Cambridge University Press

978-0-521-87133-4 — Cognitive Neurorehabilitation

2nd Edition
Excerpt
More Information

SECTION 1

Principles of cognitive
neurorehabilitation

© in this web service Cambridge University Press www.cambridge.org

Cambridge University Press
978-0-521-87133-4 — Cognitive Neurorehabilitation
2nd Edition
Excerpt
More Information

Introduction to Section 1

Gordon Winocur

Cognitive neurorehabilitation is predicated on two neurorehabilitation. For example, an understanding

fundamental principles: (1) that the brain has of alterations at the neuronal or neurotransmitter
within it an inherent plasticity that enables it to level are critical to the development of pharmaco-
recover from damage that gives rise to cognitive logical therapies, whereas an appreciation of reor-
impairment, and (2) that individuals have the ganisation of neural circuitry through neuroimaging,
capacity to make behavioral adjustments necessi- may be more relevant to designing behavioral thera-
tated by changing circumstances. At the same time, pies that invoke the use of appropriate strategies. The
the development of the field rests on recognition chapter broadly reviews research that demonstrates
of and adherence to other essential principles. that brain plasticity is influenced by a wide range of
Foremost are those principles that guide the con- factors that include age, etiology of damage, and type
duct and assessment of research that ultimately of experiences. Kolb and Gibb end with the interest-
fuels the development of successful rehabilitation ing observation that brain plasticity may not always
programs. Too often, these principles receive insuf- work in the patient’s interest and cite, as an example,
ficient attention and, as a result, the foundations on counter-productive changes that occur in relation to
which treatment programs rest can be a little shaky. developmental disorders. This is important for cog-
It is appropriate then that the first section of this nitive neurorehabilitation because brain-damaged
book, which is intended to be comprehensive and individuals often develop maladaptive responses
broader in scope than the first edition, is devoted to and their neurophysiological imprint must be over-
some of the over-riding principles that are central to come for rehabilitation to proceed successfully.
cognitive neurorehabilitation, and to its continued In their chapter, Hunter and McEwen focus on
development as a form of clinical practice within the relationship between adverse life experiences
rehabilitation medicine. and chronic dysregulation of the neuroendocrine
The first two chapters cover neuroplasticity and axis which, combined, increase the risk of various
the brain’s potential for reorganization in ways that diseases (e.g., type 2 diabetes, stroke), that include
make optimal use of residual resources in promot- cognitive impairment as part of their symptomatol-
ing cognitive recovery. Kolb and Gibb review prin- ogy. Attention is directed to stress-induced abnor-
ciples that govern brain plasticity and factors that mal release of adrenal steroids (e.g., cortisol) and
affect its expression following regional damage. In their toxic effects on brain regions such as the hip-
describing some of the cellular and physiological pocampus. In contrast, there may be benefits from
mechanisms underlying plasticity, Kolb and Gibb treatment with gonadal steroids (e.g., androgens,
make the important point that the level of analysis estrogens). Hunter and McEwen acknowledge the
must depend on the questions being asked. controversy in this area but also remind us of the
This has important implications for cognitive considerable evidence linking such treatment to

Cognitive Neurorehabilitation, Second Edition: Evidence and Application, ed. Donald T. Stuss, Gordon Winocur and
Ian H. Robertson. Published by Cambridge University Press. © Cambridge University Press 2008.

