast, conclude that the first suspect
they consider (initial belief) must be guilty, without evaluating the evidence.
Once they understand the process, children practice this skill frequently
during sessions and for homework over the course of the program.
Middle school and high school students with depressive symptoms often
catastrophize about problems. That is, they believe that terrible outcomes are
likely to occur following negative events and that these outcomes will
endure. For example, after an embarrassing experience at school, a child
may become convinced that she will be ridiculed by most of her peers for
the remainder of the school year. She may imagine that even her closest
friends will come to reject her, and that she will have no one to talk to or
socialize with when she begins high school next year. This type of thinking
process leads to considerable anxiety and hopelessness. The story of Chicken
Little is useful for illustrating the consequences of catastrophic thinking,
and Dont be a Chicken Little is a slogan that is used throughout the
program. T o counter catastrophic thinking, participants are taught to consider other possible outcomes of the situation. In the worst-best-most likely
procedure, children first list the disastrous outcomes they fear. Then, they
are encouraged to list the best possible outcomes they can imagine for the
situation. These best outcomes are often equally unlikely and equally extreme. For example, a fantastic outcome for the above situation may be

312

GILLHAM ET AL.

that other children in the school feel sympathetic toward the girl and, as
a result, invite her to many parties. She becomes the most popular girl in
the school and enters high school with everyone fighting to be her friend.
Considering the positive extreme helps children realize that the disasters
they imagine are also unlikely to occur. The next step in this process is for
children to list several outcomes that are more likely, to evaluate these
outcomes, and to select those that seem most realistic given the situation
(e.g., Several of the girls peers will tease her over the next week or so).
Once they have identified the most likely outcomes, they are taught how
to plan for and deal with the negative aspects of these outcomes.

Skill Acquisition
Although cognitive skills provide hope that problems will dissipate
and that change is possible, they do not provide the skills required to effect
that change. The second half of the program teaches skills that can be
used to solve a variety of problems, including emotion-control techniques,
relaxation, assertiveness and negotiation techniques, and a method for
countering procrastination. In addition, students learn an approach to problem solving that involves generating a variety of possible solutions for
problems they encounter and then evaluating these solutions and deciding
between them. This process may be especially helpful in counteracting
hopelessness, because children learn that usually there are a variety of paths
through which positive outcomes can be achieved. As with the cognitive
component, techniques are introduced through skits, stories, and cartoon
characters (e.g., Say-it-straight Samantha) that illustrate specific skills.
Students are encouraged to role-play possible solutions in the group and to
practice the skill for homework between sessions.
Research on POP

In our first study of POP, fifth and sixth graders were identified as atrisk for depression based on reports of parental conflict and current depressive
symptoms. The 69 children who participated in the intervention were compared with a matched control group of children. Initial results indicated
that POP significantly improved explanatory style and reduced depressive
symptoms (Jaycox, Reivich, Gillham, 6r Seligman, 1994). These effects
endured through a 2-year follow-up period (Gillham, Reivich, Jaycox, &
Seligman, 1995). There was a dramatic reduction in reports of moderateto-severe levels of depressive symptoms across the 2-year follow-up period.
For example, at the 2-year follow-up, 44% of control participants reported
moderate-to-severe levels of symptoms as compared with only 22% of
POP participants. Mediational analyses suggested that improvements in
BUILDING OPTIMISM AND PREVENTING DEPRESSIVE SYMPTOMS

313

explanatory style were, in part, responsible for the program's effect on

depressive symptoms. It is important to note that data from 2%- and 3-year
follow-ups indicate the prevention effect for depressive symptoms was no
longer significant (Gillham 6r Reivich, 1999b), suggesting that booster
sessions may be important for long-term prevention of depression.
This study had several methodological limitations, including the lack
of random assignment to condition and reliance on self-report measures. In
addition, there was no test of treatment specificity. Thus, it is possible that
POP'S effects were due to social support or other factors not specific to CBT.
Research that addresses these limitations is currently underway. Gillham
and Seligman (1999) recently completed a study in which children were
randomly assigned to POP or a control group. The children participating
in this study were followed for 12 months after POP groups ended. POP
participants reported improvements in explanatory style and depressive
symptoms following the intervention. There was some evidence that POP
prevented depressive symptoms over the follow-up period. Although average
levels of depressive symptoms were not significantly different at 6 or 12
months, peak levels of symptoms across the follow-up were significantly
lower in POP participants than in participants of the control group. Current
studies are underway to investigate POP'S effectiveness relative to other
interventions, the effects of POP on children's behavior as observed by
parents and teachers, and POP'S ability to prevent clinical diagnoses of
depression.
Building the Foundation for an Optimistic Explanatory Style

CBT offers promise as a means of boosting optimism and explanatory

style, and as a means of preventing depression in late childhood and adolescence. However, there are many other possible paths through which optimistic interpretive styles can be promoted, and these may be effective for young
children as well. If children learn optimism and explanatory style from
adults, a powerful intervention may be to boost explanatory style in parents
and teachers. Children's explanatory style may benefit when their depressed
parents participate in CBT, for example, or when their teachers are educated
about the effects of attributing failure to lack of ability. If uncontrollable
stressors breed pessimism, therapeutic and societal interventions that target
violence, abuse, parental psychopathology, and poverty may promote optimistic styles in children. If repeated failure leads children to attribute events
to permanent causes and to expect continued failure, they may benefit from
interventions that improve their academic, athletic, artistic, social, and
other skills. Finally, providing children with control and mastery experiences
may help lay the foundation for an optimistic interpretive style. By facing
challenges, struggling, and, eventually, succeeding, children may learn to

3 14

GILLHAM ET a.

view even large problems as temporary and changeable. All of these paths
should increase childrens resiliency.

SUMMARY AND CONCLUSION

In summary, analysis of recent research suggests that depressive symptoms and depressive disorders are common during middle school and rise
dramatically during the high school years (Angold, 1988; Hankin et al.,
1998; Lewinsohn et al., 1993).Cognitive vulnerabilities, such as a pessimistic
explanatory style or dispositional pessimism, that are implicated in adult
depression are relevant to childhood depression at least by the end of
elementary school or beginning of middle school. Our interpretation of
these findings suggests that the middle school years offer an important
window for intervention programs. We are encouraged by recent research
indicating that CBT techniques effective in treating adult depression can
also be used to foster an optimistic explanatory style and prevent depressive
symptoms in children. Despite these findings, researchers still know very
little about the origins of optimistic interpretive styles or how to build these
styles in children. I t is likely that there are many pathways through which
optimistic styles can be fostered in children. Discovering these pathways is
an important and exciting area for future research.

REFERENCES
Abramson, L. Y., Metalsky, G. I., & Alloy, L. B. (1989). Hopelessness depression:
A theory-based subtype of depression. Psychological Review, 96, 358-372.
Anderson, J., Williams, S., McGee, R., & Silva, P. A. (1987). The prevalence of
DSM-111diagnosesin a large sample of pre-adolescent children from the general
population. Archives of General Psychiatry, 44, 69-76.
Angold, A. ( 1988). Childhood and adolescent depression: I. Epidemiological and
aetiological aspects. British Journal of Psychiatry, 152, 601-617.
Aydin, G. (1988). The remediation of childrens helpless explanatory style and
related unpopularity. Cognitive Therapy and Research, 12, 155-165.
Beck, A. T. (1976). Cognitive therapy and the emotional disorders. New York: Intemational Universities Press.
Beck, A. T., Hollon, S. D., Young, J. E., Bedrosian, R. C., & Budenz, D. (1985).
Treatment of depression with cognitive therapy and amitriptyline. Archives of
General Psychiatry, 42, 142-148.
Bemporad, J. R. (1994). Dynamic and interpersonal theories of depression. In
W. M. Reynolds & H. F. Johnson (Eds.),Handbook of depression in children and
adolescents (pp. 81-95). New York: Plenum Press.
BUILDING OPTIMISM AND PREVENTING DEPRESSIVE SYMPTOMS

315

Cecil, M. A., & Medway, F. J. (1986). Attribution retraining with low achieving
and learned helpless children. Techniques, 2, 173-1 8 1.
Centers for Disease Control. (1991). Attempted suicide among high school students-United States. 1990. Morbidity and Mortality Weekly Report, 40, 633635.
Cicchetti, D., & Schneider-Rosen,K. (1986). An organizational approach to childhood depression. In M. Rutter, C. E. Izard, & P. B. Read (Eds.), Depression
in young people: Developmental and clinical perspectives (pp. 71-134). New York:
Guilford Press.
Cicchetti, D., & Toth, S. L. (1998). The development of depression in children
and adolescents. American Psychologist, 53, 221-242.
Clarke, G. N., Hawkins, W., Murphy, M., Sheeber, L., Lewinsohn, P. M., & Seeley,
J. (1995). Targeted prevention of unipolar depressive disorder in an at-risk
sample of high school adolescents: A randomized trial of a group cognitive
intervention.Journu1of the American Acudemy of Child and Adolescent Psychiatry,
34(3), 312-321.
Clarke, G. N., Rohde, P., Lewinsohn, P. M., Hops, H., & Seeley, J. R. (1999).
Cognitive-behavioral treatment of adolescent depression: Efficacy of acute
group treatment and booster sessions.Journal of the American Acudemy of Child
and Adolescent Psychiatry, 38, 272-279.
Craske, M. ( 1985). Improving persistence through observational learning and attribution retraining. British Journul of Educational Psychology, 55, 138-147.
Craske, M. (1988). Learned helplessness, self-worth motivation and attribution
retraining for primary school children. British Journal of Educational Psychology,
58, 152-164.
Damon, W., & Hart, D. (1988). Self-understandingin childhood and adolescence. New
York: Cambridge University Press.
DeRubeis, R. J., & Hollon, S. D. (1995). Explanatory style in the treatment of
depression. In G. M. Buchanan & M. E. P. Seligman (Eds.), Explanatory style
(pp. 99-1 12). Hillsdale, NJ: Erlbaum.
Downey, G., & Coyne, J. C. (1990). Children of depressed parents: An integrative
review. Psychological Bulletin, 108, 50-76.
Dweck, C. S. (1975). The role of expectations and attributions in the alleviation
of learned helplessness.Journalof Personality and Social Psychology, 31,674-685.
Dweck, C. S., Davidson, W., Nelson, S., & Enna, B. (1978). Sex differences
in learned helplessness: 11. The contingencies of evaluative feedback in the
classroom, and 111. An experimental analysis. Developmental Psychology, 14,
268-276.
Dweck, C. S., & Leggett, E. L. (1988). A social-cognitive approach to motivation
and personality. Psychological Review, 95, 256-273.
Ellis, A. (1962). Reason and emotion in psychotherapy. New York: Lyle Stuart.

316

GILLHAM ET AL.

Evans, M. D., Hollon, S. D., DeRubeis, R. J., & Piasecki, J. M. (1992). Differential
relapse following cognitive therapy and pharmacotherapy for depression. Archives of General Psychiacry, 49, 802-808.
Forsterling, F. (1985). Attribution retraining: A review. Psychological Bulletin, 98,
495-512.
Fowler, J. W., & Peterson, P. L. (1981). Increasing reading persistence and altering
attributional style of learned helpless children. Journul of Educational Psychology,
73, 251-260.
Frey, K. S., & Ruble, D. N. (1987). What children say about classroom performance:
Sex and grade differences in perceived competence. Child Development, 58,
1066- 1078.
Frey, K. S., &Ruble, D. N. (1990). Strategies for comparative evaluation: Maintaining a sense of competence across the life span. In R. J. Sternberg & J. Kolligian
(Eds.), Competence considered (pp. 167-189). New Haven, CT:Yale University Press.
Garber, J., & Flynn, C. (1998). Origins of depressive cognitive style. In D. K. Routh
& R. J. DeRubeis (Eds.), The science of clinical psychology: Accomplishments and
future directions (pp. 53-93). Washington, DC: American Psychological Association.
Garber, J., Quiggle, N., & Shanley, N. (1990). Cognitions and depression in children
and adolescents. In R. E. Ingram (Ed.), Contemporary psychological approaches
to depression (pp. 87-115). New York: Plenum Press.
Garber, J., & Robinson, N. S. (1997). Cognitive vulnerability in children at risk
for depression. Cognition and Emotion, 11, 619-635.
Garber, J., Weiss, B., & Shanley, N. (1993). Cognitions, depressive symptoms, and
development in adolescents. Journal of Abnormal Psychology, 102, 47-57.
Garrison, C. Z., Schluchter, M. D., Schoenbach, V. J., & Kaplan, B. K. (1989).
Epidemiology of depressive symptoms in young adolescents. Journal of the
American Academy of Child and Adolescent Psychiatry, 28, 343-35 1.
Gillham, J. E., & Reivich, K. J. (1999a). Depressive symptoms through the middle
school years. Unpublished manuscript, University of Pennsylvania, Philadelphia.
Gillham, J. E., & Reivich, K. J. (199913). Prevention of depressive symptoms in
school children: A research update. Psychological Science, 10, 461-462.
Gillham, J. E., Reivich, K. J., Jaycox, L. H., & Seligman, M. E. P. (1995). Prevention
of depressive symptoms in schoolchildren: Two-year follow-up. Psychological
Science, 6, 343-351.
Gillham, J. E., & Seligman, M. E. P. (1999). School-based prevention of depressive
symptoms in children. Does including parents help? Unpublished manuscript,
University of Pennsylvania, Philadelphia.
Gillham, J. E., Shattk, A. J., & Freres, D. R. (2000). Preventing depression: A
review of cognitive-behavioral and family interventions. Applied and Preventive
Psychology, 9 63-88.
BUILDING OPTIMISM AND PREVENTING DEPRESSIVE SYMPTOMS

317

Gladstone, T. R. G., & Kaslow, N. J. (1995). Depression and attributions in children

and adolescents: A meta-analytic review. Journal of Abnormal Child Psychology,
23, 597-606.
Gold, E. R. (1986). Long-term effects of sexual victimization in childhood: An
attributional approach. Journal of Consulting and Clinical Psychology, 54, 471475.
Hammen, C., Adrian, C., & Hiroto, D. (1988). A longitudinal test of the attributional vulnerability model in children at risk for depression. British Journal of
Clinical Psychology, 27, 37-46.
Hankin, B. L., Abramson, L. Y., Moffitt, T. E., Silva, P. A., & McGee, R. (1998).
Development of depression from preadolescence to young adulthood: Emerging
gender differences in a 10-year longitudinal study. Journal of A b n d Psychology, 107, 128-140.
Harter, S. ( 1990). Developmental differences in the nature of self-representations:
Implications for the understanding, assessment, and treatment of maladaptive
behavior. Cognitive Therapy and Research, 14, 113-142.
Ilardi, S. S., Craighead, E. W., & Evans, D. D. (1997). Modeling relapse in unipolar
depression: The effects of dysfunctional cognitions and personality disorders.
Journal of Consulting and Clinical Psychology, 65, 381-391.
Inhelder, B., & Piaget, J. (1958). The growth of logical thinking from childhood to
adolescence. New York: Basic Books.
Jaycox, L. H., Reivich, K. J., Gillham, J. E., & Seligman, M. E. P. (1994). Prevention
of depressive symptoms in school children. Behaviour Research and Therapy,
32, 801-816.
Joiner, T. E., & Wagner, K. D. (1995a). Attributional style and depression in
children and adolescents: A meta-analytic review. Clinical Psychology Review,
15, 777-798.
Joiner, T. E., & Wagner, K. D. (1995b). Parental, child-centered attributions
and outcome: A meta-analytic review with conceptual and methodological
implications. Journal of Abnormal Child Psychology, 24, 37-52.
Kashani, J. H., Rosenberg, R. K., & Reid, J. C. (1989). Developmental perspectives
in child and adolescent depressive symptoms in a community sample. American
Journal of Psychiatry, 141, 1397-1402.
Kaslow, N. J., Brown, R. T., & Mee, L. L. (1994). Cognitive and behavioral
correlates of childhood depression: A developmental perspective. In W. M.
Reynolds & H. F. Johnston (Eds.), Handbook of depression in children and
adolescents: issues in clinical child psychology (pp. 97-121). New York: Plenum
Press.
Kaslow, N. J., & Thompson, M. P. (1998). Applying the criteria for empirically
supported treatments to studies of psychosocial interventions for child and
adolescent depression. Journal of Clinical Child Psychology, 27, 146-1 55.
Kaufman, J. (1991). Depressive disorders in maltreated children. Journal of the
American Academy of Child and Adolescent Psychiatry, 30, 257-265.

