THEORETICAL APPROACHES

Medically unexplained model assumes that dissociation is a defensive response evolved

to protect the individual from the overwhelming affect associated

symptoms: a new model with trauma.3 This traumatic affect is then ‘converted’ into somatic
symptoms, allowing the affect to be expressed without its traumatic
origins being acknowledged. The resulting reduction in negative
Richard J Brown affect is the ‘primary gain’ from symptoms. Any extrinsic benefits
arising from the symptoms (e.g. attention from others, financial
gain) are ‘secondary gains’. The symptoms themselves are thought
to be symbolic representations of the unconscious psychological
conflict that is being converted.
One advantage of the conversion model is that clinicians often
encounter patients with MUS where the symptoms seem to serve
Doctors frequently encounter patients with somatic symptoms some kind of psychological function, have been precipitated by
that defy adequate medical explanation. Many such ‘medically conflict or stress, or confer obvious gains. In practice, however,
unexplained symptoms’ (MUS) develop into chronic problems it is extremely difficult to establish an aetiological link between
associated with considerable distress, disability and resource uti- stress, defence, gain and symptoms. Moreover, these are often
lization. Despite the importance of this problem, the development absent in many cases and are no more common in MUS than in
of more effective treatments for MUS is hindered by inadequate general medical illness. The same is true for other phenomena
understanding of the mechanisms of these conditions. often viewed as evidence for the conversion model, such as la
belle indifférence (the lack or paucity of concern about disability
and the prospect for recovery).4
Previous models
Historically, there have been three basic models of MUS. All of Somatization
these models have been refined since their initial development, The somatization concept also assumes that MUS reflect psycho-
although the basic elements remain in each case. logical distress that is being experienced or expressed as somatic
symptoms.5,6 In this case, however, it is assumed that MUS have a
Dissociation multifactorial aetiology, with biological, psychological and social
The dissociation model assumes that traumatic experiences create variables all regarded as important (Figure 1). Current cognitive–
a deficit in attention that prevents memories from being integrated behavioural therapy (CBT) models of MUS overlap closely with the
with the patient’s personality.1 As the patient has no control over somatization concept and identify other relevant factors.
these ‘dissociated’ memories, they may be activated by internal The main strength of the somatization model is its emphasis
and external events producing a spontaneous experience of the on the biopsychosocial nature of MUS and the empirical support
memory content. Due to the patient’s attentional deficit, however, linking these conditions with various biological, psychological and
the content is experienced as current reality rather than a memory. social factors. However, at present, the precise aetiological role of
Broadly speaking, unexplained symptoms arise when the dissoci- these factors is unclear. Like the conversion model, the somatiza-
ated memories are of somatic sensations. tion concept is also limited by its assumption that symptoms are
Consistent with the dissociation model, research has found that necessarily a product of psychological distress, which is not well
MUS patients often report early traumatic experiences. There is also established empirically. Moreover, it can be difficult in practice to
some evidence of attentional deficits and of dissociations between identify a clear link between a patient’s distress and their somatic
different forms of cognitive processing in patients with MUS. symptoms.
Research in this area remains limited, particularly as there are doubts More generally, none of the available models provides a precise
about how dissociation should be defined and measured.2 explanation of how it is possible for physical symptoms to exist in the
absence of obvious physiological changes. Although physiological
Conversion processes (e.g. autonomic arousal, muscle tension, deconditioning)
The dissociation model assumes that memory fragmentation is a can account for many MUS, this is unlikely to be true in all cases,
pathological consequence of trauma in individuals who are con- particularly for certain unexplained neurological symptoms (e.g.
stitutionally vulnerable to dissociation. In contrast, the conversion sensory loss, pseudohallucinations, non-epileptic seizures).

Medically unexplained symptoms: a new model

In the remainder of this contribution, a novel model of MUS is
presented. Rather than replacing existing approaches in this area,
the model incorporates the dissociation, conversion and somatiza-
Richard J Brown is Lecturer in Clinical Psychology at the University of tion models within a general framework based on research and
Manchester, UK, and Clinical Psychologist with Manchester Mental theory from academic psychology. In so doing, the model explains
Health and Social Care Trust. He completed his PhD at University College how it is possible for physical symptoms to exist in the absence
London, and trained in clinical psychology in Manchester. His research of physiological perturbation, or where physiological factors are
interests include somatization, dissociation and the cognitive psychology insufficient to account for the clinical presentation. In addition,
of consciousness. the model demonstrates that unexplained symptoms can arise

PSYCHIATRY 5:2 43 © 2006 Elsevier Ltd

 THEORETICAL APPROACHES

Factors involved in creation and maintenance of medically unexplained symptoms according to the
somatization and cognitive–behavioural models

