Perry, A., 2012. CAT with People who Hear Distressing Voices. Reformulation, Summer, pp.16-22.
The CAT model provides us with the ideal tools for making sense of hallucinatory voices because the language of voice and dialogue is at its heart (Ryle & Kerr 2002). The aim of this article is to integrate some of the recent developments from outside of CAT in understanding voice-hearing with the developing CAT models of psychosis. In addition, the article aims to explore how CAT tools could be applied to different types of voice experiences and also how the CAT model of voice hearing could be usefully developed. Anonymised and altered case examples are provided for illustration.
There is a weight of evidence that traditional categorical approaches to diagnosis are flawed and unreliable (Bentall et al. 2007; Bentall 2003; Carpenter et al. 2009). It may therefore be most useful think of psychosis and ‘Schizophrenia’ in terms of specific types of experiences (i.e. voices, paranoia, flashbacks etc.) which have interrelated but separate developmental trajectories. By using a CAT framework to examine ideas from both mainstream cognitive science and the more radical service user-researcher perspective, I aim to present some possibilities as to what such trajectories might look like.
The CAT account of Psychosis described by Ian Kerr & colleagues (Kerr et al. 2006; Kerr et al. 2003) describes psychosis as the result of a breakdown of normal ego functions (such as reality testing, relationship formation, impulse & affect control) in the face of extreme or conflicting emotional demands. Rather than the traditional Freudian view of the Ego becoming overwhelmed by the intolerable drives and demands of the Id, Kerr and colleagues see this breakdown as arising from a combination of neurodevelopmental vulnerability and psychosocial stress. The intolerable feelings resulting from any kind of problematic RRP can therefore lead a person to shift into a “Psychotic State”, if he or she has certain vulnerabilities to doing so. Experiences such as voices are conceptualised as “muddled, amplified or distorted” versions of a person’s Reciprocal Role repertoire and “disordered enactment and dissociation of RRs” (Kerr, Crowley, & Beard, 2006. p 175). To give an example: a lived experience (such as a critical other) gives rise to a corresponding internalised dialogical ‘voice’ (e.g. critical-criticised), which is in turn defensively dissociated and experienced as an externalised (critical) voice.
Kerr & colleagues’ model draws upon the widely used ‘Stress-Vulnerability’ paradigm (Zubin & Spring 1977) but also points out its weaknesses – namely that it doesn’t allow for the complex and dynamic interplay between ‘stresses’ and ‘vulnerabilities’. The CAT model draws upon developmental neuropsychology (e.g. Fox et al. 1994) to develop the idea further by showing how certain neuro-cognitive difficulties (for example, a problem understanding the world of others) may themselves invite difficult or harsh responses from other people. In addition, internalised RRPs would be ‘self-stressful’ to the individual even when the original source of the distress was not present. Extreme or amplified enactments of procedures in response to psychosis (such as hyper-vigilance for a persecutor) may also lead to amplified experiences of distress. The resultant disruption to normal levels of stress hormones and neurotransmitters could in turn be toxic to a person’s brain, particularly if it is still developing.
Kerr et al. acknowledge that there is much debate about the specific types of neurodevelopmental factors (such as information processing biases, motivation, emotional regulation) that could lead to a vulnerability for psychosis and they conclude that “it is generally accepted that the impairment of consensual reality-testing and of executive function associated with... all psychotic disorders will be generated by both vulnerability factors and psychosocial stressors” (Kerr, Crowley, & Beard, 2006. p. 174). Richard Bentall and colleagues (Bentall et al. 2007; Bentall 2010) present a summary of the evidence for the various potential biological factors and conclude that despite huge amounts of investment and research interest, there is no strong evidence for any particular factor. Evidence such as a correlation between a Schizophrenia diagnosis and enlarged brain ventricles is often cited as proof of a biological vulnerability. However, Bentall and colleagues present a weight of evidence that such factors are far better predictors of general psychopathology and that any link to a specific psychotic experience, such as hearing voices, is statistically weak at best. They also demonstrate how all psychotic experiences (dissociation, unusual beliefs, hallucinations, ‘negative’ symptoms) can be clearly explained in terms of normal human development and responses to early trauma (some of which I will describe in more detail later). It therefore seems important to champion the contribution that CAT can make to understanding the psychological roots of psychosis and to push harder against the dominance of an overly biological view.