318

GILLHAM ET AL..

Keating, D. P. (1980). Thinking process in adolescence. In J. Adelson (Ed.), Handbook of adokscent psychology (pp. 21 1-246). New York: Wiley & Sons.
Kovacs, M. ( I 989). Affective disorders in children and adolescents. American Psychologist, 44, 209-215.
Langer, E. J., & Park, K. (1990). Incompetence: A conceptual reconsideration. In
R. J. Stemberg & J. Kolligian (Eds.), Competence considered (pp. 149-166).
New Haven, CT:Yale University Press.
Lewinsohn, P. M., Hops, H., Roberts, R., & Seeley,J. (1993). Adolescent psychopathology: I. Prevalence and incidence of depression and other DSM-111-R disorders in high school students. Journal of Abnormal Psychology, 102, 110-120.
McCauley, E., Mitchell, J. R., Burke, P., & Moss, S. (1988). Cognitive attributes
of depression in children and adolescents. Journal of Consulting and Clinical
Psycholoa, 56, 903-908.
McConville, B. J., Boag, L. C., & Purohit, A. P. (1973). Three types of childhood
depression. Canadian Psychiatric Association Journal, 18, 133-137.
Miserandino, M. (1998). Attribution retraining as a method of improving athletic
performance. Journal of Spurt Behavior, 21, 286-297.
Nolen-Hoeksema,S. ( 1986).Developmental studies of explanatory style and learned
helplessness in children. Unpublished doctoral dissertation, University of
Pennsylvania, Philadelphia.
Nolen-Hoeksema, S., Girgus, J. S., & Seligman, M. E. P. (1986). Depression in
children of families in turmoil. Unpublished manuscript,University of Pennsylvania, Philadelphia.
Nolen-Hoeksema, S., Girgus, J. S., & Seligman, M. E. P. (1992). Predictors and
consequences of childhood depressive symptoms: A five-year longitudinal
study. Journal of Abnormal Psychology, 101, 405-422.
Okolo, C. M. (1992). The effects of computer-based attribution retraining on
the attributions, persistence, and mathematics computation of students with
learning disabilities. Journal of Learning Disabilities, 25, 327-334.
Piaget, J. (1926). Judgment and reasoning in the child. New York: Harcourt, Brace
& World.
Piaget, J. (1959). The language and thought of the chitd. New York: Basic Books.
Pintrich, P. R., & Blumenfeld, P. C. (1985). Classroom experience and childrens
self-perceptionsof ability, effort, and conduct. Journal of Educational Psychology,
77, 646-657.
Robins, C. I., & Hayes, A. M. (1995). The role of causal attributions in the
prediction of depression. In G. M. Buchanan & M. E. P. Seligman (Eds.),
Expwplanatory styk (pp. 71-98). Hillsdale, NJ: Erlbaum.
Ruble, D. N.?Boggiano, A. K., Feldman, N. S., & Loebl, J. H. ( 1980).Developmental
analysis of the role of social comparison in self-evaluation. Developmental
Psychology, 16, 105-115.
BUlLDlNG OPTIMISM AND PREVENTING DEPRESSIVE SYMPTOMS

319

Rutter, M. (1986). The developmental psychopathology of depression. In M. Rutter,

C. E. Izard, & P. B. Read (Eds.), Depression in young people: Developmental and
clinical perspectives (pp. 3-30). New York: Guilford Press.
Scheier, M. F., & Carver, C. S. (1992). Effects of optimism on psychological
and physical well-being: Theoretical overview and empirical update. Cognitive
Therapy and Research, 16, 201-228.
Seligman, M. E. P. (1975). Helplessness: On depression, development, and death.
San Francisco: W. H. Freeman & Co.
Seligman, M. E. P., Castellon, C., Cacciola, J., Schulman, P., Luborsky, L., Ollove,
M., & Downing, R. (1988). Explanatory style change during cognitive therapy
for unipolar depression. Journal of Abnormal Psychology, 97, 13-18.
Seligman, M. E. P., Reivich, K. J., Jaycox, L. H., & Gillham, J. (1995). The optimistic
child. New York: Houghton-Mifflin.
Seligman, M. E. P., Schulman, P., DeRubeis, R. J., 61 Hollon, S. D. (1999).
The prevention of depression and anxiety. Prevention B Treatment, 2, 8,
journals.apa.org.
Shattk, A. J., Gillham, J. E., & Reivich, K. J. (in press). Promoting hope in children
and adolescents. In J. E. Gillham (Ed.), The science of optimism and hope: Essays
in honor of Martin E. P. Seligman. Radnor, PA: Templeton Foundation Press.
Turner, L. A., Dofny, E. M., & Dutka, S. (1997). Effect of strategy and attribution
training on strategy maintenance and transfer. American Journal on Mental
Retardation, 98, 445-454.

3 20

GILLHAM ET AL.

15
OPTIMISM, PESSIMISM, AND
PSYCHOTHERAPY: IMPLICATIONS
FOR CLINICAL PRACTICE
JAMES L. PRETZER AND CHAILLE A. WALSH

Art, a 43-year-old engineer who has never had a long-term romantic

relationship, is anticipating an upcoming date. In discussinghis anticipations with his therapist (the first author) he reports thinking, What
will I have to say? Shell think Im boring. I dont have anything
to offer. He feels anxious, discouraged, and depressed. When he has
experienced these thoughts and feelings in the past he has typically
postponed, canceled, or avoided social interactions, and this avoidance
has prolonged his social anxiety and isolation.

Art is quite pessimistic regarding his upcoming date. In fact, his thinking regarding a wide range of social interactions is consistently pessimistic.
This pessimism plays a n important role in perpetuating Arts problems with
relationships. Because he anticipates that social interactions will go badly
and will accomplish little, he usually fails to initiate them and thus remains
isolated. In addition, the growing body of research on the impact of optimism
and pessimism in many areas of life (as discussed throughout this volume)

32 1

suggests that Arts pessimism may have effects that extend far beyond his
love life. Should Arts therapist focus narrowly on reducing his social anxiety
and avoidance, address his pessimism as well, or is it more important to
address his pessimism than to focus on specific concerns about social interactions? If his therapist chooses to address Arts pessimism to some extent,
how can this be accomplished? How will the therapist effectively modify
Arts pessimistic outlook? This chapter will address the question of whether
it is useful to modify adults positive and negative expectancies and will
present a cognitive-behavioral approach to increasing optimism.
A number of cognitive-behavioral approaches to psychotherapy have
emphasized the importance of negative thoughts in understanding and treating many forms of psychopathology. In particular, Becks Cognitive Therapy
(A. T. Beck, Rush, Shaw, & Emery, 1979) is known for its emphasis on
identifying and modifying negative thoughts and has received substantial
empirical support (see DeRubeis & Crits-Cristoph, 1998).Seligman (1990)
argues that the psychotherapeutic interventions used in cognitive therapy
(CT) directly modify optimism and pessimism and asserts that CT is effective
specifically because it decreases pessimism and increases optimism. Thus,
the cognitive perspective on psychotherapy seems especially relevant.

OPTIMISM, PESSIMISM, AND PSYCHOPATHOLOGY:

A COGNITIVE PERSPECTIVE
The Role of Negative Thinking in Psychopathology
Consider our case example. Art was thinking, What will I have to
say? Shell think Im boring. I dont have anything to offer. He was feeling
anxious, discouraged, and depressed, and he was inclined to avoid his upcoming date. How are we to understand Arts pessimistic outlook and its impact
on his mood and his behavior?
The cognitive view of human functioning (Beck et al., 1979) emphasizes three aspects of cognition (see Figure 15.1).An individuals automatic
thoughts, his or her immediate, spontaneous appraisal of the situation, are
seen as playing a central role in eliciting and shaping the individuals
emotional and behavioral response to a situation. CT asserts that individuals
automatically evaluate the many situations they encounter or anticipate
A variety of cognitive and cognitive-behavioral approaches to psychotherapy have been developed
in recent years. Although these approaches have much in common, there are important conceptual
and technical differences among them. To minimize confusion, the specihc approach developed by
Aaron T.Beck and his colleagues will be referred to as cognitive therapy (CT),whereas the term
cognitive-behvioral will be used as a collective term to refer to the full range of cognitive and
cognitive-behavioral approaches.

322

PRETZER A N D WALSH

Figure 15.1. The basic cognitive model.

throughout the day and that the content of these evaluations has an important impact on their emotional and behavioral responses. If the individuals
evaluation of the situation is realistic, his or her emotional and behavioral
responses are likely to prove adaptive. If the evaluation of the situation is
unrealistic, his or her emotional and behavioral responses are likely to prove
dysfunctional.
Humans do not start from scratch in perceiving and interpreting
each event or situation they encounter. Rather, in each situation, the
individuals response is influenced by sets of interrelated concepts, beliefs,
and assumptions, which CT therapists refer to as schemas. As the individual
proceeds through daily life, relevant schemas are automatically activated
and used in responding to the situations the individual encounters. For
OPTIMISM, PESSIMISM, AND PSYCHOTHERAPY

323

example, imagine that Art is proceeding through his busy day as an engineer.
As he works on a new project, consults a reference book, or goes for a cup
of coffee, his beliefs and assumptions about relationships are inactive because
they are not relevant to the situation. Thus they have little impact on his
mood or behavior. However, as he approaches the coffee pot and sees that
a female colleague is standing there, his schemas related to male-female
interactions are activated automatically and have an immediate impact.
Arts schema for male-female interactions includes an assumption that
women are repelled by boring men and are attracted to cool men. It
also includes an unspoken assumption that he is boring. Thus, when this
schema was activated in real-life situations, Art immediately attempted to
appear cool, to monitor whether he was being successful in these attempts,
and to monitor whether the woman seemed bored.
Because Art was alert for signs of disinterest or boredom, he was quick
to recognize those occasions when others did, indeed, find him boring and,
because he was not equally vigilant for times when others enjoyed their
interactions with him, he developed a biased view of their reactions to him.
Furthermore, when Art encountered a n occasion on which a woman showed
clear evidence of enjoying his company, he was inclined to discount the
experience (i.e., They dont really mean it, Theyre only saying that to
be nice, or If they get to know me better theyll find out that Im not as
interesting as they think.) because this experience was inconsistent with
his assumptions. In short, Arts vigilance for experiences that appeared to
confirm his preconceptions and his tendency to discount experiences that
were inconsistent with his worldview tended to confirm and perpetuate his
assumptions about himself and others.
In addition to their beliefs and assumptions about the way things
are, individuals also hold beliefs and assumptions about how one should
deal with various situations that might be termed interpersonal strategies. For
example, Art believed that the way to be attractive to women was to act
cool (i.e., to be sure that he did not appear anxious and to be careful not
to show too much interest in the woman). His attempts to put this belief
into practice had a major impact on his interpersonal relationships. Not
only did his careful attempts to act cool result in his appearing stiff and
awkward, but his self-consciousness also raised his level of anxiety and made
it increasingly difficult for him to act cool. Because concealing any sign of
anxiety was an important part of Arts strategy for appearing cool, he feared
that it would become apparent to others and defeat his attempts. If he was
not confident that he could conceal his anxiety, Art would become flustered and conclude that he had failed in his efforts to act cool. He had
experienced many occasions on which his strategy of attempting to act cool
had failed; however, he consistently attributed his failures to having not
acted cool enough and, therefore, resolved to try harder the next time. It

324

PRETZER AND WALSH

did not appear that he ever considered the possibility that this might be
an ineffective strategy. (For a more detailed discussion of the role of schemas
in psychopathology, see Pretzer 6r Beck, 1996.)
Another aspect of cognition that can contribute to persistent misperceptions of situations are systematic errors in reasoning, termed cognitive
distortions. Human beings are prone to a variety of errors in logic, which
can contribute to misinterpretations of events and amplify the impact of
schemas (see Table 15.1). For example, Art manifested a strong tendency
to focus selectively on his faults and shortcomings, and he tended to overlook
his good points. This tendency strongly reinforced his negative outlook.
The cognitive view is that, given the complexity of daily life and the
ambiguity of many interpersonal interactions, occasional misperceptions and
misinterpretations of events are inevitable. However, isolated misperceptions
and misinterpretations give rise to isolated maladaptive responses, which
are easily corrected by subsequent experiences. For seriously maladaptive
responses to develop, a systematic bias in perception, recall, or interpretation
would be required. This would result in more persistent maladaptive responses than would result from normal misperceptions and misinterpretations. If it also distorted the feedback process, either by strongly biasing the
interpretation of events or by influencing the responses of others, it could
result in very persistent maladaptive responses.
Since the individuals schemas, beliefs, and assumptions have a major
impact on the perception, recall, and interpretation of events, they are one
possible source of such a systematic bias. However, they are not the only
possible source of a systematic bias in the perception, recall, and interpretation of events. Despite CTs name, the model is not exclusively cognitive.
Rather, the cognitive model focuses on the interplay between cognition,
affect, and behavior in psychopathology (see Figure 15.2). The effect of Arts
beliefs and assumptions on his perception of events and on his interpersonal
behavior was discussed above, but the cycle does not end with the effects
of cognition on behavior. A persons interpersonal behavior influences the
responses of others, and their responses can, in turn, result in experiences
that can influence that individuals beliefs and assumptions. For example,
a number of Arts friends and family members responded to his anxiety,
pessimism, and lack of success in relationships with advice that reinforced
his tendency to attribute failures to inadequacies on his part. This advice
also reinforced his assumption that he had to try to act cool (e.g., Dont
be so nervous, it will drive women away, You have to be more confident,
or Dont say that, say
instead.).
Although the cognitive model assumes that the individuals interpretation of events shapes his or her emotional response to the situation, CT
also argues that the individuals emotional state has important effects on
cognition. A large body of research has demonstrated that affect tends to
~

OPTIMISM, PESSIMISM, AND PSYCHOTHERAPY

325

Table 15.1.
Common Cognitive Distortions
Distortion

Definition
~~

Dichotomous
thinking

Viewing experiences in terms of two mutually exclusive categories

with no shades of gray in between. For example, believing that
one is either a success or a failure and that anything short of a
perfect performance is a total failure.

Overgeneralization

Perceiving a particular event as being characteristic of life in general rather than as being one event among many. For example,
concluding that an inconsiderate response from ones spouse
shows that she doesnt care despite her having showed consideration on other occasions.

Selective abstraction

Focusing on one aspect of a complex situation to the exclusion of

other relevant aspects of that situation. For example, focusing on
the one negative comment in a performance evaluation received
at work and overlooking the positive comments.
Discounting positive experiences that would conflict with the individuals negative views. For example, rejecting positive feedback
from friends and colleagueson the grounds that theyre only saying
that to be nice, rather than consideringwhetherthe feedback could
be valid.
Assuming that one knows what others are thinking or how others
are reacting despite having little or no evidence. For example,
thinking, I just know he thought I was an idiot! despite the other
persons having given no apparent indication of his reactions.
Reacting as though expectations about future events are establishedfacts rather than recognizingthem as fears, hopes, or predictions. For example, thinking Hes leaving me, I just know it! and
acting as though this is definitely true.
Treating actual or anticipated negative events as intolerablecatastrophes ratherthan seeing them in perspective.For example, thinking, Oh my God, what if I faint! without considering that although
fainting may be unpleasant or embarrassing, it is not terribly dangerous.

