top of page
  • Facebook
  • Twitter
  • YouTube
  • Facebook
  • Twitter
  • YouTube
  • Facebook
  • Twitter
  • YouTube

How assisting client self-observation has changed in CAT

Cite as: Leiman, M. (2019). How assisting client self-observation has changed in CAT. International Journal of Cognitive Analytic Therapy and Relational Mental Health, 3(1), 28–34.

Int. Journal of CAT & RMH Vol.3 No.1 December 2019/ISSN2059-9919 


Assisting client self-observation is the fundamental task of therapists in any model of psychotherapy. In what follows, I will summarise my understanding on how this task is conceptualised in CAT at various stages of its development. The key issue is, should we understand self-observation as an ability, or is it a jointly generated activity that is, primarily, dependent on the therapists’ ways of timing and addressing the focal themes of therapy?

1. Early understanding

Looking back, in the pre-CAT period of traps, dilemmas and snags, client self-observation seemed unproblematic. It was assumed that clients recognised their problematic procedures when provided with portable and accurate tools for self-reflection.


This straightforward assumption worked quite well. A lot was accomplished only in 10-12 sessions. The reason for the success was in the accuracy of reformulation, which required that it was accessible to the client.


Traps described self-fulfilling anticipation. By monitoring their traps, clients quickly recognised the assumptions by which they predicted the consequences of their actions. Generalised assumptions were described as either-or or if-then dilemmas. The cleverness of dilemma formulations is hidden in the negative pole. ‘Either I keep feelings bottled up or make a mess’ describes the preferred, protective alternative and the implied risk, if the precaution fails. ‘If then’ dilemmas, in turn, describe risky situations that will lead to an intolerable position, if the risk materialises: ‘If close, then smothered’.


Dilemma formulations were effective, because they targeted the outer border of the client’s conscious self observation and because their wording was emblematic. ‘Smothered’ contains a collection of referential routes, which allow clients to explore further the undesirable or feared alternative.


The elegance of these brief formulations lies in their dynamic power as tools that expand the client’s zone of proximal development. Their use agreed with the classical psychoanalytic view on starting the therapeutic work from defences. Traps, dilemmas, and snags nicely captured the rich variety of ways by which we try to manage anxiety, helplessness, and forbidden wishes.

2. Development of complex reformulations

The way of assisting client self observation changed quite radically by the end of the 1980s, when the simple, portable tools were replaced by reformulation letters and diagrams. Letters were mainly a way of helping the as yet inexperienced psychotherapists to organise the client’s complex clinical material in some coherent way and help them infer the salient trap, dilemma, and snag formulations, or the TPPs on the basis of this material. True, a lot of clinical experience is required to identify the client’s recurring action patterns in a free dialogue during the first session and to generate accurate mini-formulations that begin to guide client self-observation.


Diagrams were inspired by Mardi Horowitz’s configuration analysis that, originally, was mainly used for research purposes. Tony Ryle’s innovative idea was to extend their use to help dissociative clients perceive and remember their disruptive ways of dealing with helplessness or difficult interpersonal situations. The early visualisations resembled the SSSDs, or Self State Sequential Diagrams, that evolved in the mid-1990s. They named alternating emotional states and identified risky situations that provoked state shifts.


Such visualisations began to grow more complex, and by the turn of the decade some of them resembled roadmaps of Birmingham as Tony Ryle noted. At that stage, CAT theory stepped in to discipline the anarchistic variety of maps. Reciprocal roles were summarised in the core of the diagram and the client’s various ways of dealing with unmanageable roles were depicted by loops around the core. While the description organised the maps, the dissociated states and shifts between them were lost in this diagrammatic format. After some years, this shortcoming was amended by the SSSDs.


As tools for self observation, diagrams were difficult, because they suffered from the same problem that had stimulated the development of reformulation letters. Instead of simple and portable tools that evolve in the course of joint reflection of client problems, writing a letter or constructing a diagram during the first sessions became an arduous task for the therapist and their sharing with the clients a ceremonial process.