TD experiences relentless voices which began following the traumatic bereavement of his brother and a period of heavy amphetamine use. He felt both responsible for his brother’s death and also that he had lost the only source of protection from a punitive and shaming family environment. TD suspected that the voices must be due to a malicious campaign involving small electronic speakers because otherwise why hadn’t they gone when he stopped taking drugs and why did they completely go when he went on holiday (out of range of the speakers)? I did a short-term piece of work with him and his care-coordinator (Figure 1). The diagram we drew helped us to see how the shaming g ashamed RRP had been around a long time before it had become a voice. Sleeplessness and drug use seemed to be the catalyst that allowed the split to occur and it was then maintained by his behaviour. We saw how he fuelled the voices’ games by frantically trying to prove them wrong and by trying to distance himself from what they were saying – by preventing this part of himself from having a real voice, it had become more and more assertive in the only way available. In addition, the CAT framework helped us to see how being on holiday where no-one knew him was instantly less shaming and so the voices were not triggered. This alternative explanation for TD’s voices was one which he could intuitively accept and which seemed within his ZPD. This allowed us space to talk about alternative explanations without getting caught up in whether they were “real” or “not real”. The new explanation also helped to undermine some of the power that the voices had by identifying how he was keeping them going. I am hopeful that it will lead to future therapeutic work relating to the parts of himself that are warded off.
There is growing evidence (e.g. Beavan 2011) that people have wildly different relationships (both positive and negative) with their voices and that these relationships are most problematic when the voices are seen as powerful and in control (P. Gilbert et al. 2001). Could it be that there are specific types of reciprocal roles and procedural responses which might make someone more likely to hear voices, rather than suffer psychological distress in some other way? In the case example described above, it seemed clear to me that TD was not simply misconstruing a disowned part of himself as coming from outside, but rather that he was actively engaged in a battle where the more the voice was suppressed, the more aggressive and assertive it became. It therefore seemed like the specific problems of ego function that he experienced (such as parts of the self seeming to operate independently) could be seen to relate to specific procedures that he engaged in. Conversely, other aspects of ego functioning seemed to remain relatively intact. For example, his ideas about speakers in the walls etc. might suggest a difficulty with ‘reality testing’ but in fact he was extremely calm and reasonable about his theories, he didn’t jump to conclusions and he actively sought exceptions and alternative evidence. He told me that the only reason he thought the voices might be real people was because they sounded like real people.
In their work looking at the links between trauma and voices, Romme and Escher (2006) identify some of the coping styles that could specifically lead to voice hearing. Their studies showed a common finding that the onset of voice hearing seems to depend not on the nature of the trauma but on the way the individual coped with it. They conducted a qualitative study with voice hearers who’d experienced trauma and identified seven common pathways from trauma to voice hearing. All seven pathways describe an attempt to find a solution when faced with traumatic memories and associated emotions which feel impossible to accept or tolerate. In CAT terms, this could perhaps be seen as the need to resolve a dilemma between reality and oblivion. The authors also describe how different solutions are likely to impact on a person’s emotional development in different ways and may lead to different types of voice hearing experiences. Each Pathway is presented in Table 1 with some suggestions of related CAT states and procedures.
The first pathway describes how children who experience sexual abuse often replace themselves with a dissociated substitute-self whilst the abuse is being carried out. This self is no longer needed when the abuse has stopped and the traumatic memories have been repressed, but it returns to consciousness when the wall of repression breaks down. This voice is likely to be at the emotional age that the child was and it acts as a reminder of what happened. Not surprisingly, individuals often respond fearfully or aggressively to this type of voice and so inadvertently re-enact the abuse and rejection that the ‘child self’ originally experienced. The voice’s attempts to be heard and cared for become increasingly desperate and aggressive and so the person and the voice become increasingly alienated from one another.