Cognitive Emotional Behavioural Temperamental Physiological Social

Illness beliefs Fear Medical consulting Negative affectivity Current/recent Reaction of significant others
illness or injury
Attention to body Anxiety Avoidance Absorption Emotional arousal Iatrogenesis
Illness attribution Depression Reassurance- Hypnotic Previous illness Insurance system
seeking susceptibility experience
Illness worry Anger Body checking Perfectionism Normal physical Compensation system
perturbations
Stress Information- Personality disorder Hormonal changes Illness ‘models’
seeking
Interpersonal Reduced activity Genetic factors Reaction of colleagues
conflict levels
Trauma Defensive postures

from subtle alterations in normal psychological processes rather controlled by a hierarchical set of memory programmes (‘schemas’),
than mental disorder per se. which are triggered automatically by the creation of primary repre-
The basic premise of the model is that many MUS can be sentations. Schemas represent the processing operations involved in
regarded as alterations in bodily consciousness and control. In different acts. At the top of the hierarchy are high-level programmes,
order to explain these alterations, it is first necessary to model such as driving a car. At the next level are programmes for different
how normal consciousness and control operate in the brain. components of the high-level acts (e.g. changing gear). At lower
levels are simple programmes for the most basic parts of an act (e.g.
Consciousness moving the left foot downwards 8 cm). Once high-level schemas
Cognitive psychology has shown that the contents of conscious- are triggered, each of the component programmes is also primed
ness are determined partly by incoming stimuli and partly by for automatic activation by environmental events. In this way, very
information in memory. In other words, the contents of con- complex behaviours can be triggered quickly and efficiently, simply
sciousness are an interpretation of the stimulus world, based by perceiving the environment. Action controlled in this way is
on existing knowledge in the system; this interpretation is one perceived as effortless and intuitive.
possible account of the world out of a number of competing As existing memory programmes are insufficient to deal with
alternatives. According to the current model, the receipt of sensory novel situations, another level of action control is also required. In
information activates a number of different ‘hypotheses’ about this account, novel situations are managed by a secondary atten-
that information in memory. The most ‘active’ hypothesis is then tional system (SAS), which controls action by biasing the activa-
selected by a primary attentional system (PAS) and combined tion level of schemas. Action controlled in this way is perceived
with sensory data to produce an account of the environment or as effortful, conscious and deliberate. Experientially, we typically
‘primary representation’. Primary representations are the basic ‘locate’ ourselves at the level of the SAS (rather than the PAS) as
contents of consciousness. The choice of hypothesis depends on this is the province of self-consciousness.
the activation levels of hypotheses in memory. Activation levels
vary according to the nature of incoming stimuli, how well learnt Unexplained symptoms as alterations in consciousness and control
each hypothesis is, current goals and the activation of related There are many instances (e.g. hallucinations, hypnotic
information in memory. phenomena, certain illusions) where experience is more consistent
As this process is performed pre-consciously, the individual is with what we believe about the world than information from the
unaware that their conscious experience is a theory about the world senses. Such examples demonstrate that material in memory can
rather than a direct record of it. The individual therefore takes their often override sense data as the contents of consciousness are
experience at face value, assuming that it is accurate. created. According to the current model, this process is central
to the generation of some MUS. By this view, these phenomena
Cognitive control result from the selection of an inappropriate hypothesis (or ‘rogue
Once a basic account of the world has been generated it can be representation’) during the creation of bodily consciousness and
used to control action. The model assumes that routine actions are the control of action, reflecting over-activation of the representation

PSYCHIATRY 5:2 44 © 2006 Elsevier Ltd

 THEORETICAL APPROACHES

Creation of conscious experience and Misperception due to over-activation

control of action: veridical perception of hypothesis in memory

RUN!! Illness behaviour/worry etc.

Behavioural programmes triggered Behavioural programmes triggered

Conscious percept ‘snake!’ Conscious experience ‘Pain!’