Disqualifying the
positive

Mind reading

Fortune telling

Catastrophizing

Maximization and
minimization
Emotional reasoning
Should statements

Labeling
Personalization

326

Treating some aspects of the situation, personal characteristics,

or experiencesas trivial and others as very important, independent
of their actual significance. For example, thinking Sure, Im good
at my job, but so what, my parents dont respect me.
Assuming that ones emotional reactions necessarily reflect the
true situation. For example, concluding that because one feels
hopeless, the situation must really be hopeless.
The use of should and have to statements that are not actually
true to provide motivationor control over ones behavior. For example, thinking 1 shouldnt feel aggravated. Shes my mother, I have
to listen to her.
Attaching a global label to oneself rather than referring to specific
events or actions. For example, thinking Im a failure! rather than
Boy, I blew that one!
Assuming that one is the cause of a particular external event when,
in fact, other factors are responsible. For example, thinking, She
wasnt very friendly today, she must be mad at me, without considering that factors other than ones own behavior may be affecting
the other individuals mood.

PRETZER AND WALSH

Schemas

Automatic
Thoughts

Biased Perception
and Recall

_ _ _ _ _ ~ ~

Responses
of Others

Figure 15.2. The role of cognition in psychopathology.

influence both cognition and behavior in mood-congruent ways (Isen, 1984).

For example, even a mild, experimentally induced depressed mood biases
perception and recall in a depression-congruent way (see Watkins, Mathews,
Williamson, & Fuller, 1992 for a recent discussion). A depressed mood
increases the likelihood that the individual will focus on negative aspects
of the situation and preferentially recall negative past experiences. Although
this phenomenon has not been investigated extensively for most other
moods, it appears that many moods tend to bias perception and recall in a
mood-congruent way. For example, as a n individuals level of anxiety increases, attentional processes appear to be biased in favor of signs of threat
(Watkins et al., 1992). This phenomenon lays the foundation for a potentially self-perpetuating cycle where the individuals automatic thoughts elicit
a particular mood, the mood biases perception and recall in a mood-congruent way, and this increases the likelihood of additional mood-eliciting automatic thoughts, which elicit more of the mood in question, which further
biases perception and recall, and so on, until the cycle is interrupted.
Affect can play an important role in an individuals functioning in
another way as well. As Taylor and Rachman (1991, 1992) have noted,
individuals may fear certain emotions and may strive to avoid the emotion
itself; may seek to escape from experiencing the emotion as quickly as
possible; or may attempt to avoid thoughts, memories, or situations that
they expect to elicit the emotion. At the time when Art sought treatment,
he frequently avoided situations, thoughts, and memories that led to his
feeling anxious. In particular, he avoided interactions with women who he
found attractive because he anticipated feeling anxious and because he was
convinced that he would fail anyway. He also avoided thinking about social
OPTIMISM, PESSlMlSM, AND PSYCHOTHERAPY

327

interactions because simply thinking about a conversation or an upcoming

date elicited significant anxiety. This cognitive avoidance made it difficult
for him to think effectively about his problems and find ways of overcoming
them because he would have to tolerate anxiety in order to do so.
It is important to note that the cognitive model does not assert that
cognition causes psychopathology. Cognition is viewed as an important part
of the cycles through which human beings perceive and respond to events,
and thus, is viewed as having an important role in pathological responses
to events. As an important part of these cycles, cognition is also seen as a
promising point for intervention, but not as the sole cause of psychopathology
or as the sole point for effective intervention.

From Negative Thinking to Optimism or Pessimism

CT has traditionally focused on the impact of specific thoughts, beliefs,

and cognitive distortions rather than emphasizing broad biases in cognition
such as dispositional pessimism. Research into CT has focused on the question of whether it provides an effective treatment approach and whether
particular thoughts and assumptions contribute to psychopathology. Relatively little attention has been focused on the question of whether persistent
individual differences in optimism or pessimism are relevant to psychotherapy or the question of whether such individual differences can be modified.
For example, a recent volume that provides a comprehensive review of
research into cognitive factors that contribute to vulnerability to depression
does not explicitly discuss optimism and pessimism (Ingram, Miranda, &
Segal, 1998).
Only in recent years have a few investigators begun to consider optimism and pessimism from a cognitive or cognitive-behavioral perspective.
However, if one understands optimism and pessimism in terms of generalized
expectancies (Scheier & Carver, 1992); in terms of attributional style (Peterson & Seligman, 1984); or in terms of a consistent positive or negative
bias in the interpretation of events (Riskind, Sarampote, & Mercier, 1996),
it is not hard at all to conceptualize optimism and pessimism from a cognitive perspective.
One of the most widely known cognitive-behavioral perspectives on
optimism and pessimism has been developed by Seligman (1990). In his
research into what is termed Lamed hlpkssness, Seligman and his colleagues
demonstrated that laboratory animals could learn either to passively tolerate
negative events (electric shock) or to persistently seek to escape the electric
shock. In subsequently exploring this phenomenon in human beings, he
found that some individuals were quick to give up after a few negative
experiences, while others persisted despite recurrent failures. In explaining
this phenomenon, Seligman focused on the attributions these individuals
328

PRETZER AND WALSH

made. He argues that individuals who believed that nothing they could
ever do would change their situation became hopeless and depressed. Those
who believed that they could influence the outcome remained optimistic
and persisted in efforts to overcome misfortune.
From Seligmans perspective (1990), optimism and pessimism are a
matter of the explanatory style that the individual acquires over the course
of development. He emphasizes three aspects of individuals explanatory
style: personalization, pervasiveness, and permanence. Consider our case
example. In it, Art consistently tended to attribute failures in social interactions to internal causes (She doesnt like me because Im boring.); his
attributions regarding his failures were quite global (This is the way I am.);
and he attributed his failures to factors he saw as permanent (This is the
way I am going to be.). In theory, the results would be quite different for
an individual who had the same experience of rejection but made different
attributions. For example, one could attribute rejection to external causes
(perhaps I must not be her type, or I guess she isnt interested in dating.).
The attributions could be specific rather than general (Sues not interested,
maybe Mary would like to go out.). Or responsibility for the rejection could
be attributed to factors that are temporary rather than permanent (e.g., Ill
try again when she is in a better mood, or Maybe I need to get better at
asking women out.). Seligman argues that pessimism is a result of persistently attributing negative events to personal, pervasive, and permanent
causes, whereas optimism results from attributing misfortune to temporary,
specific, and external causes.
Riskind and his colleagues (Riskind et al., 1996) present a somewhat
different cognitive perspective on optimism and pessimism. They note studies
where the correlation between positive and negative thought content is
low (Ingram & Wisnicki, 1988); where a decrease in negative thinking does
not necessarily result in an increase in positive thinking (Watson, Clark,
& Carey, 1988); and where the outcome of CT is more strongly correlated
with increases in positive thinking than with decreases in negative thinking
(Garamoni, Reynolds, Thase, Frank, & Fascizka, 1992). Consequently, they
argue that pessimism is not simply the result of specific negative thoughts
and that, to adequately understand pessimism and develop an effective intervention approach, the standard cognitive model needs additional development. They start from A. T. Becks early suggestion ( 1967) that rigid, absolute
schemas may lead to greater vulnerability to hopelessness and pessimism, and
they argue that some individuals hold schemas that seem to promote pessimism
or suppress optimism. Examples include beliefs such as If I think too positively, Ill just be disappointed, If Im optimistic, Ill overlook something and
fail, or If I think positively, Im just kidding myself.
Riskind and his colleagues point out that when optimism-suppressing
schemas are present, successful treatment could result in a reduction in
OPTIMISM, PESSIMISM, AND PSYCHOTHERAPY

329

negative thinking without a concomitant rise in positive thinking. In a

recent study, Stewart and colleagues (1993) found that a subgroup of depressed patients who did not believe that they could positively affect their
own futures responded poorly to CT. Riskind et al. (1996) suggest that the
impact of optimism-suppressing schemas may explain this finding and that
interventions designed to address optimism-suppressing schemas and to increase the frequency of positive automatic thoughts might overcome the
poor treatment response observed in such demoralized individuals. This
perspective suggests that modifications to standard CT might significantly
increase its effectiveness as an approach to decreasing pessimism and increasing optimism.

HOW CAN PSYCHOTHERAPISTS MODIFY OPTIMISM

OR PESSIMISM?
Three Cognitive Approaches

In terms of intervention, Seligmans and his colleagues proposed methods for helping children develop into optimists rather than pessimists have
been discussed in chapter 14 and are beyond the scope of this one. In
discussing treatment of adults who have a pessimistic, depressogenic explanatory style, Seligman advocates CT as a treatment approach. In fact, he goes
as far as to assert that the effectiveness of CT as a treatment of depression
is due to its ability to reduce pessimism (Seligman, 1990, p. 91). Seligman
seems to assume that all that a therapist needs to do to produce a lasting
decrease in pessimism is to help the individual identify pessimistic thoughts
and dispute them. He does not propose that therapists make any adaptations
or adjustments to standard CT when the goal is to decrease pessimism
and increase optimism.
Fresco and his colleagues (Fresco, Craighead, et al., 1995; Fresco,
Sampson, et al., 1995) do not present a detailed theoretical discussion of
optimism and pessimism but propose adapting standard cognitive-behavioral
interventions to focus more directly on increasing optimism. The implicit
assumption is that interventions that are explicitly directed toward increasing optimism will prove more efficient and effective than interventions that
target negative thinking in the hopes of indirectly increasing optimism.
Riskind and his colleagues (Riskind et al., 1996) present the most
extensive proposal regarding how to best modify standard CT to maximize
its impact on optimism or pessimism. These authors argue that standard
focuses on decreasing the amount of negative thinking, and this is not
necessarily sufficient to produce increases in optimism. They advocate using

330

PRETZER AND WALSH

cognitive techniques explicitly to challenge optimism-suppressing schemas,

to enhance positive thinking, and to increase optimistic thought content
as a way of increasing the effectiveness of CT. Regarding our case example,
this perspective would suggest that interventions specifically intended to
increase Arts optimism may be more effective than either interventions
focused on decreasing his negative thinking and his pessimism or than
interventions focused on decreasing his social anxiety and improving his
social skills.

Toward an Integrative Approach

There are both commonalties and differences among the approaches
to increasing optimism advocated by Seligman, by Riskind and his colleagues,
and by Fresco and colleagues. Despite the differences between the approaches, they can be integrated into a more comprehensive approach toward
increasing optimism. In doing so, it is important to note that the intervention
approaches advocated by Seligman (1990) and Riskind and his colleagues
(1996) are intended to be used in the context of ongoing CT and that,
although the approach developed by Fresco and his colleagues (Fresco,
Craighead, et al., 1995; Fresco, Sampson, et al., 1995) was designed as a
self-administered treatment, the techniques used are equally appropriate in
the context of ongoing CT. We propose integrating the three approaches
toward increasing optimism within a more comprehensive approach to CT.

Foundations for Effective Intervention

Beck and other cognitive therapists (e.g., J. S. Beck, 1994; A. T. Beck
et al., 1979; Freeman, Pretzer, Fleming, 6r Simon, 1990) have written
extensively about structuring treatment to maximize the effectiveness of
cognitive-behavioral interventions. For example, they have emphasized the
idea that cognitive-behavioral interventions are more effective if they are
applied in the context of a collaborative relationship between therapist and
client, and they have argued that if a therapist and client are working
together toward goals that the client values, this collaboration reduces
resistance and increases the clients active involvement in therapy. The
collaborative approach that Beck advocates goes beyond simply agreeing on
mutually acceptable goals to include approaches such as guided discovery,
collaborative empiricism, and behavioral experiments, which are designed to structure therapy so that the client and therapist work together
with the client playing an active role both in developing an understanding
of the problems and in the change process.
Cognitive therapists have emphasized the importance of intervening
flexibly based on a clear formulation of the client and his or her problems

OPTIMISM, PESSIMISM, AND PSYCHOTHERAPY

33 1

rather than applying invented techniques in a cookbook manner (A. T.

Beck et al., 1979; Freeman et al., 1990; Persons, 1989). (For a more detailed
discussion of these issues, see J. S. Beck, 1994; A. T. Beck et al., 1979;
Freeman et al., 1990, & Persons, 1989.)
Increasing Motivution for Change
Some clients who enter treatment are strongly and consistently motivated to change from the outset. When this is the case, treatment can
proceed directly to interventions designed to develop a better understanding
of the individuals problems or to interventions designed to produce the
desired changes. However, many individuals enter therapy without strong,
consistent motivation to change. When this is the case, it can be important
for treatment to include interventions designed to increase and consolidate
motivation for change. One such intervention would involve helping the
client to clearly identify the costs of his or her pessimism and the likely
benefits of a more optimistic outlook. At this stage it can be important to
check for any fears the client has regarding the consequences of change. If
these fears present a significant impediment, it can be important to address
them, both with cognitive interventions and behavioral experiments. When
a clients lack of consistent motivation for change is a particular problem,
the therapist can use the motivational interviewing approach developed by
Miller and Rollnick (1991).
Identifying and Challenging Pessimistic Thoughts
As Seligman ( 1990)has suggested, a variety of standard CT interventions that focus on identifying and modifying dysfunctional thoughts can
be used in an attempt to replace a clients pessimistic thinking with more
optimistic thinking. Take our case example, where Art found himself in
situations where he assumed that a particular woman would have no interest
in him. His therapist helped him to look critically at his thoughts, to consider
any indications that she was not interested as well as any indications of
interest on her part, and to draw realistic conclusions about her likely level
of interest. In many situations, Art did not have sufficient information to
know whether the woman was likely to be interested in him, so his therapist
presented the idea of initiating a conversation as a way of finding out. When
Art tried this he was surprised to discover that several women showed
interest in him.
Although this type of intervention approach was initially designed to
modify peoples dysfunctional cognitions believed to contribute to psychopathology, it is flexible enough to use without significant modification for
increasing optimism. However, given the evidence that interventions that

332

PRETZER AND WALSH

reduce the amount of negative thinking do not necessarily result in a

corresponding increase in positive thinking, these interventions alone may
not prove sufficient to produce lasting increases in optimism (see Watson,
Clark, & Carey, 1988).

Identifying Alternatives to Pessimism

If individuals are to adopt less pessimistic outlooks, what sort of alternative view would be best? Clients often assume that the alternative the
therapist has in mind is the polar opposite (i-e., instead of expecting a
negative outcome, trying always to expect a positive outcome). Is this indeed
the most promising alternative? Consider our case example: Arts tendency
was to approach his date with a mindset such as Ill be boring. Shell have
a lousy time. She wont want to go out with me again. Theres no point to
trying. Do we, as therapists, want to try to get him to switch to the other
extreme ( W l be charming. Shell have a great time. Well date for a few
months and then get married and live happily ever after.)?
Certainly, if Art could believe this alternative view, it would make a
big difference both in his mood and in his behavior. However, two problems
are apparent. First, it is hard to imagine how Art could reach a point where
he honestly believes this alternative view because it is so discrepant from
his current view. Second, suppose Art was able to adopt this view, what
would happen when he went on his date? It seems that there is a significant
chance that his experiences on the date would disconfirm his optimistic
view. It would appear that any increase in optimism that is based on Arts
adopting such an extreme positive view will likely prove to be transitory.
There are a variety of other alternatives to Arts extreme pessimism that
are worth considering (see Table 15.2). We advocate working collaboratively
with the client to consider the alternatives and choose a promising one to
try out in real life. Theoretically, it seems that the ideal alternative would
be one that the client finds relatively easy to believe, that results in beneficial
changes in the individuals mood and behavior, that is practical to implement
in real-life situations, and that is likely to be supported by his or her
experience in daily life.
To his therapist, Art appeared to be a fairly ordinary guy who was
neither remarkably charming nor remarkably boring. Art and his therapist
chose to combine the approaches labeled adopt a mildly positive view
and accept realistically negative aspects of the situation and plan how to
cope with them (i.e., Im an OK guy. Some women will be interested in
me and some wont. I should try to meet women who like me the way I
am rather than trying to look cool. If I get nervous, it is better to go ahead
anyway than to avoid it.). This alternative worked well in real-life situations.