A lot of clinical experience is required to stay within the limits of what the client can access. During my years of CAT trainer and supervisor, I rarely saw diagrams to which I did not have any reservations. Their main advantage was that they demonstrated the therapist’s developmental level on understanding reciprocal roles and their relationship to coping procedures. I was often concerned about how the therapist’s construction might look to the client and whether it actually helped in focusing the therapy on salient issues.

4. Self-observation as an ability

In current psychotherapies of personality disorders, Freud’s dynamic understanding has tacitly been replaced by the idea of self-observation as a developing capacity. This is true of the more cognitively oriented therapies like dialectical-behavioural therapy, emotion-focused therapy, and schema-focused therapy. However, it is also shared by the more psychodynamically oriented models, e.g., mentalization-based therapy and CAT.


History tends to repeat itself here. Dissociation posed a puzzling problem for Janet and Freud, who wanted to develop psychological treatments for mentally ill patients. Their ways departed early, as Janet continued to use hypnosis as one of his treatment techniques. It was an effective method to transcend the blocs in self-observation, but one problem was that hypnotic trance could not be induced for all patients. Free association became Freud’s technique of approaching dissociated mental contents, which had an additional advantage. It showed that shifting mental states are object-dependent.


A hundred years after Janet and Freud, dissociation was again tackled by the above-mentioned models. The problem was identical, but the explanatory principles were now obtained from attachment theory and cognitive script or schema models. Self-observation was understood as a mental capacity that could be impaired by maladaptive early development and relationship schemas. This view is strongly endorsed in all current psychotherapeutic approaches that attempt to treat borderline, personality disordered clients.


Harri Valkonen (2018) examined the three initial assessment interviews of three clients with BPD diagnosis. The study yielded two main results. First, the clients’ self-observation was strongly object- dependent. Their ability to reflect on their memories, experiences, and actions was fully appropriate in ‘conflict free’ domains. They could even describe their state shifts and dissociated self-states, but the problem was that their self-observation was partly state dependent. Dissociation was a problematic way of coping with dangerous memories and unmanageable experiences. It blocked further exploration during the interview and it also seemed to wipe away, before the next session, what had been jointly achieved.


Secondly, the quality of clients’ self-observation did not only depend on the objects of observation. The psychotherapist’s way of conducting the interview and addressing sensitive issues had a decisive impact on the clients’ responses. Reading the transcribed excerpts from the sessions is quite startling. The power of what we address, or do not address is much bigger than we tend to think.


The therapeutic relationship is indeed indispensable for self- observation. Clients may gradually access unmanageable themes when trust and safety develop in the relationship. This happens when they have repeated experiences of a containing and non-collusive therapist who is not afraid of what may emerge and who does not judge. The microanalysis of the assessment interviews showed how difficult it is to implement this general principle in practice.

5. It is all about timing

The development of self-understanding is a dynamic process, depending on the client’s growing trust in the therapist’s capacity to tolerate the client’s emotionally unmanageable experiences. Trust helps in accessing and expressing mental contents that the client has not been able to manage alone.


Fredrik Falkenström’s account of Ms B in his paper on the levels of self-observation (Falkenström, 2007) illustrates the joint power of trusting and the right moment for the development of client insight. Ms B had a history of drug addiction but had been clear for six years. She sought help because renewed craving for drugs and fears of relapsing. She had had both institutional treatment and psychotherapy, which had helped her to get off drugs. In her third therapy session, Ms B spoke in an agitated manner and switched between topics as if being afraid of touching anything in more detail. She did not leave any room for the therapist’s comments. Something that troubled her was becoming closer to conscious perception.


At one point the therapist said: ‘Wait a minute – could you, just for a moment, try to sense – how do you feel, right now, sitting in this chair?’ (Falkenström, 2004, p. 569). Focusing on the here-and-now was quite a bold intervention, because the therapist could not know what might appear.


Ms B stared at the therapist, settled back in her chair, and tears welled up in her eyes. She said, ‘I feel empty’, looking fearful. She added a metaphoric description, as if having a black hole inside.


In terms of classical CAT dilemma formulations, Ms B had only two options. ‘Either keep on running or falling into a black hole’. She showed the defensive alternative, but the therapist’s direct request to stop helped her get hold on the feared alternative.


