Cognitive Therapy and Serious Mental Illness. An Interacting Cognitive Subsystems approach.

Isabel Clarke.

Psychology Department. Royal South Hants Hospital. Southampton. U.K.



The increasing application of Cognitive Therapy to the more enduring forms of psychopathology represented by the DSMIV ‘Axis II Disorders’ has led to the piecemeal development of the discipline, and the incorporation of approaches from other therapeutic modalities, and from wider sources, such as Buddhist meditation. The present paper proposes the development of the Cognitive rationale, using as a foundation the research based insights provided by Teasedale’s ‘Interacting Cognitive Subsystem’ model (Teasdale and Barnard 1995). By emphasising the close relationship between the emotional (implicational) subsystem and states of bodily arousal, this restores aversive arousal states to a central place in the understanding of psychopathology, and clinical practice. The role of threatening information about the self received through early relationships in leading to chronic aversive arousal states, whether high arousal as in anxiety, or low, as in depression, in Axis II disorders, is considered. The implications of the tension between this aversive information and the basic human endeavour of constructing the self are discussed, and a clincal example is used to illustrate the therapeutic approaches suggested by this perspective.



Cognitive therapy has recently been advancing rapidly on a number of fronts; both in terms of the fundamental research understanding of cognition, memory and arousal, and clinically into areas of ever greater and more complex pathology. As well as responding to the available challenge, practitioners are answering the call from on high to concentrate efforts on those with serious mental illness. I am here using this term to denote disorder affecting the organisation of the personality, as in DSMIV ‘Axis II disorders’, but not including psychosis. Methodology is developed and borrowed in response to this more demanding client group. In the resulting diversity, the coherent and clearly communicable rationale that is central to the enterprise of engaging therapy clients collaboratively with the cognitive model can get left behind. I am here proposing to draw together some of these strands, both theoretical and therapeutic, in a simple rationale based on Teasdale’ s Interacting Cognitive Subsystem model (Teasdale & Barnard 1993 ) to address these concerns.


This paper can be regarded as a contribution to a debate as opposed to a wholly original formulation. As well as drawing on Teasdale, I refer to a number of the contributors to Power and Brewin (1997)’s recent edited volume, "The Transformation of Meaning", and relate my approach to Dialectical Behaviour Therapy (Linehan 1993) and Cognitive Analytic Therapy programmes for the treatment of Borderline Personality Disorder (Ryle 1997). The constructivist cognitive tradition is an organising influence: as Anderson writes "Constructivist therapy is not so much a technique as a philosophical context." (Anderson (1990), quoted in Neimeyer (1993). This context informs my view of the construction of the self. In particular, I cite Greenberg and Pascual-Leone’s chapter from the Power and Brewin book. Constructivists such Guidano and Liotti (1983) have been at the forefront of recognizing the influence of early attachment relationships on fundamental assumptions about the self and the world. Recognition of the profound impact of early relationships has also been an essential feature of Beck’s theorizing for many years (Beck & Emery 1985). This is one example of the trend towards the widening of the boundaries of cognitive therapy referred to at the beginning, as ideas once characteristic of other therapeutic modalities are incorporated.


Employing the Interacting Cognitive Subsystem Model

I am not here attempting a full exposition of the Interacting Cognitive Subsystems model, which can be found in Teasdale and Barnard (1993), and Teasdale’s chapter in Power and Brewin (1997), among other places. I am proposing to concentrate on the following features of the model. Firstly, as an information processing model, it is based on experimental evidence for different forms of coding information; for instance, immediate and sensory based, verbal and logically based, or a more holistic, meaning based coding. These and other distinct codes form the basis for nine postulated subsystems; three are sensory and proprioceptive, two involve higher order pattern recognition; two, the production of response, and two are yet higher order, meaning based systems on which I will now focus, the propositional and the implicational. Memory is integral to each subsystem, and likewise distinguished by separate codes. Thus, the logical, propositional, memory is verbally coded, whereas the implicational memory, that records meaning at a more generic level, is encoded in a rich variety of sensory modalities, and is therefore more immediate and vivid.

