Cognitive Analytic Therapy

Annabel McDonald
Sanaz Fallahkhair
Jessica A Eccles 


Cognitive Analytic Therapy (CAT) is a brief form of psychotherapy that was initially developed for use by the British National Health Service (NHS). Its use has now spread across Europe and it is also recognised in Australia and New Zealand. Clients generally receive 8-24 weekly sessions of therapy that each last around an hour. The approach is collaborative where the therapist and client work together to understand habitual relationship and behavioural patterns. Once these have been identified, the pair then consider alternative more functional ways of thinking and behaving. This work involves considering poor early formative relationships and considering how their nature can be replicated throughout the ongoing years. This insight leads to the possibility of developing to a more mature and successful forms of relationships.



Cognitive Analytic Therapy was first developed in the early 1980’s by Anthony Ryle [1] in London. He realised that patients receiving psychodynamic therapy generally revealed a few key problems early in therapy and that these were repeated through a series of behaviours that he referred to as snags, dilemmas and traps [2] Snags were behavioural loops where the patient failed to attempt movement in a new direction because he ‘knew’ that it would fail. Dilemmas represented black and white thinking where the patient was unable to find a ‘third way’ – if I see people I get hurt, if I don’t see them I am lonely. Traps described repeated behaviour that ‘proved’ pessimistic viewpoints, thus, people don’t like listening to me because I’m boring, so I don’t talk – which makes me boring. He started to examine patients from the perspective of their relationships rather than as individuals and made use of Kelly’s Personal Construct Theory [3].This is a system that can be used to evaluate the patient’s perceived views of those around him by comparing their relative strengths across a number of criteria such as kindness or reliability.

The next phase in the development of CAT was Ryle’s introduction of object relations theory to his understanding of personal development. He developed a theory describing the development of mind, based on Bowlby’s attachment theory [4], Vygotsky’s sign mediation and social formation of mind [2] and Bakhtin’s work on dialogism [5].

This sees us as moulded by our environment, particularly our primary caregivers, from birth. We carry the internalised voices of all those who have surrounded us throughout our lives, conveying information at both the individual and the cultural level. These voices govern our relationships even those within ourselves where they can allow us to treat ourselves kindly or, at the other extreme, with a continuous volley of self-criticism.

CAT is seen as a transdiagnostic therapy [6] and is one of the few therapies that offers a complete therapeutic approach to personality disorders [7] in addition to anxiety, depression, eating disorders and the survivors of childhood abuse [8].


The CAT Understanding of the Aetiology of Psychological Distress

Object relations recognised the human psyche as being primarily person-seeking [9] with communication through attunement and empathy long before language develops. These ‘proto-conversations’ can bring a baby through periods of stress as he is calmed by those around him [10, 11]. This introduces the baby to a predictable environment where his developing brain is not deluged with stress hormones. This stage of the baby’s development appears to be a wash of emotional responses to both the internal and external influences around him that have yet to be understood in the language which has not yet developed. He is learning, through direct experience, to interpret and anticipate his surroundings as well as to copy those around him. These are true relationships and are likely to leave their mark his mental state and orientation to others for the rest of his life. As language comes into the picture, he is able to start to learn about the world indirectly and to form a cognitive view which offers a further basis for emotional regulation. Not only is he now able to manage his emotions by copying and internalising his caregiver’s soothing activities; he can now receive spoken information about the nature of those emotions and their transitory nature. Through these interactions he is able to develop a perspective on his emotional experiences.

Unfortunately, many psychological patients fail to receive this idealised upbringing – they are left at the mercy of emotions which do not appear to have any pattern and cannot be controlled. The problem is compounded if the child’s environment is also stressful due to abuse or neglect as they will be further harmed by a need to avoid the resulting torrent of emotions. These situations tend to result in the construction of defences, perhaps a refusal to care about the actions of others or a need to hide from all contact. These defences in themselves can be damaging – the child hiding from all contact is likely to develop habitual behaviours of this nature, perhaps resulting in an avoidant personality disorder. The former child is ripe for the development of an asocial personality disorder. The lack of perspective on emotions and the construction of ultimately unhelpful need not be so severe as to result in a personality disorder. They can simply leave the child with feelings of insecurity or a lack of self-esteem which may make them vulnerable to anxiety and depression when they meet challenges later in life.