OPTIMISM, PESSIMISM, AND PSYCHOTHERAPY

333

Table 15.2.
Alternatives to Pessimism in Approaching a Date
Alternative approach

Specific thoughts

Anticipate the best possible

outcome
Adopt a less pessimistic view

111 be charming. Shell have a great time. Well

get married and live happily ever after.
I probably wont be charming enough but 11
1
give it a try anyway. Theres a chance that I
could do OK.

Adopt a mildly positive view

Im a decent guy. If I dont get so anxious and

pressured it will go OK. If I keep trying, 111 end
up getting more comfortable with dating and
it will go better.

Try to be realistic in anticipating

likely outcomes

111 give it a good try but the odds are that 111
need to date a number of women before I find
a relationship that works out.
When I get self-conscious, Itend to hold back
and get tongue-tied. When I get involved in
the conversation, it goes better. 111 make a
point of speaking up.

Accept realistically negative

aspects of the situation and plan
how to cope with them
Be mindful of negative emotions
without letting them determine
the outcome
Focus on pursuing positive
outcomes rather than on
avoiding negative outcomes

Im nervous and self-conscious but thats just

the way I react when dating. 111 go ahead and
give it a good try anyway.
How can I approach it so that we both have
a good time? 111 pick an activity that we
both like.

Specific Cognitive and Behavioral Interventions

In addition to using standard CT interventions, therapists can tailor
standard techniques to focus more specifically on optimism. Fresco and his
colleagues (Fresco, Craighead, et al., 1995; Fresco, Sampson, et al., 1995)
adapted the cognitive-behavioral techniques of self-monitoring and cognitive restructuring by combining them into a self-administered format and
instructing participants to keep a daily diary in which they monitored
both good and bad events and rated the causes of the events in terms of their
stability, internality, and globality. He directed the research participants to
then think of other possible causes of the events and to rate those causes
on the same dimensions. In theory, tailoring standard interventions to
address aspects of cognition that are directly related to optimism or pessimism
should increase their impact.
It is also possible to design interventions that are specifically designed
to decrease pessimism or increase optimism. One such technique, which
Riskind et al. (1996) propose, consists of the therapist helping the client
to identify and modify dysfunctional optimism-suppressing beliefs. Their
assumption is that an individual who holds a belief such as I dont deserve

334

PRETZER AND WALSH

positive things or good outcomes or Happiness, positive thinking, and

optimism are illusions will be more prone to engage in optimistic thinking
if the therapist can help him or her adopt an alternative view that supports
optimism. In Arts case, he assumed that optimistic thinking was dangerous
because an optimistic outlook would result in his attempting social interactions with attractive women who would then find him boring and reject
him. As long as he believed that optimistic thinking would lead directly to
rejection, he was reluctant to adopt this more positive outlook. After his
therapist helped him examine these beliefs and adopt alternative views
(Avoiding rejection at all costs results in my being isolated. I can take
small chances and cope with rejection if it happens. If I take small chances,
I can find out if women react the way I anticipate and I can learn how to get
relationships to go better.), he was more willing to approach interpersonal
situations with a guardedly optimistic outlook.
A second intervention that Riskind et al. (1996) propose is termed
positive visualizution. In this technique, the therapist asks the client to choose
a problematic situation and to visually rehearse attaining a positive outcome.
The client is instructed to imagine the outcome he or she would like to
achieve rather than visualizing the outcome he or she expects and to visualize
the specific steps needed to achieve these results. The visualization can
specifically target the challenges the client would be likely to encounter in
real life. Thus, Art might be asked to visualize a casual conversation with
an attractive acquaintance from beginning to end, to imagine encountering
the problems which he would normally encounter in such a situation, to
imagine coping successfully with each problem as it arises, and to imagine
a successful outcome. In order for Art to be able to imagine coping successfully with the problems he encounters, the therapist would need to help
him to identify these and to identify promising ways of coping with them.
Riskind et al. ( 1996)present two variations on the positive visualization
technique. The first is the story board technique, where the client is
instructed to visualize a series of discrete scenes that lead to a desired
outcome. For example, a salesman who has difficulty making unsolicited
calls might be asked to first visualize making calls to potential customers,
then to visualize some of the calls resulting in appointments, and finally to
visualize some of the appointments resulting in sales. At each step, the
client would imagine coping with any difficulties that arise and would
imagine a successful outcome, such as being pleased that he or she carried
out the task. The second variation on the positive visualization technique
is the invulnerability training technique, where the client would visualize
feeling good about how he or she handled the situation when things do not
turn out the way the client hoped they would. Riskind and his colleagues
propose that when positive visualizations of these types are repeated across
a variety of situations and different areas of life, there may be a point at
OPTIMISM, PESSZMISM , AND PSYCHOTHERAPY

335

which the effects of the visualizations generalize and have beneficial effects
that are not confined to the specific situations visualized.
The silver lining technique (Riskind et al., 1996) is a simple process
in which the client is assigned the task of taking a negative experience and
identifying one genuinely positive element in it. The clients task is not
simply to mouth positive platitudes but to identify genuinely positive aspects
of the experience. Riskind argues that this simple exercise can have a
significant impact. For example, at the outset of therapy, whenever Art had
a social interaction that did not result in a date, he was inclined to declare
it to have been a total failure. After identifying the realistic positive outcomes
of some of these failures (i.e., At least I got myself to try, Im getting
less anxious about striking up a conversation, She seems to be getting
more comfortable too, or I didnt get a date but she did show some
interest.) he found it easier to persist long enough with his efforts to start
getting some dates.
A fourth technique recommended by Riskind et al. (1996) is pump
priming, a technique based on the principle of cognitive priming. Events
relevant to a particular cognitive construct or schema can increase the
salience of the schema and increase the likelihood of its being spontaneously
elicited by subsequent events. Riskind gives the example of a n individual
who reads an article about cats over coffee in the morning and subsequently
throughout the day is more likely to think spontaneously about cats, to
notice references to cats, and to recall memories of cats. He suggests that
by having individuals intentionally attend to experiences relevant to a
schema that they wish to encourage, to intentionally repeat relevant selfstatements, or to visualize relevant stimuli, it may be possible to increase
spontaneous attention to positive experiences, increase spontaneous use of
an optimistic explanatory style, and so forth.
It is likely that a variety of other cognitive-behavioral interventions
could be modified to more specifically address optimism and pessimism.
For example, Burnss ( 1980) antiprocrastination sheet technique asks the
client to record his or her anticipations about how difficult and how rewarding it will be to perform a particular task, and then, following completion
of the task, to rate how difficult the task actually turned out to be and how
rewarding it actually was to perform the task. This technique can easily be
modified to target the specific expectancies that are relevant for decreasing
an individuals pessimism or increasing their optimism.
The antipessimism sheet shown in Figure 15.3 is one possible way
of doing this. Clients are asked to identify upcoming events that they are
likely to approach with a pessimistic outlook and to list each event in the
first column of the form. Before the event, the individual is asked to record
the best they can foresee happening, the worst they can foresee happening,
and the outcome that seems most likely. Following the event, they are asked
336

PRETZER AND WALSH

1
Upcoming Event

Before the event:

The best I can
foresee
happening is:

I can
The
foresee
happening is:

The outcome
which seems
most likely is:

kfter the event:

The actual
Dutcome was:

Figure 75.3.Antipessimism sheet.

to record the actual outcome. After a number of events have been recorded,
therapist and client can examine the correspondence between the individuals anticipations and the actual outcome, the impact a pessimistic outlook
has on the actual outcome, and what the client can do to influence the
outcome.
The antipessimism sheet was not used with Art because it had not yet
been developed. Instead, the same type of intervention was used in a less
structured way. Art was asked to rate the level of anxiety he anticipated
experiencing in upcoming interactions, the degree of interest he anticipated
that the woman would display, and the level of satisfaction he expected to
experience following the interaction. Following the interactions, he and his
therapist discussed his actual experience, looked for consistent biases in his
expectancies, and identified positive outcomes, which Art tended to overlook (i.e., decreases in his anxiety, opportunities to test his preconceptions,
and improvements in his social skills). This process seemed to facilitate
Arts recognizing that his anticipations were unduly negative and to increase
his ability to appreciate the positive aspects of interactions that produced
mixed results.

Relapse Prevention
When therapy has been successful, there remains a risk of relapse at
some point in the future. Borrowing ideas from Marlatt and Gordons (1985)
work on relapse prevention in the treatment of substance abuse, a cognitive
therapist at the conclusion of treatment typically helps the client identify
OPTIMISM, PESSIMISM, AND PSYCHOTHERAPY

337

high-risk situations where there is an increased risk of relapse, plan how

to cope with these events if they occur, and practice the skills needed for
the client to cope effectively. For example, as Art had his initial success in
interpersonal relationships, he gradually accepted the idea that he was not
inherently boring and began to believe that people could genuinely be
interested in him once they got to know him. However, if he were to
experience rejection in a relationship or if he were to encounter someone
who accused him of being boring, he could easily revert to his original
negative outlook, especially if he experienced several such experiences. Arts
treatment concluded with his therapist helping him identify adaptive selfstatements to use in coping with rejection (i.e., Things didnt work out
with her but I know from experience that other women can be interested
in me.) and plan active steps for him to take in coping with rejection (i.e.,
Avoiding rejection just makes things worse, I need to continue reaching
out to others even if it is hard at first.).

CAN COGNITIVE-BEHAVIORAL INTERVENTIONS PRODUCE

LASTING CHANGES IN OPTIMISM AND PESSIMISM?
Only a limited number of studies have directly examined the effectiveness of cognitive-behavioral interventions designed specifically to modify
optimism or pessimism.2 Fresco and colleagues combined the cognitivebehavioral techniques of self-monitoring and cognitive restructuring into a
self-administered format and investigated the effects of 1 month of selfadministered optimism training on pessimistic college students (Fresco,
Craighead, et al., 1995; Fresco, Sampson, et al., 1995). Analysis of student
participants attributional styles before and after treatment indicated that
1 month of minimally supervised, self-guided intervention produced a greater
reduction in pessimistic explanatory style than in a no-treatment approach
in the control group of students.
Riskind and his colleagues (1996) compared the effects of three group
treatments (standard cognitive therapy, optimism training, and cognitive
priming) with progressive relaxation training (as a control group) in a sample
of 83 college students. This studys results yielded promising initial support
There is considerable evidence that CT and related cognitive-behavioral approaches are effective
treatments for depression and, because interventions designed to reduce negative thinking play a
prominent role in these intervention approaches, it is tempting to interpret this as providing
evidence that CT increases optimism or decreases pessimism (see Seligman, 1990). However, this
line of reasoning assumes that the efficacy of CT for depression is due to its effects on negative
cognitions and that interventions that modify specific negative cognitions lead to lasting changes in
levels of optimism or pessimism. Because neither assumption is beyond debate, research into the
efficacy of CT as a treatment for depression is not directly relevant to the question of whether
cognitive-behavioral interventions can be used to increase optimism.

338

PRETZER AND WALSH

for the effectiveness of the optimism training condition. The optimism

training group had scores superior to those of the control group on four
out of five measures. This included a significantly higher level of optimistic interpretations for negative events and a higher level of positive selfstatements. It is interesting to note that optimism trainings effects appeared
to be specific to positive thinking. It did not demonstrate a significant effect
on the level of negative self-statements. In this study, the other treatment
approaches did noticeably less well. Standard cognitive therapy scores were
superior to those of the control group on two out of five measures (positive
self-statements and problem-solving self-efficacy),while the cognitive priming treatment was not significantly more effective than that of the control
group (progressive relaxation training) on any of the measures.
The findings reported by these investigators are encouraging. Unfortunately, these studies have only been conducted with college undergraduates
and have not yet been replicated, thus, their generalizability is severely
limited. It goes without saying that more extensive research with a broader
range of populations will be needed before we have grounds for drawing
firm conclusions.

THE EFFECTS OF MODIFYING OPTIMISM AND PESSIMISM

IN ADULTS
The association between optimism and a wide variety of positive outcomes (Scheier & Carver, 1992; see also chap. 9, this volume) makes it
tempting to assume that interventions that increase an individuals optimism
will produce manifold benefits. However, this proposition has not yet received extensive scrutiny. Fresco, Craighead, et al. (1995) found that, in a
sample of 125 college undergraduates, 1 month of self-administered optimism
training produced a significant increase in the use of an optimistic explanatory style, and a nonsignificant trend toward lower depression scores and
health symptom scores. A hierarchical multiple regression analysis found
that the cross-product of attributional style by life stress significantly predicted level of depression and health symptoms for control participants but
not for participants who received the optimism training. These results were
interpreted as suggesting that 1 month of optimism training may have
moderated the effects of the diathesis-stress interaction in pessimistic college
students. However, findings such as this provide encouragement rather than
a convincing demonstration of the benefits of optimism training.
The authors clinical experience is consistent with the research, which
suggests that cognitive-behavioral interventions can produce significant
changes in optimism and pessimism and that this can have substantial
benefits. In Arts case, initial interventions led him to reevaluate his
OPTlMlSM, PESSlMlSM, W D PSYCHOTHERAPY

339

pessimism regarding relationships. After five sessions, he began initiating

conversations and was surprised to find that some women responded positively. By Session 7, he had begun dating and experimenting with allowing
himself to be genuine rather than trying to appear cool. At that point in
therapy, he was considering a sexual relationship and talked with his therapist
about how to let his girlfriend know that he had an erectile dysfunction.
He subsequently discussed his sexual dysfunction with his girlfriend, and
they initiated a sexual relationship.
By Session 10, Art spontaneously expressed the opinion that anyone
can be interesting once you get to know them and, as he did so, he seemed
to be expressing an honest conviction that this applied to him as well as
to others. After Session 11, Art reported that he felt comfortable while
dating and while facing other social situations. He was dating actively
and, although he was not completely satisfied with his primary romantic
relationship, he expressed confidence that he could successfully pursue relationships on his own. Treatment was terminated by mutual agreement with
the option of Arts resuming treatment in the future if he encountered difficulty.

ISSUES TO CONSIDER REGARDING OPTIMISM AND PESSIMISM

IN PSYCHOTHERAPY
Should the Goal of Intervention Be Optimism, Realism,
or Something Else?
When CT originated as a treatment for depression, there was a strong
focus on decreasing the amount of negative thinking or on counteracting
negative thoughts due to the hypothesized role of negative thinking in
depression. As cognitive-behavioral investigators have begun considering
optimism and pessimism, some seem to assume that optimism and pessimism
are polar opposites and that, therefore, interventions that decrease negative
thinking will necessarily increase positive thinking. However, the previously
cited studies that found little correlation between positive and negative
thought content, found that a decrease in negative thinking did not necessarily result in an increase in positive thinking, and found that therapeutic
outcome was more strongly correlated with changes in positive thinking
than with changes in negative thinking seem to present a strong challenge
to this assumption (Garamoni et al., 1992; Ingram & Wisnicki, 1988; Watson
et al., 1988).
Riskind et al. (1996) interpret these findings as showing the importance
of working explicitly to enhance positive thinking and optimistic thought
content rather than simply attempting to decrease negative thinking. How-

340

PRETZER AND WALSH

ever, a number of other perspectives have been presented. Kendall (1982)

argued that the goal of intervention should be to increase the amount of
non-negative thinking rather than attempting to increase positive thinking. The states of mind model asserts that one should focus on the balance
between positive and negative cognitions rather than looking at either in
isolation (Bruch, 1997; Schwartz & Garamoni, 1989). Fresco, Craighead,
et al. (1995) suggest that realistic thinking might be a better alternative
than either optimism or pessimism. Other, more radical alternatives include
acceptance (Hayes, Jacobson, Follette, & Doucher, 1994; Wilson, 1996)
or mindfulness (Linehan, 1993, pp. 144-147) as alternatives to either
optimism or pessimism. Clearly this is a topic in need of further exploration.