Another area illuminated by the research into short term memory and human - computer interaction on which the theory is based (e.g. Barnard 1985) is the need for a transformation process in the interchange of information between one subsystem/coding and another, and that this is constrained by the limitation of the processing capacity. Thus connections are made more immediately within a particular memory store than between the data stored in different memory stores and coded differently. An example of this that will be important for the argument that follows is the rapid connection made between events of personal significance stored in the implicational memory. For instance, memory of earlier socially shaming events might flash into the mind of a socially anxious individual entering an unfamiliar social situation. These memories will be more immediately accessible than logical information about, for instance, interesting possible topics of conversation, stored in the propositional memory.


The other feature of the system that is central to the current argument is the immediate connection between the implicational and body state subsystems, and the much more indirect route by which information about arousal reaches the propositional subsystem. This is intuitively understandable because of the relationship between emotion and arousal. Recent neuroscientific advances clarify this direct connection between sensory appraisal of salient information and immediate autonomic arousal. (See for instance LeDoux (1993), cited in Greenberg and Pascual-Leone (1997)


Arising out of this connection to body state information, the implicational subsystem has a monopoly on emotional meaning, and a feedback loop can be set up between its appraisal of threat, and arousal, whereas the propositional subsystem can remain relatively detached from this influence. The social phobic’s recall of past social disasters will evoke an immediate physical arousal response that does nothing to help the situation. On the contrary, the experience of arousal with its associations with fear simply confirms the individual’s hunch that there is something to be frightened of. Thus a vicious circle is set up, relatively uninfluenced by a cool, "propositional" appraisal of the situation.

The final feature of the model relevent to this discussion is the central place accorded to the interchange between the propositional and implicational subsystems, which Teasdale calls " the central engine of cognition", there being no central executive beyond this interchange ( Teasdale and Barnard P.78). The limitations of processing capacity already noted makes possible the establishment of habitual patterns of response in these exchanges, which he identifies with "schematic models". As these are resistent to revision and the incorporation of new information, they can maintain maladaptive responses, such as could be amenable to modification through cognitive behaviour therapy. For this modification to take place, the individual needs to be able to stand aside from the habitual response and process new information at the propositional level. I am going to suggest that states of arousal, which influence the implicational level, have a crucial role in maintaining these schematic patterns, and that attention to issues of arousal facilitates bringing the propositional subsystem to bear on the situation to create a new response.


A similar distinction between logical and emotional information processing contained in the propositional and implicational subsystems is currently appearing in a number of guises, for instance (Brewin 1989), as verbally and situationally accessible memory; and Segal (1988) who refers to automatic and conscious processing. Ellis’s distinction between inference and evaluation can be seen in the same light; according to his theory it is evaluations, or hot cognitions, rather than inferences that are associated with emotional problems (Ellis, 1962). Teasdale employs Ellis’s terminology of "hot cognition" to denote implicational level processing, and "cold cognition" for propositional level processing.



The Implicational Subsystem and Personal Meaning.

Cognitive therapy has always been centrally concerned with meaning and the self. Both Beck and Ellis trace dysfunctional thinking patterns in the moment back to beliefs about the self. The idea that the threat lies not in the objective situation, but in the meaning attributed to the situation is also fundamental, so that the focus of therapy lies in the "transformation" of this meaning, to borrow a phrase from the title of Power and Brewin’s (1997) book, already referred to. Teasdale’s paper in this book emphasises the point that the implicational subsystem is concerned with matters that relate directly to the self and therefore to personal meaning. This is a point I wish to develop further, in order to suggest that rather than simply uncovering and challenging beliefs about the self, therapy can become involved in the ongoing process of the construction of the self. I argue that this is a process which continues throughout life, though based on the formative stage of the construction of the self which takes place within the context of the primary relationship(s) in early life; that it only makes sense in terms of relationship, whether intimate, or on a wider social stage, and that it is perhaps the central preoccupation of the human being. The predominant focus of the internal dialogue on matters relating to the safety, and status of the self, and the linked issue of significant relationships, demonstrates this.