© in this web service Cambridge University Press www.cambridge.org

Cambridge University Press
978-0-521-87133-4 — Cognitive Neurorehabilitation
2nd Edition
Excerpt
More Information

4 Section 1. Principles of cognitive neurorehabilitation

positive effects on dendritic growth and synaptic exploratory studies have a rightful place in the
function in the hippocampus. Under some circum- investigative process and, properly conducted, can
stances, the authors suggest, this form of cognitive even help to shape RCTs, rendering them more
therapy may be warranted. Hunter and McEwen informative in the larger picture. A substantial por-
assess the potential for pharmacologic treatment tion of the chapter is devoted to Rothi’s multistage
of cognitive problems that are related to metabolic model for conducting neurorehabilitation research.
disorders and, in line with other contributors (e.g., According to the model, in the initial discovery stage,
Kolb and Gibb; Dawson and Winocur; Dixon, clear hypotheses are proposed and evaluated at the
Garrett and Bäckman) stress the importance of basic science level. This stage might involve animal
combining such interventions with a healthy life- models or normal humans but, in the second, trans-
style, physical exercise, and a psychosocial environ- lation stage, the research is extended to the targeted
ment that is both stimulating and supportive. populations. The third, innovation, stage entails
Consistent with the brain’s capacity for reorganiza- exploratory clinical study using the principles estab-
tion is the principle of compensation, whereby cog- lished in stages 1 and 2. At this point, a treatment
nitive abilities are recovered through a combination should be ready for a phase I clinical trial (evalua-
of adaptive changes in brain function and behavior. tion/formalization stage), in which optimal condi-
Dixon et al. provide a lucid account of conditions that tions are created for demonstrating a treatment
give rise to compensatory approaches and the mech- effect. A positive outcome would lead to a more
anisms by which compensation is mediated. On a rigorous phase II trial (efficacy stage) that would
cautionary note, they point out that, because of take into account effect sizes and possibly involve
constraints and diminished resources resulting multiple sites. A phase III clinical trial (effectiveness
from damage to functional systems, successful stage) would then be conducted to establish the full
compensation in one area may entail losses in potential of the treatment program. Finally, it is
other areas. Because these losses may have negative necessary to consider issues related to delivery of
consequences, they must be taken into account in the program and its impact, taking into account
rehabilitation programs aimed at promoting com- practical considerations such as costs and benefits.
pensation. As well, Dixon et al. emphasize the Cicerone takes on a similar cause in his probing
important interplay between compensation that chapter. He too acknowledges the important ad-
occurs between neural and behavioral levels, and vances made through RCTs and, like Rodriguez
the need for them to influence each other if optimal and Rothi, argues that there are important benefits
benefits are to be achieved. The authors cite several to be derived from other approaches. He suggests,
studies of behavioral training-induced cognitive for example, that observational studies may be even
improvements and, importantly, provide related more useful than RCTS in certain situations – such
evidence that such improvements are accompanied as evaluating behaviors that have a relatively low
by stable and long-lasting biological changes. Dixon rate of occurrence, or in assessing naturally occur-
et al. further underscore the importance of compen- ring services. Cicerone’s position is that we need
sation to rehabilitation practice by arguing for the more good RCTs in neurorehabilitation research
universality of the process and its relevance to dif- and, to achieve that, investigators must strive to
ferent types of injury and abnormality. avoid design pitfalls which occur all too frequently.
Ultimately, progress in neurorehabilitation prac- As examples, he cites unsuccessful randomization,
tice will flow from quality research and this is the failure to employ double-blind designs and masked-
focus of Rodriguez and Rothi’s thought-provoking outcome assignments, and inadequate long-term
chapter. The authors acknowledge the randomized follow-up. Moreover, it is important to be cognizant
control trial (RCT) as the gold standard in rehabil- of the need to report fully all relevant procedures,
itation research. At the same time, they argue that provide information on treatment compliance and

© in this web service Cambridge University Press www.cambridge.org

Cambridge University Press
978-0-521-87133-4 — Cognitive Neurorehabilitation
2nd Edition
Excerpt
More Information