Most active hypothesis combined with sensory Pain hypothesis inappropriately

data to provide primary representation selected due to over-activation

Activation of hypotheses about sensory Previous Previous

experience Activation of hypotheses in memory experience
information in memory

‘Rope’ ‘Basket’ ‘Ant hill’ ‘Snake’ Current ‘Itch’ ‘Tingle’ ‘Nothing’ ‘Pain’ Current
context context

Sensory information Sensory information: normal bodily sensations

in memory (Figure 2). The result is a compelling distortion in the the individuals themselves. In other cases, the representations will
body image experience as a somatic symptom. As individuals are have been acquired through exposure to physical states in others
aware only of the products of this process and not the process itself, or via sociocultural transmission (e.g. the Internet).
they interpret their symptom experience as an accurate account
of their bodily state and behave accordingly. Symptom maintenance
The nature of the resulting symptom depends on the rogue If unexplained symptoms reflect the over-activation of information
representation in question. Symptoms characterized by alterations in memory, what causes this over-activation? The primary factor
in experience (e.g. pain, nausea, pseudohallucinations) will be in this respect is the repeated re-activation of this information by
associated with perceptual representations corresponding to the the secondary attentional system. Anything that increases symp-
symptom in question. Symptoms characterized by an inability to tom-focused attention therefore contributes to the development
control perception, action, or cognition (e.g. blindness, paralysis, and maintenance of MUS (Figure 3). This category encompasses
amnesia) will be associated with behavioural representations (i.e. a wide range of biopsychosocial factors, overlapping closely with
schemas) specifying the processing parameters of the relevant the somatization model (Figure 4).
state (e.g. inhibition of visual information, movement or memory
retrieval). These schemas are activated automatically by the Misattribution of symptoms: research shows that attention is
creation of primary representations. One important implication of diverted more to symptoms that are attributed to serious rather
this is that unexplained symptoms are produced by psychological than benign causes. Many variables can influence whether an
mechanisms but are not produced deliberately, as they involve the unexplained symptom is misinterpreted as evidence of serious
primary rather than the secondary attentional system. illness. Thus, unusual, painful or disabling symptoms are more
likely to be interpreted as serious than familiar or less distressing
Origins of rogue representations symptoms. Similarly, the related beliefs that symptoms necessarily
In many cases, rogue representations will be memories of previous indicate illness and that health is a state devoid of symptoms both
illness episodes and physical states (e.g. of emotional arousal) in predispose towards misattribution, as does medical information

PSYCHIATRY 5:2 45 © 2006 Elsevier Ltd

 THEORETICAL APPROACHES

Factors involved in the development of Overlap between current model and previous
symptom chronicity accounts of MUS
Physiological Traumatic Emotional Overlap with dissociation model
disturbance experience arousal
• Attentional dysfunction
• Separation between different levels of mental processing
• Activation of illness memories
Symptom
representations Overlap with conversion model
• Defensive avoidance of traumatic affect
• Possibility that symptoms can recapitulate aspects of
traumatic experiences (see Brown, 2004)
Attention
to symptoms
Overlap with somatization model
• Biopsychosocial model
• Symptom misinterpretation
Selection of • Emphasis on cognitive–behavioural therapy as treatment
rogue representation modality

Attribution of Distress
experience to illness and disability

a catastrophic interpretation of their symptoms. Avoiding activity

can also cause secondary physical problems, as can defensive
Illness Illness worry postures that the individual may adopt in the mistaken belief that
behaviour and rumination
they minimize damage and disability.

Adapted from Brown R J, Psychol Bull 2004; 5: 793–812. Illness worry and rumination: patients with MUS often spend a
lot of time worrying and ruminating about their symptoms and the
fact that doctors have been unable to explain and treat them. Worry
3
and rumination both involve symptom-focused processing by the
secondary attentional system and therefore play an important role
that is inaccurate, inappropriate or ambiguous. Misdiagnosis in the continuing activation of rogue representations. Over time,
and/or medical mismanagement are likely to play a particularly the individual may get ‘locked into’ this form of processing activity
important role in symptom misattribution. making it increasingly difficult to ignore symptoms.7

Negative emotional states: symptom misinterpretation is often a Personality: individuals high in trait anxiety (or negative affectiv-
significant source of anxiety. This not only contributes to further ity) are more likely to attend to physical symptoms, to misattribute
misattribution but also increases the degree to which individuals them to physical illness and to worry about them excessively.
focus on their body for further signs of illness.7 This is also true Negative affectivity therefore contributes to the activation of rogue
for other negative emotional states such as depression. In addition, representations on several levels.
such states are associated with bodily changes (e.g. autonomic
arousal) that are a source of further symptoms. Defensive use of body focus: focusing attention on the body can
also be an effective way of avoiding the cognitive and emotional
Illness behaviour: once symptoms have been attributed to a activity associated with traumatic events. This could account for
physical cause, the individual may engage in a range of behaviours the relationship between early abuse and MUS.4 Once developed,
directed at bringing about symptom relief or reducing negative somatic symptoms also provide a way of expressing negative affect
affect. Many such illness behaviours have the effect of increasing without acknowledging its psychosocial source.
attention to the body, particularly when performed excessively.
This includes physical and mental checking of the body to confirm Treatment implications
symptom status, seeking information about symptoms (e.g. on the The model provides a detailed rationale for the use of CBT with
Internet), reassurance-seeking and repeated medical consulting. MUS,8 aids in the formulation of these conditions and identifies
Other illness behaviours, such as doctor shopping, avoidance of an important mechanism of change in many cases. It may be
medical examination/information and reducing activity levels, particularly useful in cases where it is difficult to identify obvious
indirectly affect symptom-focused attention by preventing the physiological perturbations that could account for symptoms,
individual from being exposed to information that could disconfirm providing both clinicians and clients with a non-blaming way