The final thread of my argument concerns the role of arousal in this process of self construction and in psychopathology. Autonomic arousal is a biological mechanism to protect the organism from physical harm by preparing it for action, whether in the form of fight or flight in response to threat. For human beings, as for the higher animals, information on place in the social order and therefore relationship is perceived as threat, or proof of value. (I am here employing Gilbert (1992)’s evolutionary approach to human social order and arousal.) This information is registered at the implicational level and by the linked arousal system.


I am here adopting a model developed by the constructivist cognitive therapists, referred to above, and will specifically be citing the argument of Greenberg and Pasual-Leone (1997). According to this, the self is a construction forged out of cumulative information on threat and value in relation to the individual. Relating this to the ICS perspective, this information would be laid down along with rich sensory data in the implicational memory store, but reflected upon and integrated at the propositional level. Teasdale, in the same volume (P. 146), makes the distinction between the propositional understanding of "self as object", and the implicational level experiencing of "self as subject". It is central to understanding the type of emotional difficulties tackled through cognitive therapy that these two, and the related information about them in the two separate memory stores, can be quite distinct in certain circumstances. Hence, depressogenic schemata containing information about the unacceptability of the self can lurk in the implicational memory store, untapped in normal life, until triggered by some circumstance in the present that awakes echoes of that particular memory content (see, for example, Segal 1988).


The Implicational Memory and Autonomic Arousal.

Regulation of arousal is central to an ICS formulation, since cognitive restructuring represents the propositional level appraisal of implicational level material, and states of high autonomic arousal pose an obstacle to this. Physiologically the state of "hypocapnia" or decreased alveolar CO2, produced by the hyperventilation characteristic of autonomic arousal reduces blood flow to the brain (Fried 1993). Subjectively this produces the experience of "tunnel vision" where concentration on threat-related information, drawn more from the implicational memory than from current sensory data, excludes all other considerations. When arousal levels rise towards panic, thinking becomes paralysed into confusion.


The shift from behaviour therapy to cognitive therapy over the last 20 or so years has led to a reduction in emphasis on regulation of arousal (for instance by progressive relaxation techniques (Jacobson 1964). Recent trends to re-instate this aspect are to be welcomed, such as Linehan’s "Distress Tolerance" and Emotion Regulation" within Dialectical Behaviour Therapy skills training (Linehan 1993). Attention regulation breaks into the cycle of arousal and concentration on threat at a different point. Recent examples are Wells et al. (1997)’s investigation of techniques of attention and concentration training, and Linehan’s adoption of Buddhist mindfulness techniques to regulate attention so that it sits fairly between implicational and propositional systems. Teasdale, Williams and Segal have adopted the same approach, as expounded by Kabat-Zinn (1996), in a study in progress, into relapse prevention in depression (reported at the 1998 BABPC conference).


A Developmental Perspective.

I will now develop these ideas by linking an ICS based understanding of the construction of the self, to issues of arousal regulation to illuminate sources of psychopathology. Taking as my starting point the sort of social evolutionary perspective expounded by Gilbert (1992) I suggest that an individual’s sense of self is constructed out of their experience of being in relationship from birth (and very probably before that in the womb) and throughout their subsequent experience. There is abundant evidence in the intersubjectivity literature to suggest the fundamental role of the infant - caregiver dyad in the creation of meaning, communication and therefore a sense of self in the infant. Both the attachment theory school (e.g. Ainsworth et al. 1979) and researchers coming from a cognitive developmental position (e.g Pawlby 1977, and Lewis et al. 1980) have explored and established this through extensive experimentation. Greenberg and Pascual-Leone summarise this process in Power and Brewin (1997) thus:


"Infants’ emotional systems are involved in rapid evaluation of what is good and bad for them. Thus infants, right from birth, experience feelings and, as soon as they can construct schemes of sufficient complexity, they use these feelings to construct a conscious personal sense of self. A major determinant of this self construction is their intersubjective experience, with their caretakers, associated with their own automatic emotional reactions. An individual’s sense of self is permanently organised around emotional schemes formed in primary attachment relationships. Affect regulation develops with maturation, but also with the way caretakers react to the child’s emotions; these experiences determine the affectively based sense of self."