Are Increases in Optimism Following Treatment a Side Effect of

Decreased Depression?
The increase in optimism observed in depressed individuals following
a successful course of CT has been interpreted as demonstrating that CT
is effective as a means of increasing optimism (Seligman, 1990). However,
an alternative hypothesis is that the decrease in pessimism or increase in
optimism observed following successful treatment of depression is a result
of the decrease in depression rather than a direct result of CT on the
individuals levels of optimism or pessimism.
At this point, some empirical evidence appears to support this hypothesis. For example, in an empirical investigation of the proposed characteristics
of Depressive Personality Disorder (see American Psychiatric Association,
Diagnostic and Statistical Manual of Mental Disorders; DSM-IV, 1994), Hartlage, Arduino, and Alloy ( 1998)found that pessimism appeared to be a statedependent concomitant of depression rather than an enduring personality
characteristic which predisposed individuals to depression. However, other
evidence suggests that an individuals level of optimism is not simply an
artifact of the individuals level of depression and actually may protect the
individual against the development of depression. For example, Carver and
Gaines (1987) found that level of optimism measured during the third
trimester of pregnancy predicted level of postpartum depression even after
statistically controlling for initial level of dysphoria.
It appears that it will be important for researchers to design studies of
the effectiveness of interventions intended to decrease pessimism or increase
optimism so that it is possible to distinguish changes in these qualities that
are a direct effect of the interventions from changes that are a side effect
of changes in depression. It will be important for investigators to remember
that viewing pessimism as an effect of depression and the hypothesis that
optimism confers resistance to depression are not necessarily incompatible.
OPTIMISM, PESSIMISM, AND PSYCHOTHERAPY

34 I

It is possible that pessimism is, to some extent, a result of depression, whereas

optimism, to some extent, tends to reduce the likelihood of an individual
becoming depressed.
Is Too Much Optimism Risky?

A number of authors suggest that optimism can be hazardous in some

situations and that individuals should be cautious or selective in adopting
an optimistic outlook. For example, Seligman (1990) suggests that one
should not use an optimistic outlook in situations where the cost of failure
would be high (p. 209). Similarly, Freeman and DeWolf (1992) assert that
an overly optimistic outlook can contribute to poor decision making, failure
to accept responsibility for ones actions, persisting with a maladaptive
strategy, and other problems (pp. 84-90). If there are any significant risks
to adopting a more optimistic approach to life, treatments designed to
increase optimism will need to take them into account.
Scheier and Carver (1992) provide a good discussion of a variety
of ways in which optimism, particularly unrealistic optimism, might have
disadvantages. They conclude that, although the picture is still incomplete,
optimism appears to have substantial advantages over pessimism. Some
research does support the idea that there may be risks, as well as benefits,
to optimism. For example, in a study of cognitions related to risk-taking
behavior, Fromme, Katz, and Rivet (1997) found that positive outcome
expectancies were more consistently associated with risk taking than were
beliefs about potential negative consequences. Clearly, this is an issue that
calls for further examination and that will need to be considered as approaches to increasing optimism are developed and tested.
Should Cultural Differences Be Considered?
Most available research on optimism and pessimism has been conducted
on Caucasian American participants, and much of the authors clinical
experience has been with Caucasian American clients. However, there is
reason to believe that attention to cultural differences may be important
both in conducting research on this topic and in intervening clinically.
Chang (1996) examined differences in optimism, pessimism, and coping
between Asian American and Caucasian American college students. He
found consistent differences between the two groups of subjects that suggested optimism may have an important influence on adjustment for Asian
American students, and pessimism may be more important for Caucasian
American students. In discussing the clinical implications of these findings
Chang suggests that interventions designed to reduce pessimism may be
appropriate for Caucasian Americans, and interventions designed to increase

342

PRETZER AND WALSH

optimism may be more appropriate for Asian Americans. After practicing

for some time in an ethnically diverse community, the authors experience
is consistent with the hypothesis that cultural differences can have an
important impact on levels of optimism and pessimism. Further research
will be needed to examine a broader range of cultures and to clarify the
extent to which cultural differences are important. (For a more detailed
discussion of cultural influences on optimism and pessimism, see chap. 12,
this volume.)

Do the Benefits of Modifying Optimism and

Pessimism Justify the Costs?
Given the assertions that researchers make regarding the manifold
benefits of optimism and the potential for developing interventions that
result in lasting increases in optimism, one could wax enthusiastic about
such interventions. However, at a time when health care budgets face
financial constraints and managed care organizations play an increasing
role in determining which intervention approaches will receive financial
reimbursement, it is increasingly important for investigators to go beyond
examining the effectiveness of a given treatment approach to also consider
whether it is cost effective. An intervention approach is likely to receive
much greater acceptance by professionals, the public, and those who control
the funding if it is clear that the benefits of intervention substantially
outweigh the costs.

CONCLUSION
Shall we be optimistic and emphasize the great benefits which could
potentially result from intervention approaches designed to increase dispositional optimism, or shall we be pessimistic and emphasize the unanswered
questions and significant problems faced by investigators in this field?Perhaps
we betray our own biases in advocating a realistic perspective. Interventions designed to increase optimism have potential to produce substantial
benefits if they prove to be both effective and cost effective; however, they
will need further development and investigation to reach that stage.
Significant challenges remain. In particular, it is clear that cognitivebehavioral interventions can be effective and efficient in producing shortterm change in specific cognitions. It is not yet clear if these same intervention approaches can produce broad, long-term changes of the sort needed
to produce persistent increases in dispositional optimism. Although some
proponents are enthusiastic about the potential of cognitive-behavioral
interventions to produce lasting changes (J. S. Beck, 1994; Pretzer & Beck,
OPTIMISM, PESSIMISM, AND PSYCHOTHERAPY

343

1996; Seligman, 1990) the available research has focused much more attention on the immediate, specific effects of CT and less attention on the
broad, long-term effects. It is ironic, perhaps, that those investigators who
possess a high level of dispositional optimism are most likely to persist with
their research in the face of the inevitable difficulties and setbacks to
eventually provide us with an answer to this question.

REFERENCES
American Psychiatric Association. (1994). Diagnostic and statistical m n d of mental
disorders (4th ed., rev.). Washington, DC: American Psychiatric Association.
Beck, A. T. (1967). Depression: Clinical, experimental, and theoretical aspects. New
York: Harper & Row. (Reprinted as Depression: Causes and treatment. Philadelphia: University of Pennsylvania Press, 1972).
Beck, A. T., Rush, A. I., Shaw, B. F., & Emery, G. (1979). Cognitive therapy of
depression. New York: Guilford Press.
Beck, J. S. (1994). Cognitive therapy: Basics and beyond. New York: Guilford Press.
Bruch, M. A. (1997). Positive thoughts or cognitive balance as a moderator of the
negative life events-dysphoriarelationship:A reexamination. Cognitive Therapy
and Research, 21, 25-38.
Bums, D. D. (1980). Feelinggood: The new mood therapy. New York: William Morrow.
Carver, C. S., & Gaines, J. G. (1987). Optimism, pessimism, and postpartum
depression. Cognitive Therapy and Research, 11, 449-462.
Chang, E. C. (1996). Cultural differences in optimism, pessimism, and coping:
Predictors of subsequent adjustment in Asian American and Caucasian American college students. Journal of Counseling Psychology, 43, 113-123.
DeRubeis, R. J., & Crits-Cristoph, P. (1998). Empirically supported individual and
group psychological treatments for adult mental disorders.Journal of Consulting
and Clinical Psychology, 66, 37-52.
Freeman, A., & DeWolf, R. (1992). The 10 dumbest mistakes smart people make and
how to avoid them: Simple and sure techniques for gaining greater control of your
life. New York: Harper Collins.

Freeman, A., Pretzer, J. L., Fleming, B., & Simon, K. M. (1990). Clinical applications
of cognitive therapy. New York: Plenum Press.
Fresco, D. M., Craighead, L. W., Sampson, W. S., Watt, N. M., Favell, H. E., &
Presnell, K. E. ( 1995, March). The effects of self-administered optimism training
on attributionul style, levels of depression, and health symptoms of pessimistic college
students. Poster presented at the annual meeting of the Eastern Psychological
Association, Boston, MA.
Fresco, D. M., Sampson, W. S., Craighead, L. W., Clark, J., & Enns, C. (1995,
November). Self-administeredoptimism training: The process of lessening the impat

344

PRETZER AND WALSH

of pessimistic explanatory styk. Poster presented at the annual meeting of the

Association for Advancement of Behavior Therapy, Washington, DC.

Fromme, K., Katz, E. C., &Rivet, K. (1997). Outcome expectancies and risk-taking
behavior. Cognitive Therapy and Research, 21, 421-442.
Garamoni, G. L., Reynolds, C. F., Thase, M. E., Frank, E., & Fascizka, A. L. (1992).
Shifts in affective balance during cognitive therapy of major depression. Journal
of Consulting and Clinical Psychology, 60, 260-266.
Hartlage, S., Arduino, K., & Alloy, L. B. (1998). Depressive personality characteristics: State dependent concomitants of depressive disorder and traits independent of current depression. Journal of Abnormal Psychology, 107, 349-354.
Hayes, S. C., Jacobson, N. S., Follette, V. M., & Doucher, M. J. (1994). Acceptance
and change: Content and context in psychotherapy. Reno, NV: Context Press.
Ingram, R. E., Miranda, J., & Segal, Z. V. (1998). Cognitive vulnerability to depression.
New York: Guilford Press.
Ingram, R. E., & Wisnicki, K. S. (1988). Assessment of positive automatic cognition.
Journal of Consulting and Clinical Psychology, 56, 898-902.
Isen, A. M. (1984). Toward understanding the role of affect in cognition. In R. S.
Wyer & T. K. Skrull (Eds.), Handbook of social cognition (pp. 179-236). Hillsdale, NJ: Erlbaum.
Kendall, P. C. (1982). Cognitive processes and procedures in behavior therapy. In
C. M. Franks, G. T. Wilson, P. C. Kendall, & S. D. Hollon (Eds.), Assessment
strategies for cognitive behavioral intervention. New York: Academic Press.
Linehan, M. M. ( 1993). Cognitive-behavioral treatment of borderline personality disorder. New York: Guilford Press.

Marlatt, G. ,4.&
, Gordon, J. M. (Eds.). (1985). Relapse prevention: Maintenance
strategies in the treatment of addictive behaviors. New York: Guilford.

Miller, W. R., & Rollnick, S. (1991). Motivational interviewing: Preparing peopk to

change addictive behavior. New York: Guilford.
Persons, J. B. (1989). Cognitive therapy in practice: A case formulation approach. New
York: Norton.
Peterson, C., & Seligman, M. E. P. (1984). Causal explanations as a risk factor for
depression: Theory and evidence. Psychological Review, 91 , 347-374.
Pretzer, J. L., & Beck, A. T. (1996). A cognitive theory of personality disorders.
In J. F. Clarkin & M. F. Lenzenweger (Eds.), Major theories ofpersonality disorder
(pp. 36-105). New York: Guilford.
Riskind, J. H., Sarampote, C. S., & Mercier, M. A. (1996). For every malady
a sovereign cure: Optimism training. Journal of Cognitive Psychotherapy: An
International Quarterly, 10, 105-1 17.
Scheier, M. F., & Carver, C. S. (1992). Effects of optimism on psychological
and physical well-being: Theoretical overview and empirical update. Cognitive
Therapy and Research, 16, 201-228.
OPTIMISM, PESSIMISM, A N D PSYCHOTHERAPY

345

Schwartz, R. M., & Garamoni, G. L. (1989). Cognitive balance and psychopathology: Evaluation of an information processing model of positive and negative
states of mind. Clinical Psychology Review, 9, 271-294.
Seligman, M. E. P. (1990). Learned optimism. New York: A. A. Knopf.
Stewart, J. W., Mercier, M. A., Quitkin, F. M., McGrath, P. J., Nunes, E., Young,
J., Ocepek-Welikson,K., & Trocamo, E. (1993). Demoralization predicts nonresponse to cognitive therapy in depressed outpatients. Journal of Cognitive
Psychotherapy, 7, 105-116.
Taylor, S., & Rachman, S. J. (1991). Fear of sadness. Journal of Anxiety Disorders,
5, 375-381.
Taylor, S., & Rachman, S. J. (1992). Fear and avoidance of aversive affective states:
Dimensions and causal relations. Journal of Anxiety Disorders, 6, 15-25.
Watkins, P. C., Mathews, A., Williamson, D. A., & Fuller, R. D. (1992). Moodcongruent memory in depression: Emotional priming or elaboration? Journal
of Abnormal Psychology, 101, 581-586.
Watson, D., Clark, L. A., & Carey, G. (1988). Positive and negative affectivity
and their relation to anxiety and depressive disorders. Journal of Abnormal
Psychology, 97, 346-353.
Wilson, G. T. (1996). Acceptance and change in the treatment of eating disorders
and obesity. Behuvior Therapy, 27, 417-439.

346

PRETZER AND WALSH

16
A VERY FULL GLASS: ADDING
COMPLEXITY TO OUR THINKING
ABOUT THE IMPLICATIONS AND
APPLICATIONS OF OPTIMISM AND
PESSIMISM RESEARCH
JULIE K. NOREM AND EDWARD C. CHANG

As much of the research discussed in this volume indicates, there is

substantial evidence that optimism, in its many forms, is related to better
outcomes (e.g., coping, satisfaction, well-being) measured in a variety of
ways across a variety of contexts. This evidence makes it extremely tempting
to conclude that optimism is always to be desired over pessimism, and
further, that as educators, policy consultants, therapists, and parents we
should do everything we can to promote optimism. Indeed, as we leave the
century espousing the sentiments of Dale Carnegie and Dont Worry, Be
Happy, it seems churlish, as well as somewhat un-American, to dwell on
caveats to that conclusion.
Nevertheless, we live in a culture where cautions about overgeneralization, misapplication, or limitations-all the carefully considered details

347

written up in the presentation of original research-are easily lost in the

rush for catchy sound bites, palatable self-help, and quick-fix policy. It thus
becomes especially important that we clarify for ourselves, and for those
who might be consumers of our research, the ways in which our theories
and research present a picture that is more complicated-and much more
potentially useful-than the simple phrases optimism is good and pessimism is bad suggest. Accordingly, in this chapter we review some complexities that seem important to consider both as we design our research, and
as we disseminate it. Despite use of the term complexity in the title, the
two-part message of the chapter is rather simple: (a) There are potential
benefits and costs to both optimism and pessimism (as they are variously
defined in the literature) that may be highly sensitive to context (broadly
and variously defined); and thus, (b) our research designs, interpretations
of results, advocacy, interventions, and teaching need to be sensitive to
costs, benefits, and context.