Pathway 2 describes the process well-recognised in CAT where abusive experiences are translated into self-self RRPs. We can use the ‘split egg’ RRP diagram to represent how a dissociative wall may ward off the most vicious of these attacks and protect an individual from annihilation. It may be that the parts of the self are so much in conflict that the structure of the ego collapses entirely and the parts of the self are experienced as separate ‘entities’. Alternatively, it may be that a person is able to function reasonably well but the critical/attacking self is forced to occupy a space of its own and so be heard as a voice. A related process identified by Romme and Escher is where an individual denies feelings of aggression towards another (for example a suffocating caretaker, or an abuser who is also vulnerable) and instead channels these feelings towards him/herself. Voices are hypothesized to occur when the person’s circumstances no-longer match this familiar self-attacking RRP (e.g. when a teenager moves out of home, or when someone escapes an abusive relationship).
Pathway 3 describes an attempt to shut out all emotions (e.g. by becoming entirely consumed by activity) which is destined to fail. The authors are not clear about why some people who do this might hear voices and some people might not. Pathway 4 (often associated with being in a ‘high’ state) describes how voices sometimes represent a fantasized escape from an impossible situation. One example from my own work is a person who hears the voice of a girl he once met telling him she loves him. Pathway 5 (also linked to mania) describes how voices might be linked to someone putting him/herself under a great deal of pressure to achieve – the voice(s) represent a (disowned) critical other snapping at his/her heels. Pathway 6 describes how events in a person’s life can sometimes clash intolerably with his/her construction of the world, such as being unable to accept that a relationship is over, or finding one’s own sexual impulses abhorrent. In these circumstances, the voice is similar to an OCD intrusion, voicing mental events which the person is frantically trying to suppress. Pathway 7 is seen as akin to many religious beliefs where the voice acts as a powerful ‘rescuer’ from an intolerable situation. The person is locked in an endlessly repeating RRP (e.g. striving to please leading to mistreatment from others) and the voice presents the only exit that seems possible.
Clearly, Romme and Escher’s categories are from only a small number of studies and need replication and expansion. However, the categories do make intuitive sense to me - I can recognise them (or combinations of them) in all the clients I have worked with and can’t readily think of exceptions. It seems that the common factor across all the domains is the active suppression of a particular self-state which is not compatible with the rest. This specific type of procedure links with what Tait, Birchwood and Trower (2004) dubbed a ‘sealing over’ coping style and presents a useful extension to the CAT model of how stresses and vulnerabilities interact.
Superficially, there seems to be an overlap between Kerr and colleagues model of Psychosis and CBT models (e.g. Garety, Kuipers, Fowler, Freeman, & Bebbington, 2001) in the way they describe how ego-functioning may be compromised when people hear voices. In both models, voices are seen to occur when mental events are misconstrued (or actively fantasised) as coming from an external source – known as ‘source monitoring’ disruption (Johnson et al. 1993). Some compelling evidence for this phenomenon comes from imaging studies that show activity in brain areas and muscles associated with speech when people are hearing voices (Szechtman et al. 1998). Another striking research finding is that if voice hearers listen to distorted recordings of their own voices, they are more likely to attribute these to a different person (Frith et al. 2001), especially if the content is troubling to their sense of themselves (Bentall 2003). However, evidence that such a phenomenon exists does not necessarily mean that source monitoring errors are the whole of the story. Talking to people who hear voices, I have found such a description woefully inadequate to describe the complex relationship that people have with their voices.
The CAT model provides a useful extension to the standard cognitive explanations by integrating them with traditional Psychoanalytic concepts of psychosis. Rather than the traditional focus on unconscious internal conflict, the emphasis here is on the impact of early and current trauma and a reversion to developmentally primitive defences and self-organisation (Pollock 2001). In response to overwhelming distress, the various parts of the self become ‘out of dialogue’ with one-another and an overall sense of the self as coherent, continuous and separate from others is lost (P. Lysaker & J. Lysaker 2001). Therefore, it is not simply that an individual has the perception that a part of him/herself is separate but that this part of the self is operating as such.