From this earliest stage, information about threat and (the individual’s) value, and therefore the self and its survival, is stored in the implicational memory. For the infant there is no distinction between information about interpersonal acceptablitiy and information about physical survival - because of his/her absolute dependency, they are one and the same. This type of threatening information about the self is accessed in the propositional form of an unconditional core belief, of the "I am worthless" variety during cognitive therapy. As the child develops, threats are differentiated into those involving physical integrity, and those involving the social hierarchy. Both are stored in the implicational memory, and will trigger autonomic arousal when reactivated, but the former produce the most powerful reaction.


Propositional thinking develops with symbolic aptitude and language, and with this, the sense of "self as subject", based on appraisal of the primary, implicational level data about the self in relation to others. The "stage based" developmental theories such as Piaget and Kohlberg’s can be seen as an exposition of this developing ability to adopt a wider, less egocentric perspective as the child grows. With the transition from baby to toddler, the picture becomes more complex, as the young child recognises that different aspects of the self are more or less acceptable to other people. The distinction that here develops between the "public self", designed to fulfill the expectations of important others, and the privately acknowledged self is crucial for the practice of cognitive therapy. It is the basis of the "if" core beliefs; e.g. "I will only be loved if I please others", which reveal a message of conditional acceptance from the original caregiver, and fear of the revelation of the private self underlies shame.


The ICS model is particularly helpful for understanding disruptions and discontinuities in the development of the sense of self that will lead to vulnerability to breakdown in case of later adverse life events, and, in cases of serious mental illness and personality disorder, to a fragmented and poorly functioning self. The two distinct levels of processing and memory stores represented by the implicational and propositional levels allow memories of childhood events that are deeply threatening to the acceptibility of the self to be overlayed by later experience, until triggered by a later life event. For instance, entering an intimate relationship, or having a baby can bring a previously well functioning individual face to face with early, threatening experiences of relationship along with associated arousal state, and thus trigger a breakdown. The current, well functioning, understanding of the self is overwhelmed, and makes way for the re-experience of an earlier, more threatened, persona.


A Clinical Example.

In order to explore this process in the case of personality disorder, I am going to introduce a (composite) clinical example which will be used as an illustration

through the remainder of this paper. A thirty year old woman, whom I will call Alice, has caused concern to the services for some time because of self destructive behaviours. She sometimes presents with cutting and suicide attempts, and at other times with intermittent drug and alcohol abuse; she also experiences dissociated states. Her children are a cause for concern, both because of questions around the stability of the home life Alice is able to provide, and because of a propensity to form abusive partnerships. On the positive side, Alice impresses people with her determination and intelligence, and can also display an engaging, efficient, well functioning aspect. However, just as things appear to be progressing well, helping and healthy relationships are rejected, and self destructive and risky behaviours resurface. It is remarked that it is like dealing with two, or more, different people.