WHAT IS CONTEXT?
The complexity referred to in the chapter title immediately arises when
one considers the contexts that might be important to understanding the
functions and consequences of optimism and pessimism. Without claiming
to exhaust the list of possibilities under this heading, some of these important
contexts that have implications for optimism and pessimism include

1. the specific kind of optimism and pessimism under scrutiny

2. the other personality characteristics of individuals who are
optimistic and pessimistic (i.e., the intrapsychic context)
3 . the particular outcome variables considered
4. the interpersonal and social contexts within a culture
5. the life-span developmental context
6. the larger cultural context.
Varieties of Optimism and Pessimism
As is clear from the table of contents for this volume, optimism and
pessimism are used as umbrella terms within the field. They in fact refer
to several distinct concepts, often require modifiers (e.g., naive optimism)
or are used as modifiers (e.g., pessimistic attributional style), and only
sometimes correspond to everyday usage. Among the concepts included in
the literature are dispositional optimism and pessimism (Scheier & Carver,
1985); optimistic and pessimistic attributional or explanatory styles (Peterson &Seligman, 1987); naive optimism (Epstein & Meier, 1989); optimis348

NOREM AND CHANG

tic biases or illusions (Taylor & Brown, 1988);neurotic and rational pessimism (Kelman, 1945); unrealistic optimism (Weinstein, 1980); unrealistic
pessimism (Dolinski, Gromski, & Zawisza, 1987); realistic pessimism (Frese,
1992); and defensive pessimism and strategic optimism (Norem & Cantor,
1986a). That list does not even include terms that are strongly related, such
as hope (as previously discussed by Snyder, Sympson, Michael, & Cheavens,
chap. 5, this volume).
Of these concepts, dispositional optimism and pessimism-defined as
the tendency to hold generally positive or negative expectations-probably
best represent the lay or colloquial understanding of the terms (as discussed
by Carver & Scheier, chap. 2, this volume, and Scheier, Carver, & Bridges,
chap. 9, this volume). Even these concepts, however, are more complex
than they seem at first glance, in that the initial assumption that optimism
and pessimism are opposite ends of the same continuum has been questioned
(as mentioned in the introduction to this volume). There is now reason to
believe that they represent two unipolar dimensions: The opposite of optimism is the lack of optimism, which is distinguishable from the presence
of pessimism (Marshall, Wortman, Kusulas, Hervig, & Vickers, 1992). Similarly, the opposite of pessimism is the lack of pessimism, which is distinguishable from the presence of optimism (Robinson-Whelen, Kim, MacCallum,
& Kiecolt-Glaser, 1997).
Distinctions among these constructs are more than semantic to the
extent that the different constructs have different consequences. There is
some evidence, for example, that lack of optimism may have different
consequences than does the presence of pessimism. Chang (1998) found
that pessimism among college students was related to physical symptom
reports, whereas optimism was not. In another study of college students,
Chang, Maydeu-Olivares, and DZurilla ( 1997) found that pessimism, but
not optimism, was related again to dysphoric symptoms. Schulz, Bookwala,
Knapp, Scheier, and Williamson ( 1996) found that optimism was unrelated
to survival in cancer patients, but that pessimism was correlated with earlier
death. Raikkonen, Matthews, Flory, Owens, and Gump (1999) found that
pessimistic adults (as well as those who were dispositionally anxious) had
higher ambulatory blood pressure and greater negative mood than did optimistic adults. When the optimistic participants were in a negative mood,
however, their blood pressure was as high as the pessimistic participants.
Obviously we should be cautious about basing an intervention on any
one study. Nevertheless, the results of these studies illustrate some ways we
need to be careful in thinking about intervention. T o the extent that
optimism and pessimism are stable dispositions, and assuming some potential
for change at the disposition level, it may be that potential for change varies
depending on whether one is trying to create an optimistic disposition, or
to dampen a pessimistic one. Similarly, if one is designing an intervention,
A VERY FULL GLASS

349

it may very well be that different techniques need to be developed for

creating optimism and diminishing pessimism, and it will be important to
determine which goal is most important: If pessimism alone is associated
with survival, we should not expend resources trying to increase optimism
(at least until it can be shown that this has some reciprocal effect on levels
of pessimism). Although optimism was clearly associated with more positive
outcomes in Raikkonen et al.s (1999) study, it is not at all clear that
trying to make people more optimistic would effect such important health
indicators as blood pressure. That is, optimism in this study did not appear
to have protective benefits as much as pessimism seemed to have significant
costs. Of course, if blood pressure volatility is strongly moderated by mood,
it may very well be that trying to increase optimism would help people to
regulate their moods, which might, in turn, have desirable effects on blood
pressure. Nevertheless, we do not want to waste resources increasing optimism if we might more efficiently focus on some other mood regulation
strategies or biofeedback techniques.
Intrapsychic Context and Comparison of Outcomes
The results discussed above reinforce the importance of the very basic
idea that the more we know about the mechanisms that underlie a given
empirical association between optimism, pessimism, and any other variable,
the more likely it will be that our interventions or applications will be
effective. Clearly, however, as we learn more about what can be changed
and how to change it, we also need to consider precisely what we might
want to change, in which contexts. At this point, we know very little about
how to either increase dispositional optimism, or decrease dispositional
pessimism-and certainly many dispositional theorists would be pessimistic
about change at this level. Other varieties of optimism and pessimism are
theoretically more amenable to change, however. Seligman ( 1991) has
argued strongly that optimism can be learned. He and his colleagues define
optimism as a particular explanatory style in response to negative life events
(as previously discussed by Gillham, ShattC, Reivich, & Seligman, chap. 3,
this volume). Seligman is also a persuasive advocate of learning a more
optimistic style. Yet, much of the research on learned optimism has been
based almost entirely on studies of peoples pessimistic attributions for negative events, not on their optimistic attributions for positive events (Peterson
& Chang, in press). In that sense, it seems curious that so much has been
said by so many theorists about teaching people to develop an optimistic
explanatory style, when in fact, there have actually been very few direct
studies showing that an optimistic explanatory style leads to positive adjustment. Taylor (1989; Taylor & Brown, 1988) also argues for the benefits of
an optimistic bias as part of a set of mild positive illusions that are related

350

NOREM AND CHANG

to better adaptation, and that are, to some extent, malleable. However, the
adaptiveness of positive illusions has been seriously questioned (Baumeister,
1989; Colvin & Block, 1994). Norem (1989; see also chap. 4, this volume)
proposes that the cognitive strategies of defensive pessimism and strategic
optimism are potentially changeable. We talk a little more about research
on these strategies below.
As we consider what might be changed, however, there are further
complexities that might influence our answers to the question of whether,
or under what circumstances, we should try to change what we might be
able to change. Optimism and pessimism-in all the ways they can be
defined-do not exist in an intrapsychic vacuum. Instead, they are more
or less integrated into personality and are strongly related to a variety of
other personality characteristics, temperament, and past experiences. Park,
Moore, Turner, and Adler (1997), for example, take care to distinguish
between positive effects of optimism that are mediated by constructive
thinking and the less sanguine effects of naive optimism. Their results
lead them to suggest that optimism may need to be channeled through
effective thinking strategies if it is to lead to better coping and adjustment;
in other words, they found little evidence that optimism in isolation was
related to better coping.
The question of whether optimism and pessimism are truly discriminable from related constructs has plagued researchers in this area from the
start (Scheier, Carver, & Bridges, 1994). Putting aside the psychometric
complexities of this debate for the moment, we need to consider the following: Even if we can isolate the positive or negative statistical impact of a
given kind of optimism or pessimism on a particular outcome variable, it
does not follow that an application of research seeking to target pessimism
or optimism in isolation will be able to do so. In addition, it does not
necessarily follow that even if this research does target pessimism or optimism
in isolation, that there will be the further intended effect on the outcome
in question. Davidson and Prkachin (1997), in one of the few published
attempts to disentangle the effects of dispositional optimism and unrealistic
optimism, showed in two different studies that dispositional optimism and
unrealistic optimism interacted to predict coronary heart disease (CHD)
related outcomes. In the first study, optimism alone was unrelated to exercise
over time, while unrealistic optimism was related to decreased exercise over
time. Those participants who scored high on both optimism and unrealistic
optimism showed the largest decreases in exercise over time, while those
who scored high in optimism and low in unrealistic optimism showed the
greatest increases. A similar pattern was found in a second study, when
the outcome variable was knowledge of CHD prevention after classroom
instruction. In both studies, unrealistic optimism and dispositional optimism
were positively, although not strongly, correlated with each other. Davidson
A VERY FULL GLASS

35 1

and Prkachins (1997) results suggest that attempting to increase optimism

without taking into account other psychological variables would be misguided: We might have little effect on outcome, if we only targeted optimism,
or we might have the unintended, collateral effect of increasing unrealistic
optimism. At the extreme, trying to increase the optimism of those people
who are already unrealistically optimistic might have serious negative effects.
We are not suggesting that anyone has tried to do this-only that it seems
that relatively little thought is typically given to how potential changes in
optimism might influence and be influenced by other aspects of personality.
Interaction effects such as those found by Davidson and Prkachin need to
be included in our assessment of the costs and benefits of different kinds
of optimism and pessimism.
We also need to be cognizant that statistical identification of the effects
of single variables may have questionable relevance to psychology in vivo.
That is, just because we can control statistically for the relationship between
optimism and other variables (e.g., extraversion, narcissism, etc.) does not
mean that we can necessarily target optimism in ways that separate it from
other characteristics of actual people.
There are other examples of the ways in which cost-benefit analyses
of optimism and pessimism are influenced by intrapsychic context, as well
as by the particular outcome variables one examines. As reviewed in chapter
4 of this volume, Norem and her colleagues have done considerable research
on the cognitive strategies of defensive pessimism and strategic optimism
(Cantor & Norem, 1989; Norem, 1989; Norem & Cantor, 1986a, 198613,
1990; Norem & Illingworth, 1993; Spencer & Norem, 1996). Defensive
pessimism refers to a strategy individuals may use to pursue important goals:
These individuals set unrealistically low expectations and then, for a given
situation, devote considerable energy to mentally playing through or reflecting on all the possible outcomes they can imagine. Researchers have typically
contrasted defensive pessimism with strategic optimism. The latter refers to
a strategy whereby individuals set optimistic expectations for their own
performance and actively avoid extensive reflection.
Generally this research has shown that defensive pessimists perform
as well as strategic optimists, and that both groups show performance decrements and increased anxiety when prevented from using their preferred
strategies. Much of this research has also shown that strategic optimists tend
to be more satisfied and in a better mood than defensive pessimists, and
they may exhibit better health and better performance over the long term.
Often, however, a much too simplistic picture emerges from quick comparisons of outcomes in this research. Three reasons for this that are relevant
to this chapter come to mind. The first is that too little attention is given
to the fact that the individuals using these strategies are facing different
psychological situations as they approach their goals. The second is that

352

NOREM AND CHANG

this research, as with much of the research in the field, covers too restrictive
a range of variables and may ignore subtle or hidden costs of each strategy.
The third is that too much attention is paid to the particular contrast
between the outcomes of strategic optimists and defensive pessimists, when
there are other equally, or even more, relevant contrasts to consider.
People who use defensive pessimism are typically high in anxiety (on
state, trait, and physiological measures; Norem & Cantor, 1986a, 198613;
Norem & Illingworth, 1993; Spencer & Norem, 1996)-and that aspect of
intrapsychic context is fundamental to understanding the strategy. It is
tempting to think that people using defensive pessimism need to be calmed
down and reassured, and that the best thing for them is to help them become
more optimistic. In contrast, however, research suggests that it is very
important to understand the particular function of the pessimism exhibited
by people using this strategy. Its meaning and its consequences are only
seen when it is considered in conjunction with the problem that anxiety
poses for individuals motivated to pursue performance goals. Defensive pessimism (low expectations coupled with extensive reflection) offers a solution
to that problem. Defensive pessimists perform better when they are allowed to
maintain their low expectations and reflect before a task; their performance is
impaired (and they feel more anxious) if that reflective process is disrupted
(Norem & Cantor, 1986b; Norem & Illingworth, 1993; Spencer &
Norem, 1996).
Further arguing against the lets cheer em up approach to defensive
pessimists, research shows that positive mood inductions impair the performance of defensive pessimists, even though they may be effective at making
them feel better (Norem & Illingworth, 1999; Sanna, 1998). Finally, a
longitudinal study has shown significant improvement in self-esteem, as well
as better overall performance and adjustment among anxious women who
use defensive pessimism, as compared with anxious women who do not use
the strategy (Norem, 1996).
There may indeed be costs to using defensive pessimism, as well as to
other forms of pessimism, especially when one considers social outcomes.
In Western societies such as the United States, where self-enhancement
appears to be the rule, rather than the exception, an individuals expressions
of pessimism may be perceived as undesirable by more optimistic others.
Consistent with Helweg-Larsen, Sadeghian, and Webbs ( 1999) findings,
most Westerners tend to socially reject pessimists. The existence of costs
for defensive pessimism, however, does not necessarily negate its advantages
for those who use it in terms of anxiety-management and subsequent performance. In addition, the relative costs and benefits of both defensive pessimism and strategic optimism may vary significantly across culture and context. In Eastern cultures where self-effacement is normative, expressions of
optimism might be considered incongruent with the expectations of most
A VERY FULL GLASS

353

Asian people. As Chang (1996) found, Asians have a tendency to be more

pessimistic than non-Asians. Accordingly, in Eastern cultures, optimistic
individuals might incur negative interpersonal reactions from more pessimistic others. Currently, as discussed below, there is relatively little research
that has examined the potential social and personal costs of optimism.
In quick review, these results present a complex, yet coherent picture
of the costs and benefits associated with these strategies, and they begin to
suggest some problems that might be associated with trying to change the
strategies individuals use. We clearly need to take into account the other
aspects of personality and experience with which varieties of optimism and
pessimism are intertwined, as well as the varieties of outcomes for which
they might have implications, before drawing broad conclusions about what
characteristics or strategies are most desirable.

Interpersonal and Social Context

We do not have the space in this chapter to consider all the important
aspects of context that should influence how we think about applications

of optimism and pessimism research. There are, however, some examples

that we can use to illustrate briefly the importance of other aspects of context.
Both Taylor and Seligman suggest that there are particular decisionmaking contexts that benefit from the accurate, albeit negative, perspective
associated with pessimism (Seligman, 1991; Taylor & Gollwitzer, 1995).
Seligman in particular argues that there needs to be fixibility associated
with an optimistic perspective for it to be adaptive (see also Norem, 1989, on
the importance of flexibility in strategy use). Situations where the potential
downside is either relatively likely or relatively serious would seem especially to call for a balance of pessimism and optimism.
Currently, almost no research exists on the extent to which individuals
(or on which individuals) are able to achieve this kind of balance or flexibility.
One of the things likely to influence whether individuals can be flexible in
their optimism is the extent to which it may be maintained by inhibitory
or defensive processes, such as denial, repression, and projection (Norem,
1998). If and when optimism persists because of unconscious processes, then
the costs associated with those processes need to be added to our assessments
of this trait and to our understanding of how difficult individual change is
likely to be. There are at least two kinds of costs that may be associated
with what might be called defensive optimism: (a) direct health costs, such
as those associated with repression (e.g., Pennebaker, Hughes, & OHeeron,
1987); and (b) indirect costs that may accumulate precisely because less
accessible unconscious processes are likely to limit flexibility (Norem, 1998).
A number of person-perception and group process research studies
suggest that optimism, as part of the constellation of characteristics associated

354

NOREM AND CHANG

with extraversion, is highly regarded in others, at least in the United States.