PSYCHIATRY 5:2 46 © 2006 Elsevier Ltd

 THEORETICAL APPROACHES

of understanding how such symptoms could have been created FURTHER READING
and maintained. Symptom-focused attention and the distortion of Brown R J. Psychological mechanisms of medically unexplained
sensory input are also likely to be important in the maintenance symptoms: an integrative conceptual model. Psychol Bull 2004; 130:
of MUS that have obvious physiological causes (e.g. autonomic 793–812.
arousal). As such, the model may inform formulation and (Explains the model in detail.)
treatment in these cases as well. Socialization to the model can Brown R J. Medically unexplained symptoms. In Tarrier N, ed. Case
be augmented with everyday examples of how bodily experience formulation in cognitive behaviour therapy: the treatment of
often misrepresents reality, including the placebo effect, battlefield challenging and complex cases. London: Brunner-Routledge, in press.
analgesia and hypnotic phenomena. Emphasis should also be (Provides information about the cognitive–behavioural treatment of
placed on the subjective reality of symptoms, the role of normal MUS, incorporating insights from the current model.)
biopsychosocial processes and the non-threatening nature of MUS;
often unpalatable issues such as gain, denial and unconscious
conflict can be de-emphasized.
Treatment should focus on reducing the activation of rogue
representations by minimizing symptom-focused attention. This
will involve identifying and addressing factors that are maintain-
ing symptom focus, particularly catastrophic misinterpretation,
body checking, illness worry/rumination, help/reassurance-seek- Practice points
ing and avoidance of feared situations and activities. Attention
training treatment may be a particularly useful way of reducing • Not all medically unexplained symptoms reflect the somatic
symptom-focus.9 Techniques such as hypnotic and imaginal sug- presentation of emotional distress
gestion also provide direct ways of reducing the activation of rogue
representations. These may be augmented by physiotherapy and • Medically unexplained symptoms can arise from subtle
graded activity programmes that serve to activate more healthy disruptions in normal psychological mechanisms that produce
representations of the body.  distortions in bodily consciousness and control

• Symptom-focused attention is a key factor in the creation and

maintenance of unexplained complaints

• Other examples of how bodily experience often misrepresents

reality (e.g. placebo effect, battlefield analgesia, hypnotic
phenomena, dreams) can help patients understand how
symptoms can exist in the absence of medical pathology

• Cognitive–behavioural therapy for medically unexplained

REFERENCES symptoms should aim to minimize symptom-focused attention
1 Janet P. The major symptoms of hysteria. New York: Macmillan, 1907. and symptom misinterpretation, while augmenting healthy
2 Holmes E, Brown R J, Mansell W et al. Are there two qualitatively perceptions of the body
distinct forms of dissociation? A review and some clinical
implications. Clin Psychol Rev 2005; 25: 1–23.
3 Breuer J, Freud S. Studies on hysteria. In Strachey J, Strachey A, eds.
The Penguin Freud library. Volume 3. London: Penguin, 2001.
4 Brown R J. Dissociation and conversion in psychogenic illness. In
Hallett M, Fahn S, Jankovic J, Lang A E, Cloninger C R, Yudofsky S C,
eds. Psychogenic movement disorders: psychobiology and treatment
of a functional disorder. Philadelphia: Lippincott, Williams & Wilkins,
2005.
5 Lipowski Z J. Somatization: the concept and its clinical application.
Am J Psychiatry 1988; 145: 1358–68.
6 Kirmayer L J, Taillefer S. Somatoform disorders. In Turner S M, Hersen
M, eds. Adult psychopathology and diagnosis. 3rd ed. New York:
Wiley, 1997.
7 Wells A. Cognitive therapy of anxiety disorders. Chichester: Wiley,
1997.
8 Sharpe M, Peveler R, Mayou R. The psychological treatment of
patients with functional somatic symptoms: a practical guide. J
Psychosom Res 1992; 36: 515–29.
9 Wells, A. Emotional disorders and metacognition. Chichester: Wiley,
2000.

PSYCHIATRY 5:2 47 © 2006 Elsevier Ltd