Developmentally this can be understood in Alice’s case in the following way, with reference to ICS. Alice’s early experiences of relationship were: a mother who alternated between being indulgent and intrusive (when a single parent) and cold and neglectful (when with a partner). A grandmother who periodically looked after her and was a good figure, but was only spasmodically available, as mother used to take her away, and mother’s two main partners who were both physically abusive, and one also sexually abused her. Thus the information about threat and value, relevent to her developing a sense of herself, stored in her implicational memory was contradictory, and much of it, highly threatening, and therefore, when triggered, productive of an aversive state of arousal. A wide range of situations, often entailing quite idiosynchratic memories, could re-evoke this state, and Alice’s various compensatory behaviours performed the function of shifting her away from this implicational level material. Drink and drugs blocked it out; self harm was consonant with the degraded sense of self she was experiencing, and brought her back to the present, and dissociation distanced her. Unfortunately, all these reactions disrupted the smooth process whereby communication between propositional and implicational level, Teasdale’s "central engine of cognition", could process reliable information about herself in the present, and facilitate the construction of the self. Because of the constant disruption, the hurts of the past were perpetuated, not processed, and a unitary sense of self could not be achieved.


Implications for therapy.

I will conclude this paper by using Alice’s case to illustrate the implications of this model for therapy. The therapeutic task, according to this approach, is to integrate Alice’s fragmented sense of herself and to enable her to tolerate contact with the implicational level information about herself from the past, so that she can at last process it and put it behind her, and to experience the present in a new and healthy way, so that she is not constantly sucked back into dysfunctional patterns of relating. The methods that follow are familiar, though sometimes drawn from outside the strict cognitive behavioural tradition. It is the ICS rationale that draws them into a coherant whole that I wish to present.


Therapeutic Methods.

1. Formulation.

The power of the threatening information about the self, locked into place by the arousal feedback loop, provides an explanation for Alice’s difficulties. The first task of therapy is to refine this at assessment into a clear formulation, worked out in collaboration with Alice, and shared with her explicitly, either in the form of a diagram, a letter, or both. The important relationships that gave her her formative information about herself and her place in the world are central to this, and in according this central place to relationship rather than core beliefs, the approach is here closer to CAT than to CBT in its traditional form. The beliefs could be seen as essentially derived from the experience of relating. Thus the formulation would draw out those experiences, and how they are perpetuated in the present, both in how Alice relates to herself, and how she relates to others. Information gathered in the assessment phase through the use of diaries will add specific information about the trigger situations that re-evoke the threatening memories and accompanying arousal states. Understanding that there is discernable logic behind behaviour she feels driven to irrationally, and further, that this is maintained by the past rather than the present, gives Alice the hope and feeling of being understood that can be the basis for the therapeutic alliance. Three immediate goals were identified at the assessment phase; to maintain the therapeutic alliance, which could fall victim to patterns of abandonment and rejection; to reduce self destructive behaviours and to maintain a new partnership with a man who was well meaning, good for the children, but unexciting compared to previous, more abusive partners.


2. The body dimension.

It was the aversive states of arousal that kept Alice’s dysfunctional patterns locked in place, and so tackling the body dimension was a powerful precursor to any change. Alice’s relationship with her body was itself quite alienated. As well as self harming, and abusing it with substances, she tended to binge and starve, and was deeply dissatisfied with her appearance, reflecting early confusing and negative messages she received about herself.


The immediate task was to develop skills in arousal regulation, and hence impulse control. As mentioned above, DBT is particularly strong on this aspect, having developed skills teaching programmes to promote mindfulness and emotion regulation. In Alice’s case, she was taught a breathing designed to bring down arousal rapidly by focussing on the outbreathe. This skill was reinforced by attendance at a stress management course, at which she refined relaxation skills. The breathing was efficient at decreasing arousal within the session, and so could be used to help contain threatening material, and to enable her to think, at the propositional level, about areas such as childhood abuse, that had previously been experienced and re-experienced mainly at the implicational level. To use these techniques between sessions, she had to learn to "read" her body, so that she could start to bring down arousal before it reached an aversive threshold.


She also worked on promoting a better relationship with her body in general; allowing herself good food and exercise, and challenging negative beliefs about it.