These considerations led Ryle to concentrate further on relationships [2]. He proposed that children learn about the relationships to which they are exposed and that these are dialogic. Thus the bullied child will understand the relationship of dominant abuser to victim. He will understand both aspects and will therefore be able to play either of the two roles himself. This understanding of the role of abuser much simplifies the previous psychodynamic view of projecting the ‘bad’ aspects of oneself on to others. A child from a difficult background will tend to have a more limited set of learnt relationships with more negative connotations which will set a negative slant on their future relationships. It is, sadly, difficult to behave as a kind and loving parent and accept care graciously if you have never experienced that type of relationship.

The more damaged individuals who develop personality disorders may have such poor perspective regarding their emotions that they may ‘switch’ between emotional states, and thus relationships, with little awareness of the change [12]. Thus an individual with an unstable personality disorder may feel criticised and simply jump from a calm state to an angry state without any meta-awareness of the event. This can bring particular difficulties to therapy due to the volatility of the patient and the fact that he is unable to track his emotions and behaviour due to the loss of knowledge of his previous state prior to a switch.


CAT Treatment Aims: [2]

CAT aims to enable the patient to develop metacognitive skills such that he gains the perspective to be able to ‘watch’ his behaviours and the play of his emotions. Once this perspective is achieved, the patient is no longer at the mercy of those emotions but can start to understand the reasons for those feelings and consider whether they are appropriate. He is empowered to start choosing his responses to environmental stimuli. This perspective is provided by an empathic examination of the patient’s behaviours within varying situations and relationships. These include those experienced as a child, his current life and interactions with the therapist ‘in the room’. Careful observation tends to lead to recognition of a few dysfunctional patterns of behaviour that the patient repeatedly plays out within different situations.

These behaviours have generally been learnt from childhood relationships or as a defence against psychological pain caused by those relationships. An example of the former would be a patient whose mother was unreliable and would neglect her children. This would teach the patient about a relationship where disinterest caused feelings of worthlessness. He would therefore be hypersensitive for any impression of disinterest in current relationships which might instantly return him to those childhood emotions of painful worthlessness. Alternatively he might have developed a defence, as a child, of persuading himself that he did not want any attention from his mother. This could lead to patterns in later relationships of being unable to risk any dependence on others. Once the patient is able to recognise the occurrence of these patterns and to understand their origins, he is in a position to question his feelings and responses, and then to change those that he feels are unhelpful and perhaps belong in the past.

This wakening is not provided by the therapist acting as an expert – it is the patient who is the expert in his own history and behaviour. The role of the therapist is to encourage and guide the patient in his own exploration, offering him a safe and containing environment to cope with the stress of the process. The therapist is also there to help the patient see events from a more external perspective – collusion with a patient’s warped view of events, perhaps when he always sees himself as a victim, will only exacerbate his problems. This calls for an ability to challenge the patient’s views without causing irreparable damage to the therapeutic relationship. These ruptures can often be resolved by considering how the patient’s response is echoing his habitual procedures [13] It can be seen that this is a collaborative process where the therapist and patient work together to understand processes, look for patterns, and discuss alternative approaches. This very lack of ‘expertise’ on the part of the therapist opens the patient to a less certain approach where he is encouraged to accept fluidity in his thinking and search out his own views and opinions.

The patient, as an active proponent of therapy, is not encouraged to regress and develop excessive dependence on the therapist – this could be dangerous in the setting of brief therapy.


CAT in Practice: [2]

The collaborative nature of CAT therapy has significantly changed the traditional psychological formulation which is the document where the psychotherapist establishes the patient’s problems, their aetiology and maintaining factors – followed by a realistic plan for treatment. This has become a shared document in CAT which the therapist gives to the patient, usually around the fifth session. At this point the pair will make a decision as to the planned number of sessions. This is generally 16 or 24 sessions but may be limited to 8 sessions in a simpler case when the reformulation will have occurred at an earlier point. The preceding sessions enable the pair to contract the terms of therapy, explore the patient’s view of his problems and allow the therapist to develop an understanding of the patient’s past and current relationships. The empathy of the developing relationship is often accelerated by the ‘gift’ of the reformulation letter as it indicates to the patient that he has been heard and that the psychotherapist is accepting of his problems and hopeful of forward progress. The letter is sometimes written in conjunction with the patient and is a tentative document that is open to revision where the patient feels that it is inaccurate. This is an important stage in treatment as it ensures that the therapeutic relationship is based on a shared baseline understanding. The patient is encouraged to consider offering a written reply to the letter.