The Multiple Self States Model (Ryle 1997) developed for people labeled with Borderline Personality Disorder provides a useful framework for understanding the specific process that may underlie voice hearing. Using the model, we can see how people hearing voices may have lost some specific ‘level 2’ and ‘level 3’ capacities, such as being able to mobilise (or not) certain self-states and being able to self-reflect. In BPD (and Dissociative Identity Disorder), individuals seem to occupy one self-state at a time but have difficulty managing the ‘flow’ between them. In my experience of working with voice hearers, some people seem to present in a very similar way, shifting from a ‘well’ state with no voices at all to an acutely ‘psychotic’ state where voices are ever-present (similar to that described by Graham (1995)). For other clients (such as TD, described earlier) the voice seems to occur all the time (perhaps varying with distress or the triggering of certain RRPs) but ego functioning remains otherwise relatively intact. I have also noticed that the longer a voice is in someone’s life, the more it seems to develop an independent personality and agenda. In this case, I wonder if it is possible that a slightly different dissociative process is going on where two self states are occurring simultaneously?
CBT models which advocate working with an individual’s metacognitive beliefs (beliefs about their voices) have shown significant clinical benefits by helping clients to have a greater sense of power and control over their voices (Morrison 2001; Chadwick & Birchwood 1994). These approaches focus on doing experiments and finding evidence relating to the perceived omnipotence and omniscience of the voices (e.g. their ability to carry out threats) but do not advocate engaging with the voices beyond this. By pitting themselves against the voice(s) in this way, there is a risk that the client and therapist miss the opportunity to find out what the voice might be communicating, or what its needs are. CAT provides some alternative ways to negotiate power from a person’s voices in a way that maintains a dialogue and relationship. Key to this is how the SDR makes both the current and desired relationship with the voices (in the form of exits and exit RRPs) explicit. Therefore, the exits identified apply equally to all aspects of the client’s life, including the way he/she relates to voices and the way the voices respond. Lessons learned from self-self and self-other interactions can be applied to interactions with voices and vice versa. In this way the voices are brought into the work far more closely, rather than treating them as troublesome ‘symptoms’ which must be overcome. A CAT approach also means that (even when a person is entirely convinced that a voice is not part of them), power and freedom from unhelpful procedures are negotiated between the person and the voice, as they would be in any other relationship.
Using CAT as a tool for negotiating power from voices has some overlap with a pioneering and radical approach to working with voices by Dirk Corstens, Rufus May and Eleanor Longden (Corstens, May, & Longden, 2011). Their ‘Talking with Voices’ method is informed by the Voice Dialogue approach (Stone & Winkelman 1998) and sees voices as representing some of our multitude of inner selves (or self-states in CAT). The method requires practitioners to speak directly to a person’s voice(s) by asking them to sit in a different chair and either speak as the voice or act as ‘translator’. By treating the voice with respect and asking questions about its role in the person’s life and about what it needs or desires, the person gets to know what underlies the things the voice says. In this way, the relationship is changed dramatically - for example, from (coerciveïƒ g fearfully obedient) to (advising g supported). I have used this approach a handful of times and have found it startlingly effective. For example, my client (PB) heard the voice of a woman criticising and abusing him and felt angry and confused by what she said. By helping him to have a dialogue with the voice, he was able to find out that the voice was fearful that he (they) would be hurt or abandoned. The voice had become increasingly aggressive to PB because she did not feel that he was listening to her and because she felt he was being reckless with his heart in relationships. The approach helped him to get in touch with this dissociated part of himself in a way that would have been very time consuming and maybe impossible via standard psychological work. A simple CAT diagram (figure 2) helped us to use what ‘Julie’ had said to better understand his relationship with her and with himself. The dotted line shows ‘Julie’s’ RRP (which PB doesn’t recognise in himself) next to PB’s dominant RRP. The spark between the two RRPs shows the conflict between the way Julie and PB experience a situation and between the strategies they employ. The diagram not only helped us to identify possible exits from PB’s procedural repertoire but also to tentatively look at how he and ‘Julie’ took each of the 4 poles at various times in relation to one another. Entering into the client’s world to the extent that voices are unquestioningly treated as separate beings (described as ‘Radical Collaboration’ by Paul Chadwick  ) may seem like a step too far for many therapists but I believe that doing so could dramatically increase the potential of CAT-based interventions.