This high regard may be a very important influence on the social outcomes
of optimists and pessimists. Nevertheless, very little research has examined
specifically the social interactions and social networks of optimistic and
pessimistic people, especially as they evolve over time. Following from other
research, however, there is reason to hypothesize that optimists may have
broader, yet less intense social networks than do pessimists (Norem, 1987).
Considerable research remains to be done to explore the potential costs
and benefits of optimism and pessimism that may derive from differences
in the interpersonal interactions of those possessing each characteristic.
One specific and glaring absence in the literature concerns the potential
differences in others reactions to optimism and pessimism that might be
related to gender-a question that goes beyond whether there are mean
differences, to see if the correlates and outcomes of pessimism and optimism
in different contexts vary by gender.
Though there is voluminous research suggesting the advantages of selfenhancing and optimistically biased self-perceptions (e.g., Taylor & Armor,
1996; Taylor & Brown, 1988), little has been done to examine the potential
implications of these tendencies as they influence perception of other people. One wonders whether a pervasive tendency to be optimistic about
ones self-especially if it is related to self-enhancement and even denialmight lead to ignoring or denigrating others and their potential contributions
(Weinstein & Lachendro, 1982). Seldom have we seen much discussion or
research into those kinds of potential costs. In one such rare exception,
Colvin, Block, and Funder (1995) examined the social and interpersonal
costs of self-enhancement. In one set of studies, young men and women
were rated by others at 18 years of age (Study 1) and rated again 5 years
later (Study 2). Results from both studies indicated that expressions of
self-enhancement were associated with negative evaluations as made by
independent raters as well as by friends. For men, negative evaluations
included impressions such as is guileful and deceitful and hostile toward
others, whereas for women, they included impressions such as transfers
or projects blame and keeps people at a distance. In a final study involving
opposite-gender interactions (Study 3 ) , these investigators found again that
men and women who self-enhanced were negatively rated by their peers.
The consistency of their findings led Colvin et al. (1995) to raise strong
doubts about the benefits of positive illusions:
The present analyses suggest that self-enhancement, while aiding ones
self-esteem, is over the long term an ineffective interpersonal strategy
with both friends and acquaintances and, therefore, the growth or

One exception is research that suggests that enhancing and optimistic biases extend to close others.

A VERY FULL GLASS

355

development of self. A vicious cycle is generated whereby self-enhancement is rigidly and frequently used to maintain positive self-regard but
at a continual and cumulative cost of alienating ones friends and
discouraging new acquaintances. (p. 1161)
Most research focuses on variables having to do with individual outcomes, rather than with outcomes associated with larger social groups, which
is congruent with our societal focus on individual development and achievement. Consider Ryff and Keyess (1995) model and measure of psychological
well-being measured across six different dimensions of adaptive functioning
(e.g., self-acceptance, personal growth). Although we consider their model
and measure very important given the dearth of positive outcome measures
available to researchers, it is unfortunate that only one of the six scales
explicitly assesses for outcomes that go beyond the immediate person (viz.,
Positive Relations With Others). It is our contention that just as individual
optimism may have benefits for society at large-such as might be reaped
from the can-do attitude of the stereotypical American-there might also
be costs to society associated with a lack of attention to potential negative
outcomes. (What might be saved if we more routinely asked ourselves and
others with whom we make decisions What if we are wrong?) It is much
more difficult, of course, to do research on those kinds of outcomes. It does
not follow, however, that they are necessarily less important to consider.
Let us briefly note another way in which the social aspect of optimism
and pessimism has been neglected. More than half a century ago, Sanford,
Conrad, and Franck (1946) studied facets of optimism associated with the
consequences of war. Since then, and despite Banduras (1977,1986)emphasis that expectancies tied to specific contexts are more likely to be linked
to behavior, researchers have remained slow to embrace notions of situated
optimism and pessimism. (Note that we are not referring to Armor &
Taylors [1998] notion of situated optimism as previously discussed by
Aspinwall, Richter, & Hoffman, chap. 10, this volume.) Nonetheless, some
researchers have looked at optimism associated with college life and academic performance (Prola & Stern, 1984);love relationships (Carnelley &
Janoff-Bulman, 1992);mood regulation (Catanzaro, 1993); and risk for AIDS
(Taylor et al., 1992).We anticipate in the forthcoming decades more studies
involving situated) optimism and pessimism.
Life-Span Developmental Context

As is common in personality and social psychology, college students

have provided a convenience sample for much of the research on optimism
and pessimism although there are significant exceptions to this generalization. There is research, however, that suggests that the costs and benefits

356

NOREM AND CHANG

of optimism and pessimism may vary across the life span; accordingly, we
need to be sensitive to this aspect of context as we think about research
applications. Isaacowitz and Seligman ( 1998) report that among the elderly,
a realistically pessimistic perspective is associated with better adaptation to
negative life events, in contrast to the typical findings with younger participants in the sample groups. In a related manner, Robinson-Whelen et al.
( 1997)found little evidence for the power of positive thinking in predicting
anxiety, stress, depression, and self-appraised health among an older aged
group of people composed of caregivers and noncaregivers. In addition, it
is worth noting that although dispositional optimism and pessimism have
been found to be unidimensional in college student populations (Scheier
& Carver, 1985), the relation between the two variables has consistently
been lower, and in some cases nonsignificant, in samples of people from older
populations (e.g., Mroczek, Spiro, Aldwin, Ozer, & BOSS^, 1993; RobinsonWhelen et al., 1997). For example, in a sample of older Swedish adults,
Plomin et al. (1992) found that dispositional optimism was unrelated to
dispositional pessimism ( r = -.02). It is difficult to make out what these
correlational findings may mean, although they do suggest that age (or
factors associated with age, such as life experiences) may be an important
moderator of the dimensionality of optimism and pessimism. Because increasingly more people in America and in other parts of the world are living
longer, we look forward to seeing more studies that help evaluate the costs
and benefits of optimism and pessimism among the elderly population.
On the other hand, there has been a growing push to teach children
to be more optimistic and less pessimistic (Seligman, Reivich, Jaycox, &
Gillham, 1995). Again, this has been based on the seldom-challenged assumption that optimism is good and pessimism is bad. Although this trend
has caught the attention of parents and teachers alike who are eager to find
a quick and simple panacea for dealing with problematic children, we still
know too little about the form and function of optimism and pessimism
among children to even begin meaningfully considering the short- and longterm costs and benefits of such interventions.
Helping children think more optimistically may help them feel less
sad at some particular point in time-or even over time. We do not know
very much, however, about what other effects such efforts may have over
time. In fact, similar to findings that have linked high self-esteem (another
highly regarded attribute in the minds of most parents and teachers) with
expressions of hostility and anger (see e.g., Baumeister, Smart, & Boden,
1996; Kernis, Grannemann, & Barclay, 1989), efforts to promote high
optimism and to eliminate pessimism may produce negative outcomes later
on. For example, in a study involving young adults, Kasser and Ryan (1993)
found that optimism associated with ones ability to obtain financial success

A VERY FULL GLASS

357

was negatively associated with self-actualization, vitality, and global functioning, and positively associated with anxiety and behavioral disorders (viz.,
oppositional and conduct disorders).
For another example, consider Taylor et al.s ( 1992) finding that AIDSrelated optimism was positively associated with risky sexual behaviors. Although Taylor et al.s finding was based on the study of a very specific
population (viz., gay men), we find it difficult to ignore the potential implications for adolescents and young adults who remain a high-risk group for
HIV infection (see also Kiehl, 1994).Consistent with our notion of balance
throughout this chapter, Schwartz (1992) argues that deviations in either
direction from an optimal balance of positive and negative thoughts (e.g.,
optimism and pessimism, respectively), will result in sure and certain psychopathology. Hence, it may be just as maladaptive to be excessively optimistic
as it can be to be excessively pessimistic. Accordingly, parents and teachers
may want to exercise caution before running the risk of nurturing a generation of overly optimistic children who may be at high risk for disappointment
and disturbance when faced later on with many of the hardships and realities
of modern life (e.g., parental divorce, violence, illness, poverty). In fact,
data suggests that the power of positive thinking can sometimes backfire
under stressful life conditions.
According to Brown and McGills (1989) identity disruption model,
accumulation of life events inconsistent with ones self-concept or identity is
believed to result in maladjustment (e.g., poor physical health). As expected,
these investigators found that an accumulation of positive life events was
associated with greater physical symptoms for those who held negative
compared with positive self-schemas. In contrast to this model, Chang
(1999a) set out to examine the relations among optimism, pessimism,
negative life events, and adjustment in a sample of young adults. Considering the potential costs of optimism, Chang (1999a) predicted that the
accumulation of negative life events may cause greater disturbance for
optimists than for pessimists, and thus, poorer adjustment in the former
than in the latter group. In contrast, it was expected that chronic
expectations for negative outcomes would actually help pessimists abate
the harmful influence of negative life events on adjustment. Consistent
with expectations, results of this study indicated that the associations
between negative life events and poor psychological and physical outcomes
were exacerbated for optimists, but not for pessimists. In fact, for pessimists,
experiencing greater negative life events was not even significantly related
to outcomes. Hence, underscoring Tennen and Afflecks (1987) concerns
about the potential costs of dispositional optimism, these findings indicate
that when a lot of bad things happen, optimists indeed may be particularly
vulnerable. However, because Changs (1999a) study was based on a college

358

NOREM AND CHANG

student population, the generalizability of these findings to younger populations remains an open question.
In considering environmental influences further, we should not neglect
the potential role of broad social factors on optimism. For example, Schutte,
Valerio, and Carrillo (1996) found that for White Americans, the relation
between socioeconomic status and dispositional optimism was positive and
significant. In addition, more specific contexts also must be considered. For
example, Burger and Palmer (1992) found that individuals who had recently
experienced the 1989 California earthquake tended to express greater pessimism for being seriously hurt in a natural disaster than did a group of control
participants who had not experienced the earthquake. However, the former
group did not differ from the control group in expectancies concerning a
variety of other traumatic life events unrelated to natural disasters (e.g.,
mugging, divorce, heart attack). Thus, environmental determinants of pessimism for one situation may not influence pessimism for others. In a more
recent study that tracked responses to the 1994 California earthquake,
similar findings were obtained for those people who had just experienced
the earthquake (Helweg-Larsen, 1999). Particularly notable, however, was
Helweg-Larsens finding that for those people who directly experienced the
earthquake, the specificity of their lack of optimism bias (e.g., believed that
ones chances of getting seriously injured in a future earthquake was greater
than the typical student) remained relatively unchanged across a 5-month
period, compared with those who did not have such experiences. Hence,
exposures to certain types of encounters (possibly those that are life-threatening and unpredictable), appear to have a specific and enduring influence
on peoples lack of optimism. However, it is unclear if such outcomes are
adaptive, maladaptive, or a little bit of both.
As previously discussed by Zuckerman (see chap. 8, this volume), very
few studies have looked at potential biological influences on optimism and
pessimism. Moreover, of the conducted studies, very similar methodologies
have been used, namely, comparing monozygotic and dyzygotic twins in their
expressions of optimism and pessimism. Another way to explore underlying
biological influences associated with the development of optimism and
pessimism is to consider relations between biological parents and their
children. In a recent study that focused on how optimism and pessimism
moderated the link between perceived stress and adjustment in a sample of
younger and older adults, the latter group was largely composed of biological
parents of the former group (Chang, 199913). Although not a central point
of the study, dispositional optimism was found to be significantly, albeit
very modestly, correlated between biologically related parents and their
adult children ( r = .12). Given that research in behavioral genetics indicates
that a child typically obtains half of his or her genes from each biological

A VERY FULL GLASS

359

parent, it is surprising that a more robust correlation was not found. Thus,
these findings may suggest that environmental factors (and their interaction
with biological factors) play a greater role in the development of optimism
and pessimism. No doubt, we will have to wait for the results of studies
using more rigorous methodologies, such as adoption studies, to better distinguish the influences of genetics, environment, and their interaction on
optimism and pessimism. In general, as can be seen in this section much
work remains to be done on the origins of kinds of optimism and pessimism,
as well as on their development and their consequences across the life span.

Cultural Contexts

In a study of 31 countries where people were asked if the coming year

would be better or worse than the present year, Michalos (1988) found that
the average percentage of people who expressed optimism for the coming
year across a 10-year period varied by country. For example, over a 7-year
period, an average of 50% of Americans believed that the new year would
be better compared with an average of 10% among Austrians. Cultural
differences in optimism or pessimism also can be expressed in other ways.
The first author recalls receiving voluminous mail from Eastern European
Block countries requesting reprints of the first articles on defensive pessimism-evidence that the popularity of pessimism research might vary across
cultures. Beyond interest in research, there is also good reason to believe
that the implications and consequences of optimism and pessimism-their
cultural meaning generally-vary across culture.
The United States is widely perceived as a notably optimistic society
by people in other countries. Beyond this stereotype, there has been much
theorizing about aspects of American culture that promote and value optimism, such as high social mobility and an emphasis on individualism (e.g.,
Murphy, 1947). It would not be surprising if the costs and benefits of
optimism and pessimism also varied depending on cultural context. Zane,
Sue, Hu, and Kwon (1991) found that Asian Americans had higher scores
in pessimism in some domains, but their scores were less optimistic than
White Americans in other domains, and that the particular patterns of
optimism and pessimism were consistent with differences in cultural values.
Chang (1996) also found that Asian Americans were significantly more
pessimistic than were White Americans, but not significantly less optimistic.
In addition, although pessimism was negatively associated with problem
solving and with expressing emotion coping strategies for White Americans,
it was positively associated with use of these coping strategies for Asian
Americans. He interprets these data in terms of the different cultural sensibilities of the two groups, and notes that, much as with defensive pessimists,
negative outcome expectancies are correlated with active coping among
360

NOREM A N D CHANG

Asian Americans. Changs research suggests that we be extremely cautious

about trying to adapt research findings from groups composed of White
American populations.
This research also points to the potential for cultural clashes as people
of different backgrounds try to work together, as is a common part of
American life. For example, Helweg-Larsen et al.s (1999) findings indicating
that pessimistic target participants are more likely than optimistic target
participants to be socially rejected because of peoples common assumption
that pessimists are depressed, may suggest significant social and psychological
costs for many Asian Americans, who typically do express heightened pessimism (as previously discussed by Chang, chap. 12, this volume). Therefore,
to the extent that optimism and pessimism are differently valued and interpreted by individuals from different cultural backgrounds, one would expect
significant misunderstanding and confusion in clinical, educational, workrelated, and other settings, concerning efforts to impress an optimistic viewpoint. A bolder vision and model that respects differences and similarities
between people is needed if a culturally pluralistic society such as ours is
to thrive and flourish (Epps, 1974).
Cultural contexts may also influence the malleability of optimism and
pessimism. Dispositional optimism and pessimism may be especially difficult
to change in a cultural context where these qualities are perceived to be
unchangeable. Choi, Nisbett, and Norenzayan ( 1999) have noted that,
whether from Eastern or Western cultural backgrounds, people almost universally appear to make dispositional attributions (e.g., Jane is smiling
because she is a happy person). Yet, according to Choi et al., most Asians,
unlike most Westerners, appear to perceive dispositions as changeable and
socially situated (e.g., Jane is smiling because she is a happy person ever
since she successfully completed her doctoral defense). In a similar vein,
research has shown that the American tendency to believe that math ability
is fixed, permanent, and strongly gender-related may impede efforts to decrease the differences between scores of males and females on standardized
math tests. In contrast, in some countries (e.g., China), math ability is
believed to be primarily a function of effort, and there is no gender difference
in math performance (Halpern, 1992). The implications of these findings
are too numerous to consider in the present context, but at least two relevant
points come to mind. First, efforts to change dispositional optimism and
pessimism may be more effective among those who perceive dispositions as
malleable (e.g., Easterners), than among those who may perceive dispositions
as immutable (e.g., Westerners). Second, we speculate that the negative
consequences of dispositions (or explanatory style) may be most harmful
for those who believe that dispositions, in general, are unchangeable and
unaffected by situational factors. Hence, the view that optimism is always
good and pessimism is always bad may be moderated by whether one believes
A VERY FULL GLASS

361

that once an optimist always an optimist or once a pessimist always

a pessimist.
Finally, beyond our recognition of the influences of a Westernized
culture that typically pushes Americans to see themselves better off than
others (Taylor & Brown, 1988),we hope to keep things in perspective, and
suggest another important concomitant, dare we say cause, of Americans
noted optimism. For many immigrants, coming to America has often been
associated with visions of attainable prosperity, that is, of obtaining means
to live a substantially better life for oneself and for ones family (Roger,
1990). Hence, it should not be a great surprise if some of the most optimistic
people on earth have come to make America their home. By the same
token, recall the heightened pessimism of Asians discussed earlier. In Japan
where long work hours are the rule rather than the exception, people have
suffered from cardiovascular attacks and in some cases died due to karoshi,
extreme work overload (e.g., Nakagawa & Sugita, 1994; Uehata, 1991).
Thus, for some Asians, pessimism may be a realistic reflection of the demanding conditions of their lives.