3. Multi modal approaches.

Both Brewin, and Teasdale, in their chapters in Power and Brewin (1997) indicate a new way forward for CBT suggested by the ICS perspective, where guided discovery is substituted for thought challenging. They argue that simply demolishing the old way of thinking is insufficient; it is necessary to foster a new quality of implicational level experience of the self. This is in line with current trends within CBT - for instance, Padesky (1993). Where the implicational subsystem is being activated and revised, a purely verbal approach is likely to be inadequate, because of the multi modal nature of this system and its memory coding. This is where the current emphasis on using imagery in therapy (see Hackmann (1997) for instance), both to explore trauma, and to reconstruct healthy schemata fits in. The same argument supports working more directly with emotions. This is a development of CBT advocated by constructivists such as Greenberg and Pascal-Leone, in Power and Brewin (1997), but also to be found within Ellis’s ABC assessment process. As expounded by Trower, Casey & Dryden (1988), this process explores the emotion and attached threats to the self (expressed in imagery, automatic thoughts and core beliefs), physiology and behavioural impulse. By exploring, and taking seriously the experienced emotion, it is possible to reconstruct the quality of the identity-threatening information that is behind the current pathology. From there it is possible to embark on the work of naming and integrating scattered elements of the identity.


In Alice’s case, imagery work was important in identifying and defusing the particular keys to the recall of aversive implicational level material that triggered self harm and relationship breaking. For instance, some interpersonal situations seemed to match and so could trigger a flashback of the adult abuser looming over her as a child; she would panic, feel trapped and either dissociate, self harm or attack the other. In discussing corrective imagery, she came up with swimming as a situation where she felt particularly free, as opposed to the claustrophobia of the flashback. Accordingly, she rehearsed the image of swimming from the past to the present, with purposeful but relaxed strokes.


Work with the feelings enabled her to mourn her shattered childhood, access her anger with abusers and mother, and to experience it as the part of herself that knew all along that she was special, however much she was treated with contempt. Accessing this deep implicational level experience of being someone worthy of protection and right treatment was the foundation for building a healthy sense of self.


4. Working on experience in the moment.

Taken together, these new approaches shift CBT away from verbal dialogue (though in essence this remains the bedrock of therapy) towards a more holistic experience in the present - in keeping with the holistic character of the implicational subsystem. The challenge of Alice’s therapy, for both herself and the therapist, was to stay in the present; to remain in touch with threatening memories without dissociation, rejection of therapy, or other escape, so that these could be fully appraised at the propostional level. The implicational level sees things in absolutes - the "black and white thinking" of the CBT thinking errors, whereas the propositional level can handle complexity. Her therapist likened staying with this complexity, both in the sessions, and in the week inbetween, like trying to balance on the middle of a see-saw.


This struggle for balance was most evident in the relationship between Alice and the therapist, as she reacted to closeness that her past experience interpreted as dangerous, with every possible impulse to escape or attack. Talking about this - bringing it into the propositional level, was a central part of the therapy.

This way of working, familiar in psychodynamic terms as transference, is not new within CBT; see for instance Safran and Segal (1988) and Young (1994). However, it fits particularly naturally within this ICS conceptualisation with its emphasis on working in the immediacy of the moment and on the smooth communication between implicational and propositional subsystems. Thus, the slipping away from present reality into a familiar pattern can be challenged, and relationship can be experienced in a new way. The new initiatives to employ mindfulness techniques as an adjunct to cognitive therapy, cited above, also work on restoring the smooth communication between the two central subsystems, while maintaining detachment from both, so sharpening experience in the moment. Cognitive Analytic Therapy has particularly clear methods for using this experience in therapy with reference to explicit written and diagrammatic formulation material.