Introduction to Section 1 5

treatment integrity, and, in the interest of repro- In Chapter 3 of this section, Stuss and Binns bring
ducibility, to use accessible outcome measures. into focus the fundamental principle of behavioral
The chapters by Cicerone and Rodriguez and Rothi variability, as it applies to cognitive neurorehabili-
contain important messages and provide a valu- tation. It is well established that cognitively
able road-map for conducting scientifically rigor- impaired patients can be extremely variable in the
ous rehabilitation research that ultimately will expression of their cognitive problems. It follows
lead to improved treatment programs and better that variability should be taken into account rou-
outcomes. tinely in assessing recovery or the effects of cogni-
As Kolb and Gibb point out in their chapter, a tive neurorehabilitation. Yet, it is not and, as Stuss
critical consideration in evaluating cognitive neuro- and Binns point out, to assess outcome in a single
rehabilitation programs is in the selection of out- snapshot in time, runs the risk of masking benefits
come measures and this is the focus of Lincoln or, conversely, giving an exaggerated impression of
and Nair’s chapter. They begin by emphasizing improvement. Because it is so tempting to mistak-
the need to distinguish between impairments enly conclude positive effects following a single
induced by brain damage and changes in functional assessment, Stuss and Binns label variability as a
performance – neurorehabilitation may be effective “silent disorder.” Their highly enlightening chapter
in improving performance without necessarily is particularly informative on intra-individual vari-
affecting the fundamental impairment. If assess- ability, which typically receives less attention than
ment focused exclusively on cognitive impairment inter-individual variability. They discuss various
(as measured by standard cognitive tests), the ben- ways of measuring intra-individual variability (e.g.,
efits of the program may not be fully appreciated or between-task scatter; within-task dispersion), as
even detected. In selecting cognitive instruments well as mediating factors (e.g., structural – fatigue,
for measuring outcome, Lincoln and Nair empha- rhythmic changes; task demands; personality; age).
size established criteria (e.g., reliability, validity, They also provide insights into possible mecha-
sensitivity to program-induced change and practi- nisms and, in the process, distinguish between fron-
cality) and, in addition, remind us that not all tests tal lobe-mediated, general control processes, and
are well suited for assessing outcome. Some are damage-specific control processes that are said to
better suited as screening or diagnostic instruments. be regulated by brain regions associated with the
They go on to review strengths and weaknesses of functions affected. The chapter ends with a descrip-
assessment techniques in various cognitive domains tion of individual cases that dramatically make the
(e.g., attention, memory, executive function). The authors’ point that variability is a critical index of
authors also underscore the need for ecologically cognitive impairment that must be factored into
valid tests but caution that sometimes such tests neurorehabilitation programs and treatment
(e.g., the Rivermead Behavioral Memory Test) only assessment. It remains to be seen, of course, whether
indirectly measure performance in everyday activ- variability measurement will prove practical in diag-
ities. Finally, they point out that, although participa- nosis and in assessing outcome. Nevertheless, in
tion in a wide range of activities in a social context is convincingly showing that increased variability is
the ultimate aim of cognitive neurorehabilitation, an inherent part of the pathology associated with
few measures of participation rate actually exist, cognitive impairment and the recovery process,
and that progress in this area must be considered a Stuss and Binns have made a strong case for inves-
research priority. tigating the possibilities.

© in this web service Cambridge University Press www.cambridge.org

Cambridge University Press
978-0-521-87133-4 — Cognitive Neurorehabilitation
2nd Edition
Excerpt
More Information