Figure 2: Simple CAT Diagram of Voice Dialogue Work with PB
CAT is a powerful tool for framing a psychological account of how a person has come to hear voices and how he or she may find relief from distress. In addition to its accessibility for clients, it may also help us to further our theoretical understanding of the relationship between dissociation and voice hearing and also the particular RRPs that are most commonly at play. One important development might be for us to be clearer about what we mean when we use the term ‘dissociation’ – its meaning seems to range from a child becoming absorbed in the TV to the most extreme breakdown of personality function.
This paper has identified three possible options for showing a person’s voices on his/her SDR. As in the example PB (Figure 2), a person may wish to have the voice represented as a separate RRP (similar to the ‘split-egg’ SDR). For other people, it may make most sense to present the voice as resulting from a psychotic state (e.g. Graham, 1995). Thirdly, the voice could just be present on the diagram, interacting with their RRPs but not necessarily ‘owned’ by them, as with the example TD (Figure 1). In all three cases, it may be helpful to modify the standard CAT model (such as that for BPD) so that rather than moving sequentially between self states, two or more can be active at once, occupying different parts of ‘intersubjective space’.
The examples and research findings presented here suggest that a key aspect of voice hearing is the active suppression of traumatic memory or intolerable parts of the self. It therefore seems that an important part of reformulation (within the person’s ZPD) is to point out that voices may be the “canary in the coal mine”, indicating that a part of the self needs to be heard. In addition, CAT is the ideal tool for working to improve a person’s relationship with his/her voices in a genuinely relational way – operating in a grey area between self-self and self-other relationships.
Beavan, V., 2011. Towards a definition of “hearing voices”: A phenomenological approach. Psychosis, 3(1), p.63-73.
Bentall, R., 2010. Brains, Minds and Psychosis: The Myth that Mental Illnesses are Brain Diseases. In Doctoring the Mind. Why psychiatric treatments fail. London: Penguin, pp. 148-182.
Bentall, R., 2003. The Illusion of Reality. In Madness Explained: Psychosis & Human Nature. London: Penguin, pp. 347-377.
Bentall, R. et al., 2007. Prospects for a cognitive-developmental account of psychotic experiences. The British journal of clinical psychology / the British Psychological Society, 46(Pt 2), p.155-73. Available at: http://www.ncbi.nlm.nih.gov/pubmed/17524210 [Accessed September 23, 2010].
Carpenter, W. et al., 2009. The psychoses: cluster 3 of the proposed meta-structure for DSM-V and ICD-11. Psychol Med, 39(12), p.2025-42.
Chadwick, P., 2006. Person Based Cognitive Therapy for Distressing Psychosis, London: Wiley.
Chadwick, P. & Birchwood, M., 1994. The omnipotence of voices: a cognitive approach to auditory hallucinations. British Journal of Psychiatry, 164, p.190-201.
Corstens, D., May, R. & Longden, E., 2011. Talking with Voices Manual, Unpublished Manuscript.
Fox, N.A., Calkins, S.D. & Bell, M.A., 1994. Neural plasticity and development in the first two years of life: evidence from cognitive and socioemotional domains of research. Development and Psychopathology, 6, p.677-696.
Frith, F. et al., 2001. Verbal self-monitoring and auditory hallucinations in people with schizophrenia. Psychological medicine, 31, p.705-715.
Garety, P. et al., 2001. A cognitive model of the positive symptoms of psychosis. Psychological medicine, 31(2), p.189-95. Available at: http://www.ncbi.nlm.nih.gov/pubmed/11232907.
Gilbert, P. et al., 2001. An exploration of evolved mental mechanisms for dominant and subordinate behaviour in relation to auditory hallucinations in schizophrenia and critical thoughts in depression. Psychological medicine, 31, p.1117-1127.