SUMMARY AND CONCLUSION

We hope it is clear that the point of this chapter is not to refute the
voluminous literature that relates optimism to good outcomes. Rather, our
point is simply to suggest that we embrace the complexity in that literature
and recognize the gaps in our knowledge, especially in any attempt to apply
research findings in real life. As researchers who have focused on pessimism
as well as on optimism, we are perhaps especially sensitive to the zeitgeist
that tends to present optimism as uniquely and completely rose-colored.
W e feel strongly that, in fact, the picture is multicolored. The benefits and
costs of optimism and pessimism depend on individual personality; the kind
of optimism or pessimism about which one speaks; the kinds of outcomes
one focuses on; and the particular social, cultural, and life span contexts.
Thorough and thoughtful consideration of these aspects of context ought
to lead to more insightful research into the costs and benefits of optimism
and pessimism, as well as to more effective applications of that research for
the betterment of all people.

REFERENCES
Armor, D. A., & Taylor, S. E. (1998). Situated optimism: Specific outcome expectancies and self-regulation. In M. P. Zanna (Ed.), Advances in experimental
social psychology (Vol. 30, pp. 309-379). New York: Academic Press.

362

NOREM AND CHANG

Bandura, A. (1977). Self-efficacy: Toward a unifying theory of behavioral change.

Psychological Review, 84, 191-215.
Bandura, A. (1986). Social foundations of thought and action: A social cognitive theory.
Englewood Cliffs, NJ: Prentice-Hall.
Baumeister, R. F. (1989). The optimal margin of illusion. Journal of Social and
Clinical Psychology, 8, 176-189.
Baumeister, R. F., Smart, L., & Boden, J. M. (1996). Relations of threatened egotism
to violence and aggression: The dark side of high self-esteem. Psychological
Review, 103, 5-33.
Brown, J. D., & McGill, K. L. (1989). The cost of good fortune: When positive
life events produce negative health consequences. Journal of Personality and
Social Psychoogy, 57, 1103-11 10.
Burger, J. M., & Palmer, M. L. (1992). Changes in and generalization of unrealistic
optimism following experiences with stressful events: Reactions to the 1989
California earthquake. Personality and Social Psychology Bulletin, 18, 39-43.
Cantor, N., & Norem, J. K. (1989). Defensive pessimism and stress and coping.
Social Cognition, 7, 92-112.
Carnelley, K. B., & Janoff-Bulman, R. (1992). Optimism about love relationships:
General vs specific lessons from ones personal experiences. Journal of Social
and Personal Relationships, 9, 5-20.
Catanzaro, S. J. ( 1993). Mood regulation expectancies, anxiety sensitivity, and
emotional distress. Journal of Abnormal Psychology, 102, 327-330.
Chang, E. C. (1996). Cultural differences in optimism, pessimism, and coping:
Predictors of subsequent adjustment in Asian American and Caucasian American college students. Journal of Counseling Psychology, 43, 113-123.
Chang, E. C. (1998). Distinguishing between optimism and pessimism: A second
look at the optimism neuroticism hypothesis. In R. R. Hoffman, M. F.
Sherrik, & J. S. Warm (Eds.), Viewing Psychology as a whole: The integrative
science of William N. Dember (pp. 415-432). Washington, DC: American
Psychological Association.
Chang, E. C. (1999a). Optimism, pessimism, negative life events, and psychological and
physical adjustment: When expectingthe best, but experiencingthe worst, is associated
with negative outcomes. Unpublished manuscript, Department of Psychology,
University of Michigan, Ann Arbor.

Chang, E. C. (199913). Optimism-pessimism and stress appraisal in younger and older

adults: Testing a cognitive interactive model of psychological adjustment. Unpublished manuscript, Department of Psychology, University of Michigan, Ann
Arbor.
Chang, E. C., Maydeu-Olivares, A., & DZurilla, T. J. (1997). Optimism and
pessimism as partially independent constructs: Relations to positive and negative affectivity and psychological well-being. Personality and Individual Differences, 23, 433-440.
A VERY FULL GLASS

363

Choi, I., Nisbett, R. E., & Norenzayan, A. (1999). Causal attribution across cultures:
Variation and universality. Psychological Bulletin, 125, 47-63.
Colvin, C. R., & Block, J. (1994). Do positive illusions foster mental health?
An examination of the Taylor and Brown formulation. Psychological Bulletin,
116, 3-20.
Colvin, C. R., Block, J., & Funder, D. C. (1995). Overly positive self-evaluations
and personality: Negative implications for mental health. journal of Personality
and Social Psychology, 68, 1152-1162.
Davidson, K., & Prkachin, D. (1997). Optimism and unrealistic optimism have
an interacting impact of health-promoting behavior and knowledge changes.
Personality and Social Psychology Bulletin, 23, 617-625.
Dolinski, D., Gromski, W., & Zawisza, E. (1987). Unrealistic pessimism. journal of
Social Psychology, 127, 511-516.
Epps, E. G. (1974). Cultural pluralisni. Berkeley, CA: McCutchan Publishing.
Epstein, S., & Meier, P. ( 1989).Constructive thinking: A broad coping variable with
specific components. Journal of Personality and Social Psychology, 57, 332-350.
Frese, M. (1992). A plea for realistic pessimism: On objective reality, coping with
stress, and psychological dysfunction. In L. Montada, S.-H. Filipp, & M. J.
Lerner (Eds.), Life crises and experiences ofloss inadulthood (pp. 81-94). Hillsdale,
NJ: Erlbaum.
Halpern, D. F. (1992). Sex differences in cognitive abilities. Hillsdale, NJ: Erlbaum.
Helweg-Larsen, M. (1999). (The lack of) optimistic biases in response to the 1994
Northridge earthquake: The role of personal experience. Basic and Applied
Social Psychology, 21, 119-129.
Helweg-Larsen, M., Sadeghian, P., Webb, M. S. (1999). The stigma of being
pessimistically biased. Unpublished manuscript, Transylvania University, Lexington, KY.
Isaacowitz, D. M., & Seligman, M. E. P. (1998). Prevention of depression in older
adults: Theory, methodology and pitfalls. Invited paper presented in Opportunities
and Pitfalls in Adult and Older Adult Prevention Research. Symposium at
the Annual Meeting of the American Psychological Association, San Francisco, CA.
Kasser, T., & Ryan, R. M. (1993). A dark side of the American dream: Correlates
of financial success as a central life aspiration. journal of Personality and Social
65, 410-422.
Psych~log~,
Kelman, H. ( 1945). Neurotic pessimism. Psychoanalytic Review, 32, 419-448.
Kemis, M. H., Grannemann, B. D., &. Barclay, L. C. (1989). Stability and level
of self-esteem as predictors of anger arousal and hostility. Journal of Personality
and Social Psychology, 56, 1013-1022.
Kiehl, E. (1994). Attitudes towards AIDS among unrealistic optimists, health optimists,
and health defensive pessimists. Unpublished honors thesis. Psychology Department, Wellesley College, Wellesley, MA.

364

NOREM AND CHANG

Marshall, G. N., Wortman, C. B., Kusulas, J. W., Hervig, L. K., & Vickers, R. R.,
Jr. ( 1992). Distinguishing optimism from pessimism: Relations to fundamental
dimensions of mood and personality. Journal of Personality and Social Psychology,
62, 1067-1074.
Michalos, A. C. (1988). Optimism in thirty counties over a decade. Social Indicators
Research, 20, 177-180.
Mroczek, D. K., Spiro, A., Aldwin, C. M., Ozer, D. J., & B o d , R. (1993). Construct
validation of optimism and pessimism in older men: Findings from the normative aging study. Health Psychology, 12, 406-409.
Murphy, G. (1947). Personality: A biosocial approach
York: Harper & Brothers.

to origins

and structure. New

Nakagawa, T., & Sugita, M. (1994). Life style changes and psychosomatic problems
in Japan. Homeostasis in Health and Disease, 35, 180-189.
Norem, J. K. (1987). Strategic realities: Optimism anddefensive pessimism. Unpublished
doctoral dissertation, University of Michigan, Ann Arbor.
Norem, J. K. ( 1989). Cognitive strategies as personality: Effectiveness, specificity,
flexibility,and change. In D. M. Buss &N. Cantor (Eds.), Personality psychology:
Recent trends and emerging issues (pp. 45-60). New York: Springer-Verlag.
Norem, J. K. (1996, August). Cognitive strategies and the rest ofpersonality. Paper
presented at the Annual Meeting of the American Psychological Association,
Toronto, Canada.
Norem, J. K. (1998). Should we lower our defenses about defense mechanisms?
Journal of Personality, 66, 895-917.
Norem, J. K., & Cantor, N. (1986a). Anticipatory and post hoc cushioning strategies: Optimism and defensive pessimism in risky situations. Cognitive Therapy
and Research, 10, 347-362.
Norem, J. K., & Cantor, N. (198613). Defensive pessimism: Harnessing anxiety as
motivation. Journal of Personality and Social Psychology, 51, 1208-1217.
Norem, J. K., & Cantor, N. (1990). Cognitive strategies, coping and perceptions
of competence. In R. J. Sternberg & J. J. Kolligian (Us.), Competence considered
(pp. 190-204). New Haven, CT:Yale University Press.
Norem, J. K., & Illingworth, K. S. S. (1993). Strategy-dependent effects of reflecting
on self and tasks: Some implications for optimism and defensive pessimism.
Journal of Personality and Social Psychology, 65, 822-835.
Norem, J. K., & Illingworth, K. S. S. (1999). Mood and perjomnce among strategic
optimists and defensive pessimists. Unpublished manuscript, Department of Psychology, Wellesley College, Wellesley, MA.
Park, C. L., Moore, P. J., Turner, R. A., & Adler, N. E. (1997). The roles of
constructive thinking and optimism in psychological and behavioral adjustment during pregnancy. Journal of Personality and Social Psychology, 73, 584592.
A VERY FULL GLASS

365

Pennebaker, J. W., Hughes, C. F., & O'Heeron, R. C. (1987). The psychophysiology

of confession: Linking inhibitory and psychosomatic processes. journal of Personality and Social Psychology, 52, 781-793.
Peterson, C., & Chang, E. C. (in press). Optimism and flourishing. To appear in
C. L. M. Keyes (Ed.), Flourishing: The positive person and the good life. Washing
ton, DC: American Psychological Association.
Peterson, C., & Seligman, M. E. P. (1987). Explanatory style and illness. journal
of Personality, 55, 237-265.
Plomin, R., Scheier, M. F., Bergeman, C. S., Pedersen, N. L., Nesselroade, J. R.,
& McCleam, G. E. (1992). Optimism, pessimism and mental health: A twin/
adoption analysis. Personality and individual Differences, 13, 921-930.
Prola, M., & Stem, D. (1984). Optimism about college life and academic performance in college. Psychological Reports, 55, 347-350.
Raikkonen, K., Matthews, K. A., Flory, J. D., Owens, J. F., & Gump, B. B. (1999).
Effects of optimism, pessimism and trait anxiety on ambulatory blood pressure
and mood during everyday life. Journal of Personality and Social Psychology,
76, 104-113.
Robinson-Whelen, S., Kim, C., MacCallum, R. C., & Kiecolt-Glaser, J. K. (1997).
Distinguishing optimism from pessimism in older adults: Is it more important
to be optimistic or not to be pessimistic? journal of Personality and Social
P~y~hology,
73, 1345-1353.
Roger, D. (1990). Coming to America: A history of immigration and ethnicity in
American life. New York: HarperCollins.
Ryff, C. D., & Keyes, C. L. M. (1995). The structure of psychological well-being
revisited. journal of Personality and Social Psychology, 69, 719-727.
Sanford, R. N., Conrad, H. S., & Franck, K. (1946). Psychological determinants
of optimism regarding the consequences of war. Journal of Psychology, 22, 207235.
Sanna, L. J. (1998). Defensive pessimism and optimism: The bitter-sweet influence
of mood on performance and prefactual and counterfactual thinking. Cognition
and Emotion, 12, 635-665.
Scheier, M. F., & Carver, C. S. (1985). Optimism, coping, and health: Assessment
and implications of generalized outcome expectancies. Health Psychology, 4,
219-247.
Scheier, M. F., Carver, C. S., & Bridges, M. W. (1994). Distinguishing optimism
from neuroticism (and trait anxiety, self-mastery, and self-esteem): A reevaluation of the Life Orientation Test. journal of Personality and Social Psychology,
67, 1063-1078.
Schulz, R., Bookwala, J., Knapp, J. E., Scheier, M., & Williamson, G. M. (1996).
Pessimism, age, and cancer mortality. Psychology and &ng, 1 1 , 304-309.
Schutte, J. W., Valerio, J. K., & Carrillo, V. (1996). Optimism and socioeconomic
status: A cross-cultural study. Social Behuvior and Personality, 24, 9-18.

366

NOREM AND CHANG

Schwartz, R. M. (1992). State of mind model and personal construct theory: Implications for psychopathology. InternationalJournalof Personal Construct Psychology,
5 , 123-143.
Seligman, M. E. P. (1991). Learned optimism. New York: Knopf.
Seligman, M. E. P., Reivich, K., Jaycox, L. H., & Gillham, J. (1995). The optimistic
child. Boston: Houghton-Mifflin.
Spencer, S. M., & Norem, J. K. (1996). Reflection and distraction: Defensive
pessimism, strategic optimism, and performance. Personality and Social Psychology Bulletin, 22, 354-365.
Taylor, S. E. (1989). Positive ilwions: Creative self-deceptionand the healthy mind.
New York: Basic Books.
Taylor, S. E., & Armor, D. A. (1996). Positive illusions and coping with adversity.
Journal of Personality, 64,873-898.
Taylor, S. E., & Brown, J. D. (1988). Illusion and well-being: A social psychological
perspective on mental health. Psychological Bulletin, 103, 193-210.
Taylor, S. E., & Gollwitzer, R. M. (1995). The effects of mindset on positive
illusions. Journal of Personality and Social Psychology, 69, 213-226.
Taylor, S. E., Kemeny, M. E., Aspinwall, L. G., Schneider, S. G., Rodriguez, R.,
& Herbert, M. (1992). Optimism, coping, psychological distress, and high-risk
sexual behavior among men at risk for acquired immunodeficiency syndrome
(AIDS). Journal of Personality and Social Psychology, 63, 460-473.
Tennen, H., & Affleck, G. (1987). The costs and benefits of optimistic explanations
and dispositional optimism. Journal of Personality, 55, 376-393.
Uehata, T. ( 1991).Long working hours and occupational stress-relatedcardiovascular attacks among middle-aged workers in Japan. Journal of Human Ergobgy,
20, 147-153.
Weinstein, N. D. (1980). Unrealistic optimism about future life events. Journal of
Personality and Social Psychology, 39, 806-820.
Weinstein, N. D., & Lachendro, E. (1982). Egocentrism as a source of unrealistic
optimism. P e r s d t y and Social Psychology Bulletin, 8, 195-200.
Zane, N. W. S., Sue, S., Hu, L., & Kwon, J. (1991). Asian-American assertion: A
social learning analysis of cultural differences.Journal of Counseling Psychology,
38, 63-70.

A VERY FULL GLASS

367