Principles of neuroplasticity and behavior

Bryan Kolb and Robbin Gibb

Introduction Brain plasticity takes advantage of a basic,

but flexible, blueprint for cerebral organization
Behavioral neuroscience spent much of the twentieth that is formed during development
century seeking the fundamental rules of cerebral
The process of brain development is a remarkable
organization. One underlying assumption of much
feat of nature. Billions of neurons and glia must be
of that work was that there is constancy in cerebral
generated, they must migrate to their correct loca-
organization and function, both between and within
tions, and they must form neuronal networks that
mammalian species (e.g., Kaas, 2006). One unex-
can underlie functions that range from as simple as
pected principle to emerge, however, was that
postural reflexes to complex thought. Although a
although there is much constancy in cerebral func-
complete genetic blueprint for neuronal organiza-
tioning, there is remarkable variability as well. This
tion might be possible for a simple creature like the
variability reflects the brain’s capacity to alter its
nematode Caenorhabditis elegans, which has a total
structure and function in reaction to environmental
of 302 neurons, it is not remotely possible for the
diversity, thus reflecting a capacity that is often
mammalian brain to have a specific blueprint (Katz,
referred to as brain plasticity. Although this term is
2007). The best that nature can be expected to do is
now commonly used in psychology and neuro-
to produce a rough blueprint of cerebral organiza-
science, it is not easily defined and is used to refer to
tion that must be shaped by experience in order for
changes at many levels in the nervous system ranging
animals to exploit specific ecologies, including cul-
from molecular events, such as changes in gene
tures. The disadvantage of such flexibility is that it is
expression, to behavior (e.g., Shaw & McEachern,
possible to make errors, but this problem is certainly
2001). The relationship between molecular or cellular
outweighed by the advantage of having a brain that
changes and behavior is by no means clear and is
can learn complex motor or perceptual skills that
plagued by the problems inherent in inferring causa-
could scarcely have been anticipated by evolution
tion from correlation. Nonetheless, we believe that it
thousands or even millions of years before.
is possible to identify some general principles of brain
plasticity and behavior. As we do so we will attempt to
link these principles to potential clinical implications.
Cerebral functions are both localized
and distributed
Assumptions underlying brain plasticity One of the great issues in the history of brain
research relates to whether functions are discretely
As we consider the principles of brain plasticity, we localized in the brain (for a review, see Kolb &
need to consider five underlying assumptions that Whishaw, 2001). The resolution to this debate was
will color our perspective. important because the degree of localization of

Cognitive Neurorehabilitation, Second Edition: Evidence and Application, ed. Donald T. Stuss, Gordon Winocur and
Ian H. Robertson. Published by Cambridge University Press. © Cambridge University Press 2008.

© in this web service Cambridge University Press www.cambridge.org

Cambridge University Press
978-0-521-87133-4 — Cognitive Neurorehabilitation
2nd Edition
Excerpt
More Information

Chapter 1. Principles of neuroplasticity and behavior 7

function places constraints on the potential extent studying brain plasticity is to choose a surrogate
of functional plasticity. The more distributed a marker that best suits the question being asked.
function, the greater the likelihood that the neural Changes in potassium channels may be perfect for
networks underlying the function will be flexible studying presynaptic changes at specific synapses
after a brain injury. As we enter the twenty-first that might be related to simple learning in inverte-
century it is clear that functions are at once localized brates (e.g., Kandel, 1979; Lukowiak et al., 2003;
and distributed. Consider language. Although there Roberts & Glanzman, 2003) but are impractical for
are discrete language zones in the cortex, language understanding sex differences in language process-
is much more distributed across the cortex than ing. The latter might best be studied by in vivo
would have been expected from the classical neur- imaging or postmortem analysis of cell morphology
ologists (e.g., Geschwind, 1972). But there are limits (e.g., Jacobs et al., 1993). Neither level of analysis is
to distributed functions, especially in the sensory “correct.” The appropriate level must be suited for
systems. For example, information coming to the the research question at hand.
occipital lobe travels from the eye to subcortical One convenient surrogate for synaptic change in
areas, then to Visual area 1 (V1) where it is pro- laboratory studies of brain and behavior is dendritic
cessed, and then is sent on to other visual regions morphology. In this type of study entire neurons are
such as V2 and on to V3 etc. If V1 is only partially stained with a heavy metal (gold, silver or mercury)
damaged, V2 will still receive some input and can and the dendritic space is calculated (Figure 1.1). It
function, albeit not normally. Further, after partial is assumed that by knowing the space available for
damage, neural networks in V1 and V2 could reor-
ganise and possibly facilitate some type of func-
tional improvement. But if V1 is completely (or
substantially) damaged, downstream visual areas,
such as V2, will not be provided with appropriate
inputs and no amount of reorganization in V2 could
generate functional recovery. The partial localiza-
tion of functions thus places significant constraints
upon plasticity and recovery of function.