The reformulation process is developed further with the construction of a flow diagram that shows the patient’s habitual dysfunctional relationships and the manner in which the patient’s behaviours keep replaying them. The attempts of the patient to escape from the resulting pain – such as feelings of being ‘bad’ – tend to return them to the pain or the dysfunctional relationships. Thus being ill-treated as a child may teach the patient to expect relationships to fall into that of abuser and victim. The resulting feeling in a child will often be that of being ‘bad’ and of deserving to be abused. The pain of this feeling of ‘badness’ can lead to the patient abusing himself through self-harm and returning himself to the role of victim and thus restarting the cycle. This diagram is constructed by the therapist and patient together and, like the reformulation letter, helps the patient to start to recognise dysfunctional patterns in his behaviour. He will often come to realise that many of his emotions are unrelated to current issues but remnants of past distress leading to relationship styles that are no longer appropriate[14]. Many patients find this graphic representation of their behaviours very helpful in the development of their metacognitive skills.

The middle phase of therapy is very fluid in CAT and other forms of therapy may be incorporated. Examples of this might be art therapy, mentalisation and cognitive behavioural techniques. They are always directed toward the recognition and revision of the habitual behavioural patterns in the patient’s day-to-day behaviour. The diagram can, at this point, be a very helpful adjunct in that a patient can recount a recent event and the pair can then trace out the events on the diagram. Thus a patient who is upset by harsh words from his sister might be able to trace out a known process on the chart that includes ‘feeling criticised’ leading to ‘hurt’ which then makes the patient angry and so on. Thus the specific circumstances can be recognised as part of a more generalised pattern and the patient’s response can be better understood. Once the patient is able to ‘recognise’ the patterns in their daily behaviour, the pair can start to talk about ‘revision’.

Revision is the second part of the work in this middle phase. This involves the patient realising that he can choose to behave in other manners and the pair can discuss, perhaps play out, some of these alternatives. The patient can then try this when a similar situation next arises and evaluate whether this is a more effective response. The patient is now moving on from ‘recognising’ a dysfunctional behaviour to identifying an alternative response that moves him out of the cycles that normally return him to dysfunctional relationships and his inner pain. This alternative is marked on the patient’s chart as an ‘escape’ that he can use on future similar occasions so that it becomes a permanent behavioural modification.

The final part of therapy is the ending. This is a period when the patient is internalising both the work covered and the relationship with the therapist. This will normally have been a warm and empathic relationship which gives the patient another, hopefully successful, relational template to add to their repertoire. The ending can feel very threatening for the patient who may be frightened and lonely at his impending loss. It is important that the ending is managed well so as to ‘repair’ the effects of difficult previous losses and to help the patient deal with future separations. The patient is encouraged to recognise the gains from the relationship and realise that these will be carried forward within him. He needs to see himself as ready to return to independence and to accept responsibility for his ongoing journey. The final ending is marked by an exchange of ‘good bye letters’ which serve as transitional objects as he moves forward. The letter from the therapist sketches out the events of therapy and recognises gains and perhaps limitations that have occurred. It looks forward to the patient’s future path, anticipates possible problems and wishes him well.

Many therapists see the patient once or twice more, perhaps three months out, for a follow-up appointment. This helps the patient to plan forward at the end of therapy and allows the discussion of any issues that arise within the ensuing period.

The CAT process thus includes a number of recognised components or ‘tools’ but it is interesting to note that there is nothing sacrosanct about them, they are simply tools to fulfil the aims of CAT therapy and support the role of the therapist – as recognised by Rayner et al (2011). This view is reinforced by recent observations that omitting the Reformulation Letter did not make a significant difference to the results of therapy [15]


The Evidence Base for CAT

A recent review of the evidence for the effectiveness of CAT comments that the therapy has been widely adopted with little formal evidence [8] The report describes the fact that they found only five RCT’s and that the participant numbers in all types of research reports were small, typically 5-18 patients with two larger studies following 38 and 41 patients – this situation does not appear to have changed. The published reports are encouraging about efficacy of the treatment but offer insufficient evidence for NICE to recommend the therapy. This is partly a reflection of the fact that NICE examines evidence by illness category whereas CAT is a transdiagnostic treatment [6]. Another problem is that CAT tends to be used for complex patients where other forms of treatment such as CBT would be inappropriate comparators.