5. A new experience of the self.

All these approaches are designed to lead the client towards a new experience of self in which emotion can be both felt and reflected upon. As well as recognising the distortions of the past, this requires the nurturing of strengths that are there all along, but become submerged by the mobilisation to deal with supposed threat. The 1990s has seen growing importance within cognitive theory of using therapy to help people construct new healthy selves, as well as weaken existing restrictive and condemnatory self constructions (e.g. Greenberger & Padesky, 1995). In Alice’s case, this new experience was grasped in the reframing of her anger as a wholesome part of herself, and in the "balancing" relationships with the therapist and the new partner. Staying balanced, without retreating into familiar escape routes, was a constant struggle. A simple diagram, summarising the possible ways of viewing familiar situations, with the idealised on one side, the catastrophising option on the other, and the compromise, representing propostional level appraisal in the middle, helped her to locate the everyday choices. Gradually she was able to report experiencing living in the middle place, and to comment on its unfamiliar, frightening feeling of uncertainty, as well as the exhileration of taking control. This new experience of herself, reflected on and guided at the propositional level, but laid down vividly in the implicational memory as well, was the basis for a new stage in Alice’s understanding of herself, and therefore in the construction of a more healthy and unified sense of self. The old patterns were not rapidly or easily erased, but at least the way ahead became clearer.



This has been an attempt to integrate a number of new concepts and approaches within CBT into a simple and coherant whole. I see it as a contribution to a debate about the development of the therapy into new areas of complexity, and towards adopting therapeutic styles previously associated with other modalities. This sort of development can lead to charges of dilution and distortion of the model. I would argue that this approach retains and indeed strengthens the following essential features of CBT; it is research based; it entails unpacking the relationship between cognition, emotion, behaviour and bodily arousal in a way that clarifies the route to change, and it works collaboratively and respectfully with the client, sharing the full understanding of the problem with them, so that they can continue to work with the model after the therapy is over.


In this way, major developments in CBT and related therapies - such as CAT for borderline personlity disorder (Ryle 1997) and dialectical behaviour therapy (Linehan 1993), to name but two, are brought together, by postulating underlying mechanisms, built around the human imperative to develop an acceptable sense of self, within the context of an ICS information processing model of the person, which accords a central place to arousal states and memory for threat.


The example given uses this approach to treat borderline personality disorder, but it can be employed to understand the spectrum of mental health difficulties in a way that cuts across diagnostic labels. All mental health problems are seen as responses to threats to assembling an acceptable sense of self, and to the aversive arousal that accompanies such threats. Anxiety based disorders do not effectively defend against this, except perhaps, through worry according to recent work on this subject (e.g. Bouman & Meijer 1999) Similarly, depression is a physiological and psychological expression of defeat and acceptance of low status or, in more serious cases, annihilation. Substance abuse and other addictive behaviour modify the arousal directly, and temporarily block awareness of the threats to the self. I would suggest that obsessive compulsive disorder and eating disorders meet the challenge of constructing the acceptable self more ingeniously, by "changing the rules". In OCD, the threat is focussed onto particular actions or thoughts, thereby rendering it more manageable in the short term, but, since the underlying anxiety is not addressed, creating more problems in the medium to long term. In eating disorders, interpreting acceptability in terms of body weight leads into the addictive loops underlying anorexia and bulimia that make these behaviour patterns so hard to eliminate.


This discussion of the possible underlying unity behind mental health problems, based on threats to the construction of the self, seen in ICS based information processing terms is, of course, highly speculative. I hope it is suggestive of the benefits in terms of clear formulation, simple enough to share with any client, that this model offers. I would further suggest that scientific exploration of the connection between arousal, memory and sense of self could help to establish this analysis on a firmer basis.




My thanks are due to Dr. Paul Chadwick for suggestions and references, and to the two anonymous reviewers whose comments have helped to shape the final version of the paper.




Ainsworth, M.D., Blehar, M., Waters, E. and Wall, S. (1978). Patterns of Attachment. Hillsdale, NJ: Erlbaum.

Anderson, W.T. (1990) Reality isn’t what it used to be. New York: Harper & Row.

Beck, A.T. & Emery, G., with Greenberg, R.L.(1985). Anxiety disorders and phobias: a cognitive perspective. New York: Basic Books.