Changes in the brain can be shown at many

levels of analysis
Although it is ultimately the activity of neuronal
networks that controls behavior, and thus changes
in neuronal network activity that are responsible for
behavioral change, there are many ways to examine
changes in the activity of networks. Changes may be
inferred from global measures of brain activity, such
as in the various forms of in vivo imaging, but such
changes are far removed from the molecular pro-
cesses that drive them. Changes in the activity of Figure 1.1. Example of a Golgi-Cox stained pyramidal cell
networks likely reflect changes at the synapse but from layer III of the parietal cortex of the rat. A. Higher
changes in synaptic activity must result from more power magnification showing spines on an apical branch.
molecular changes such as modifications in chan- B. Higher power magnification showing spines on a basilar
nels, gene expression, and so on. The problem in branch. (Photograph courtesy of Grazyna Gorny.)

© in this web service Cambridge University Press www.cambridge.org

Cambridge University Press
978-0-521-87133-4 — Cognitive Neurorehabilitation
2nd Edition
Excerpt
More Information

8 Section 1. Principles of cognitive neurorehabilitation

synapses it is possible to infer associations between that were related to the observed sensitization we
synaptic organization and behavior – notwithstanding might find a change in synapse number in some
the problems inherent in correlational studies dis- discrete brain region such as the nucleus accum-
cussed below. The studies of Jacobs and Scheibel bens (NAcc). Both the behavioral change and the
(Jacobs et al., 1993; Jacobs & Scheibel, 1993) provide synaptic change are correlates of the drug adminis-
a good example. These researchers examined the tration. But what is the relationship between the
dendritic morphology of pyramidal neurons in post- behavioral and synaptic change? We can conclude
mortem brains of people whose educational and that the drug caused the behavioral change but it is
employment history was known. Comparison of less clear that the drug directly caused the neuronal
synapse numbers in the posterior speech zone of change. Perhaps the behavioral change caused the
people with university education, high-school edu- neuronal change or maybe both were related to
cation, or less than high-school education showed some other change in the brain. Thus, a common
that there were progressively more synapses on the criticism of studies trying to link neuronal changes
neurons from brains with more education. The to behavior is that “they are only correlates.” This is
study cannot tell us why this correlation is present true but it is hardly a reason to dismiss such studies.
but it tells us that there is some relationship The task is to try to break the correlation by showing
between experience and synaptic organization. that one change can occur without the other. The
presence of such evidence would disconfirm caus-
ality but, unfortunately, the failure to break the
To be functionally meaningful, changes
correlation is not proof of causation. Ultimately
reflecting brain plasticity must persist for at
the proof would be in showing how the synaptic
least a few days
changes arose, which would presumably involve
Changes in neuronal activity related to brain plasti- molecular analysis such as a change in gene tran-
city may be of limited duration, perhaps in the order scription. For many studies this would be an
of seconds or milliseconds. While such changes are extremely difficult challenge and often impractical.
interesting in their own right, we are focusing our It is our view that once we understand the “rules”
attention on longer-lasting changes that persist for that govern neuronal and behavioral change, we will
at least a few days. This is a practical assumption as be better able to look for molecular changes.
we think about how experiences might be related to Furthermore, we argue that a certain level of ambi-
chronic behavioral changes seen after brain injury guity in the degree of causation is perfectly justifi-
or with addiction. able at this stage of our knowledge. Understanding
the precise mechanism whereby the synaptic
changes might occur is not necessary to proceed
Correlation is not a four-letter word
with further studies aimed at improving functional
By its very nature, behavioral neuroscience searches outcome.
for neuronal correlates of behavior. Some of these
changes are directly associated with behavior but
others are more ambiguous. Consider an example. Principles of brain plasticity
If an individual is given a psychoactive drug we may
see an obvious acute behavioral change such as Although it is presumptuous to try to identify basic
increased motor activity. If the drug is taken repeat- principles of brain plasticity when so much is still
edly, we may see that there is an escalating increase unknown, we believe that the progress over the past
in the drug-dependent hyperactivity, a phenom- decade allows us to begin to identify some of the
enon referred to as drug-induced behavioral sensiti- rules underlying brain plasticity. These principles
sation. If we were to look for changes in the brain should be seen as a work in progress that will