Cognitive Analytic Therapy is an interesting combination of psychodynamic-type considerations of aetiology with the more practical and collaborative approach of behavioural therapies. It has proved very satisfying for therapists who form a close and enthusiastic group. It is a potentially valuable medium for Psychiatry trainees as it offers the potential to use a range of disciplines during the middle stages whilst offering a well-scaffolded structure in the early phases of treatment.
It is clear that the future of Cognitive Analytic Therapy will be reliant on more disciplined research to allow its value – or otherwise – to be properly demonstrated but it certainly offers another potential approach to the complex problems of personality disorder and comorbid psychiatric morbidity.



[1] Ryle A. Cognitive Analytic Therapy: Developments in theory and practice, Oxford, England, John Wiley & Sons 1995.

[2] Ryle A, Kerr IB. Introducing Cognitive Analytic Therapy: Principles and Practice, Wiley 2003.

[3] Scheer J The internet encyclopaedia of personal construct psychology [Online]. [Accessed] 2004.

[4] Gomez L. An introduction to object relations, NYU Press 1997.

[5] Holquist M. Dialogism: Bakhtin and his world, Routledge. 2003

[6] Frank M. Editorial: Cognitive analytic therapy: A case study in treatment development. British Journal of Medical Psychology 2000 73, 145-150.

[7] Bateman AW, Ryle A, Fonagy P, Kerr IB. Psychotherapy for borderline personality disorder: mentalization based therapy and cognitive analytic therapy compared. International review of psychiatry 2007. (Abingdon, England), 19, 51-62.

[8] Calvert R, Kellett S. Cognitive analytic therapy: A review of the outcome evidence base for treatment. Psychology and Psychotherapy: Theory, Research and Practice, 2007. 87, 253-277.

[9] Thomas K. ‘The Defensive Self: A Psychodynamic Perspective, Thousand Oaks, CA: SAGE Publications . 1996.

[10] Gerhardt S. Why love matters: How affection shapes a baby’s brain, Routledge. 2014.

[11] Corrigall J, Wilkinson H. . Revolutionary connections: Psychotherapy and neuroscience, Karnac Books . 2003.

[12] Ryle A. The structure and development of borderline personality disorder: a proposed model. The British Journal of Psychiatry 1997 170, 82-87.

[13] Bennett D, Parry G, Ryle A. Resolving threats to the therapeutic alliance in cognitive analytic therapy of borderline personality disorder: a task analysis. Psychology and psychotherapy . 2006, 79, 395-418.

[14] Grant J, Crawley J. Transference and projection: Mirrors to the self, McGraw-Hill Education 2002.

[15] Kellett S. Is narrative reformulation during cognitive analytic therapy helpful? [Online]. 2015 Available: [Accessed 30th May 2018].



De Graaf LE, Gerhards SA, Arntz A, Riper H, Metsemakers JFM, Evers SMAA et al. Clinical effectiveness of online computerised cognitive-behavioural therapy without support for depression in primary care: randomised trial. The British journal of psychiatry : the journal of mental science. 2009, 195, 73-80.
Kellett S. . Is narrative reformulation during cognitive analytic therapy helpful? 2015 [Online]. [Accessed].
CATCH innovation grant success 2016. [Online]. Available: [Accessed 30th May 2018].
McCormick EW. Change for the Better. 2017. London, Sage Publishing.
Meadows J, Kellett S. Development and Evaluation of Cognitive Analytic Guided Self-Help (CAT-SH) for Use in IAPT Services. Behavioural and Cognitive Psychotherapy. 2017. 45, 266-284.


About the authors

Annabel McDonald is a Consultant Psychiatrist at Kent and Medway NHS and Social Care Partnership Trust.

Sanaz Fallahkhair is a Senior Lecturer at the School of Computing, Engineering and Mathematics (CEM), University of Brighton as well as a member of “Centre for Secure, Intelligent and Usable systems”, and “Centre for Digital Media Cultures”.

Jessica Eccles is a Clinical Lecturer in Psychiatry at the National Institute for Health Research as well as a fellow of the MQ: Transforming Mental Health Through Research and Versus Arthritis organisations.