Graham, C., 1995. Dissociative psychosis: An atypical presentation and response to cognitive-analytic therapy. Irish Journal of Psychological Medicine, 12(3), p.109-111.
Johnson, M.K., Hashtroudi, S. & Lindsay, D.S., 1993. Source Monitoring. Psychological Bulletin, 114(1), p.3-28.
Kerr, I.B., Birkett, P.B.L. & Chanen, A., 2003. Clinical and service implications of a cognitive analytic therapy model of psychosis. The Australian and New Zealand journal of psychiatry, 37(5), p.515-23. Available at: http://www.ncbi.nlm.nih.gov/pubmed/14511078.
Kerr, I.B., Crowley, V. & Beard, H., 2006. A Cognitive Analytic Therapy-Based Approach to Psychotic Disorder. In J. O. Johannessen, B. V. Martindale, & J. Cullberg, eds. Evolving Psychosis. London: ISPS, pp. 172-184.
Lysaker, P. & Lysaker, J., 2001. Psychosis and the disintegration of dialogical self-structure: Problems posed by schizophrenia for the maintenance of dialogue. British Journal of Medical Psychology, 74(1), p.23-33.
Morrison, A.P., 2001. The interpretation of intrusions in psychosis: an integrative cognitive approach to hallucinations and delusions. Behavioural and Cognitive Psychotherapy, 29, p.257-276.
Pollock, P., 2001. Cognitive Analytic Therapy for Adult Survivors of Childhood Abuse: Approaches to Treatment and Case Management, London: Wiley.
Romme, M. & Escher, S., 2006. Trauma and hearing voices. In Trauma and Psychosis. London: Routledge, pp. 162-192.
Ryle, A., 1997. The Multiple Self States Model of Borderline Personality. In Cognitive Analytic Therapy for Borderline Personality Disorder. London: Wiley.
Ryle, A. & Kerr, I.B., 2002. The Main Features of CAT. In A. Ryle & I. B. Kerr, eds. Introducing Cognitive Analytic Therapy. Chichester: Wiley, pp. 6-20.
Stone, H. & Winkelman, S., 1998. Embracing Ourselves, Novato, CA: New World Library.
Szechtman, H. et al., 1998. Where the imaginal appears real: a positron emission tomography study of auditory hallucinations. Proceedings of the National Academy of Sciences, 95, p.1956-1960.
Tait, L., Birchwood, M. & Trower, P., 2004. Adapting to the challenge of psychosis: personal resilience and the use of sealing-over (avoidant) coping strategies. The British journal of psychiatryâ€¯: the journal of mental science, 185, p.410-5. Available at: http://www.ncbi.nlm.nih.gov/pubmed/15516550 [Accessed January 18, 2011].
Zubin, J. & Spring, B., 1977. Vulnerability: A new view of schizophrenia. Journal of Abnormal Psychology, (86), p.103-126.
Reflections on a Dilemma in a Supervision group: Caught between a Rock and a Hard Place
Gil-Rios, Dr. C., M., and Blunden, Dr. J., 2012. Reflections on a Dilemma in a Supervision group: Caught between a Rock and a Hard Place. Reformulation, Summer, pp.23-25.
Sleep Tight: Working Creatively with Dreams in CAT
Coulter, N. and Rushbrook, Dr S., 2012. Sleep Tight: Working Creatively with Dreams in CAT. Reformulation, Summer, pp.7-12.
Using a Cognitive Analytic Therapy approach in working with Eating Disorders: Reflections on Practice
Wicksteed, Dr. A., 2012. Using a Cognitive Analytic Therapy approach in working with Eating Disorders: Reflections on Practice. Reformulation, Summer, pp.26-31.
What is the 2005 Mental Capacity Act and how can CAT help us to make sense of the decision making process at its heart?
Brown, H. and Lloyd, L., 2012. What is the 2005 Mental Capacity Act and how can CAT help us to make sense of the decision making process at its heart?. Reformulation, Summer, pp.35-42.
This site has recently been updated to be Mobile Friendly. We are working through the pages to check everything is working properly. If you spot a problem please email firstname.lastname@example.org and we'll look into it. Thank you.