© in this web service Cambridge University Press www.cambridge.org

Cambridge University Press
978-0-521-87133-4 — Cognitive Neurorehabilitation
2nd Edition
Excerpt
More Information

Chapter 1. Principles of neuroplasticity and behavior 9

undoubtedly be expanded and further demonstra- Table 1.1. Factors affecting the synaptic organization
ted over the next decade. of the normal brain

Factor Reference
When the brain changes, this is reflected in
behavioral change Sensory and motor experience Greenough & Chang,
1989
The primary function of the brain is to produce Task learning Greenough & Chang,
behavior but behavior is not constant. We learn 1989
and remember, we create new thoughts or images, Gonadal hormones and stress Stewart & Kolb, 1988
and we change throughout our lifetime. All of these hormones
processes require changes in neural networks. It Psychoactive drugs (e.g., Robinson & Kolb,
follows that whenever neural networks change, stimulants, THC) 2004
behavior, including mental behavior, will also Neurotrophic factors (e.g., NGF, Kolb et al., 1997
bFGF)
change. A corollary of this principle is that in order
Natural rewards (e.g., social Fiorino & Kolb, 2003
to change behavior we must change the brain. This
interaction, sex)
latter idea is especially important as we search for
Aging Kramer et al., 2004
treatments for brain injuries or diseases. Stress McEwen, 2005
Anti-inflammatories (e.g., COX-2 Silasi & Kolb, 2007
inhibitors)
Plasticity is found in all nervous systems and
Diet (e.g., choline) Meck & Williams,
the principles are conserved
2003
Even the simplest animals, such as the nematode Electrical stimulation: kindling Teskey et al., 2001
C. elegans, can show simple learning that is corre- Long-term potentiation Ivanco et al., 2000
lated with neuronal plasticity (e.g., Rose & Rankin, Direct cortical stimulation Teskey et al., 2004
2001). Similarly, there is now an extensive literature
showing neuronal and other changes in invertebrates
such as sea snail Aplysia during simple learning, inclu- enduring changes, ranging from general sensory-
ding associative learning. Furthermore, it now has motor experience to psychoactive drugs to electrical
become clear that both simple and complex nervous brain stimulation (see Table 1.1). The bulk of these
systems show both pre- and postsynaptic changes studies have used morphological techniques such
and that the changes are remarkably similar (e.g., as electron microscopy or Golgi-like stains and have
Rose & Rankin, 2001). There is reason to believe, shown that experience-dependent changes can be
for example, that there are NMDA-like changes in seen in every species of animals tested, ranging
learning in both mammals and invertebrates (e.g., from fruit flies and bees to rats, cats and monkeys
Roberts & Glanzman, 2003). The details of the post- (for a review see Kolb & Whishaw, 1998). Consider a
synaptic second messengers may differ in simple and few examples.
complex systems but the general principles appear to When animals are placed in complex environ-
be conserved across both simple and complex animals. ments rather than simple laboratory cages, within
30 days there is about a 5% increase in brain weight
and cortical thickness, an increase in cortical acet-
The brain is altered by a wide range
ylcholine and neurotrophic factors (e.g., nerve
of experiences
growth factor (NGF), brain-derived neurotrophic
Virtually every experience has the potential to alter factor (BDNF), fibroblast growth factor-2 (FGF-2)),
the brain, at least briefly. It now has been shown as well as changes in physiological properties of
that a wide variety of experiences can also produce neurons such as those measured in studies of