Bouman, T.K. & Meijer, K.J. (1999). A Preliminary Study of Worry and Metacognitions in Hypochondriasis. Clinical Psychology and Psychotherapy, 6, 96 - 102.

Brewin C.R. (1989) Cognitive Change Processes in Psychotherapy. Psychological Review, 96, 379-394.

Ellis, A. (1962). Reason and emotion in psychotherapy. New York: Lyle Stuart.

Fried R. (1993) The role of Respiration in Stress and Stress Control: Toward a Theory of Stress as a Hypoxic Phenomenon. In Lehrer P.M. & Woolfolk R.L. Eds. (1993) Principles and Practice of Stress Management. N.Y. Guildford.

Gilbert, P.(1992) Depression. The Evolution of Powerlessness.

Hove:Lawrence Erlbaum Associates.

Greenberg, L.S. & Pascual-Leone, J. Emotion in the Creation of Personal Meaning. In Power,M. & Brewin,C.R.Eds.(1997) The Transformation of Meaning. Chichester: Wiley.

Greenberger, D & Padesky, C. A. Mind over Mood. A Therapist’s Manual. Guildford. London.

Guidano, V & Liotti, G. (1983). Cognitive processes and emotional disorders. New York: Guildford press.

Hackmann, A. The Transformation of Meaning in Cognitive Therapy. In Power,M. & Brewin,C.R.Eds.(1997) The Transformation of Meaning. Chichester: Wiley.

Jacobson, E. (1964) Self-operations control: a manual of tension control. Chicago. National Foundation for Progressive Relaxation.

Kabat-Zinn J, (1996) Mindfulness Meditation for Everyday Life. Piatkus.

LeDoux, J.E. (1993) Emotional networks in the brain. In M.Lewis & J.M.Haviland Eds., Handbook of Emotions. NY Holt.

Linehan, M. M., (1993) Skills Training Manual for Treating Borderline Personality Disorder. N.Y. Guildford.

Meichenbaum, D., (1976) A self-instructional approach to stress management. A proposal for stress innoculation training. In Spielberger C. and Sarason I. Eds. Stress and Anxiety in Modern Life. New York. Winston.

Neimayer, R.A., (1993) An appraisal of Constructivist Psychotherapies. Journal of Consulting and Clinical Psychology. 61. 221-234.

Padesky, C. (1993). Socratic questioning: changing minds or guided discovery ? European Congress of Behavioural and Cognitive Therapies, London.

Power,M. & Brewin,C.R.Eds.(1997) The Transformation of Meaning. Chichester: Wiley.

Ryle, A. (1995). Cognitive Analytic Therapy. Chichester, Wiley.

Ryle, A. (1997). Cognitive Analytic Therapy and Borderline Personality Disorder. Chichester, Wiley.

Safran,J.D., & Segal, Z.V. (1990). Interpersonal Processes in Cognitive Therapy. NY. Basic Books.

Segal Z.V.,(1988) Appraisal of Self-Schema Construct in Cognitive Models of Depression. Psychological Bulletin, 103, 147-162.

Teasdale, J.D. & Barnard, P.J. (1993) Affect, Cognition and Change: Remodelling Depressive Thought. Hove:Lawrence Erlbaum Associates.

Teasdale, J.D. The Transformation of Meaning: The Interacting Cognitive Subsystems Approach. In Power,M.and Brewin,C.R.Eds.(1997) The Transformation of Meaning. Chichester: Wiley.

Trower,P., Casey, A. & Dryden,W.,(1988). Cognitive Behavioural Counselling in Action. London, UK. Sage.

Wells A., White,J. & Carter, K. (1997) Attention Training: Effects on Anxiety and Beliefs in Panic and Social Phobia. Clinical Psychology and Psychotherapy, 4, 226-233.

Young, J.E. (1994) Cognitive Therapy for Personality Disorders: a Schema Focussed Approach. (Rev. ed.) Sarasota, FL: Professional Resource Press.