Falkenström describes in the paper how this momentary insight into her agitated emotional state opened a path to an understanding of the role of emptiness as a protective device against painful feelings of guilt that she had unconsciously assumed to be uncontainable by anybody.


To me, Falkenström’s account illustrates how many layers may exist in our internal world. What at first appears as the feared problem, which the client attempts to manage by hypomanic action, turns out to be a protective response to even more painful and dangerous feelings.


Initially, we only see some fragments of the ways by which clients try to cope with their lives. Their protective actions form hierarchical chains. Each layer is well guarded like Ms B’s hypomanic attempts to avoid getting in touch with the sense of emptiness. But although feared, feeling empty was not the bottom layer. It had developed early as a way of dealing with overwhelming guilt.


The therapist’s intervention struck the right chord at the right moment. One of the most difficult tasks in any therapy is to decide, when the time is ripe to help clients confront something they fear, are ashamed of, or feel guilty about. We must perturb the client, if we want to get over the stalemate that staying within the safety area tends to generate. However, we should not intervene prematurely, because it only forces the client to react, defensively, and our good intention has backfired.


Assisting client self-observation is not, primarily, an issue of what kind of tools we should use. Mini formulations, mapping, diagrams, and reformulation letters present quite an impressive array of techniques that CAT therapists have generated over the years. None of them can tell the therapist the moment at which they afford clients to approach and address experiences that they have warded off for different reasons. A reformulation letter may act as a premature interpretation, if it contains themes correctly anticipated by the therapist but not yet accessible to the client. Maps may clarify repetitive action patterns, but they too may contain too much information from the client’s viewpoint.


Every tool is permissible, when the time is ripe. Kairos, or the opportune time, is the most important tool for any psychotherapist.


Bion, W. R. (1954). Notes on the Theory of Schizophrenia.

International Journal of Psycho-Analysis, 35: 113-118.

Falkenström, F. (2007). The psychodynamics of self-observation.

Psychoanalytic Dialogues,  551-574.

Freud, S. (1894). The neuro-psychoses of defence. Standard Edition, 3: 41-61.

Freud, S. (1914). Remembering, Repeating and Working Through (Further Recommendations in the Technique of Psychoanalysis II) Standard Edition, 12:145 -156.

Valkonen, H. (2018). The dynamics of self-observation in patients with borderline personality disorder (BPD) diagnosis. Dissertations in Education, Humanities, and Theology No 125. Publications of the University of Eastern Finland, Joensuu.

3. The dynamic space

Freud’s early experience with hysterical patients showed him the presence of powerful forces that affected patient self-observation. Dissociation banished the memories of traumatic experiences from consciousness either completely, or allowed pale, affectless memory traces to enter as objects of self-observation (Freud, 1894). He found a way of approaching such objects indirectly by free association. The patient was instructed to let her mind wander freely, observing what was going on, and describing it to the analyst. The self-observing part of this practice is almost identical with current mindfulness methods. The difference is in the request to tell about what is going on.

Freud soon found out that the presence of the analyst introduced distractions in the flow of tracing the semantic networks of internal objects. At times, the patient’s associations stopped completely. This phenomenon led Freud to postulate transference, which he, initially, understood as a hinder to the analytic process. Only later he realized that patients show, in the interaction, what they cannot yet make a target of deliberate self-observation. Thus, transference became an important object of joint observation that helped patients address disavowed aspects of themselves and dissociated past experiences (Freud, 1914).


The dynamic nature of self-observation became very clear to Wilfred Bion, when he tried to work psychoanalytically with fragile and nearly psychotic clients. The general contemporary belief was that helping clients access the unconscious by accurate interpretations was mutative. Bion (1954) noticed that, for psychotic clients, interpretations provoked fragmentation and confused mental states. This finding led him to develop the idea of containment. The noxious internal objects, or unmanageable memories of past experiences, which the client could not encounter without risks of deterioration, needed to stay with their full emotional force within the therapist’s consciousness for some time. This altered the power balance between the client and the ‘unthinkable object’ and it could become a target of joint observation. Eventually, clients were able to tolerate the emotional impact of the object and modify their way of relating to it.

bottom of page