© in this web service Cambridge University Press www.cambridge.org

Cambridge University Press
978-0-521-87133-4 — Cognitive Neurorehabilitation
2nd Edition
Excerpt
More Information

10 Section 1. Principles of cognitive neurorehabilitation

long-term potentiation (LTP) (for a review see Kolb altered neuronal networks has yet to be proven but
& Whishaw, 1998). Although most studies have there is little doubt that the chronic effects of drug
focused on neocortical changes, similar changes use are not neutral to cerebral functioning. The
can also be seen in hippocampus and striatum ability of drugs to alter neuronal morphology may
(e.g., Comery et al., 1996; Juraska, 1990). The ana- be important for rehabilitation because drugs can
tomical and physiological changes are associated be combined with behavioral treatments such as
with improved performance on tests of both motor rehabilitation therapy, including cognitive therapy
and cognitive behaviors and although the data are (e.g., Gonzalez et al., 2006).
correlational, it is generally assumed that the mor- Taken together the examples described above
phological changes are responsible for the facilita- illustrate the power of experience in modulating
tion in behavior. cerebral networks and in facilitating remodeling
Experience-dependent changes in the brain do stimulated by behavioral therapies. Although expe-
not require procedures as intense as complex hous- rience is likely more effective in remodeling neural
ing, however. Increased social experience selec- networks as they are repairing after injury, improve-
tively increases synapses in the orbital frontal ment still can occur late after injury (e.g., Hodics
cortex (Fiorino & Kolb, 2003; Hamilton et al., 2003). et al., 2006). Psychomotor stimulants may provide a
We have also seen that tactile stimulation either in powerful way of reinstigating cerebral plasticity late
infancy or adulthood alters cells in sensorimotor after injury to facilitate the effectiveness of behav-
cortex (e.g., Gibb & Kolb, submitted a,b). This latter ioral therapies.
treatment has also been used in animals with cort-
ical injuries to stimulate dendritic growth and facil-
Plastic changes are age-specific
itate functional recovery. Although there is little
evidence that exercise can enhance plasticity in When weanling, adult or senescent rats were placed
the normal brain, there is growing evidence that it in a complex environment, we had anticipated that
can facilitate plastic changes in the injured lab ani- we would find larger changes in the younger ani-
mal and human brain (e.g., Gibb et al., 2005; Kramer mals but to our surprise, we found a qualitative
et al., 2006). difference in the neuronal response to the same
A final example can be seen in the effects of psy- experience. Thus, whereas rats at all ages showed
choactive drugs. Robinson & Kolb (1999a) showed an increase in dendritic length and branching in
that repeated doses of amphetamine or cocaine neocortical pyramidal cells after complex housing,
given to rats produced a persisting increase in den- rats placed in the environments as infants showed a
dritic length and spine density localized to the decrease in spine density whereas young adult or
medial prefrontal cortex (mPFC) and NAcc but not senescent rats showed an increase in spine density
to adjacent sensorimotor cortical regions. It now (Kolb et al., 2003a). A similar drop in spine density
appears that repeated doses of all psychoactive was found in later studies in which newborn rats
drugs, including prescription drugs, change neuro- were given tactile stimulation with a soft brush for
nal morphology. The details of drug-induced mor- 15 min, three times daily over the first 10 days of life
phological changes vary with the drug but the (Kolb & Gibb, submitted).
general principle is that psychoactive drugs alter The obvious question is whether the behavioral
neuronal morphology in the cerebrum and this effects to the complex housing are the same
can be seen both in dendritic measures as well as depending upon the age at experience. Our early
in a variety of more molecular measures (for a results suggest that there is an advantage in both
review, see Hyman et al., 2006). Once again, the cognitive and motor tasks and that it does not mat-
relationship between the behavioral changes, such ter when the experience occurred. There are clearly
as drug-induced behavioral sensitization, and the different ways to organize neuronal networks to

© in this web service Cambridge University Press www.cambridge.org

Cambridge University Press
978-0-521-87133-4 — Cognitive Neurorehabilitation
2nd Edition
Excerpt
More Information

Chapter 1. Principles of neuroplasticity and behavior 11

enhance both motor and cognitive behaviors. This clear that many experience-dependent changes are
point is important as we consider treatments for highly specific. The clearest examples can be seen in
brain dysfunction – there may be many ways to neuropsychological studies in which animals are
facilitate recovery. trained on cognitive or motor tasks. For example,
rats trained on a visuospatial task show specific
changes in visual cortex whereas rats trained on
Early events, including prenatal events, can
motor tasks show specific changes in motor cortex
influence the brain throughout life
(e.g., Greenough & Chang, 1989; Kolb & Cioe, sub-
Our finding that early postnatal experiences could mitted; Withers & Greenough, 1989). Such task-
alter neuronal organization led us to ask if prenatal dependent specific changes are reasonable in view
experiences might also alter cerebral organization. of the relative localization of functions in the cortex.
In one study pregnant dams were placed in complex But not all area-dependent changes are so easily
environments for 8 hours a day prior to their preg- predicted. Consider two examples.
nancy and then throughout the 3 week gestation. (In We noted above that the effect of psychoactive
different studies the dams were in the environments drugs appeared to be selective to regions that
during the day or night but it made no difference.) receive dopaminergic innervation. We therefore
Analysis of the adult brains of their infants showed a were surprised to find that the orbitofrontal cortex
decrease in dendritic length and an increase in spine (OFC), another region that receives dopaminergic
density in adulthood (Gibb et al., submitted). We innervation, showed drug-induced changes that are
were surprised both that there was a large effect of opposite to those in mPFC and NAcc (Robinson &
prenatal experience and that it was qualitatively Kolb, 2004). Thus, whereas psychomotor stimulants
different than experience either in the juvenile increased dendritic length and spine density in the
period or in adulthood. More recently we have mPFC, they decreased the same measures in
shown that a variety of prenatal experiences alter the OFC. The contrasting effects of these drugs on
brain organisation in adulthood including prenatal the two prefrontal regions are puzzling given the
tactile stimulation (i.e., stimulation of the pregnant similarity in thalamic and other connections of the
dam), exercise during pregnancy, prenatal stress two regions (e.g., Uylings et al., 2003). Curiously,
and psychoactive drugs. All of these experiences there also are differential effects of gonadal hor-
also chronically alter motor and cognitive functions, mones on the two prefrontal regions as well: mPFC
with the precise effect varying with the different neurons have more synaptic space in males whereas
experiences (for a review see Kolb et al., in press). OFC neurons have more space in females (Kolb &
Although we do not know how these prenatal Stewart, 1991). Although we do not yet know what
changes might influence the effect of postnatal such differences mean behaviorally, there can be
experiences, it is clear that prenatal experiences little doubt that the differential response of two
produce chronic effects on brain organization and such similar cortical regions to drugs and hormones
behavior. One is reminded here of the idea of cog- must be important in understanding their functions.
nitive (or neural) reserve as being key factors in the
onset of dementias (e.g., Stern, 2006). Might early
Plastic changes are time-dependent
life events influence cognitive reserve in adulthood
or senescence? There is growing evidence that plastic changes are
not constant and can change over time. The clearest
example comes from drug studies. For example,
Plastic changes are area dependent
although there are large increases in spine density
Although we are tempted to expect plastic changes in and dendritic length 2 weeks after cessation of cocaine
neuronal networks to be fairly general, it is becoming administration, these changes slowly disappear over a

© in this web service Cambridge University Press www.cambridge.org