The intrapersonal civil war

Rebecca Johnson asks what control theory can contribute to our understanding of dissociative identity disorder
When a person presents with dissociative identity disorder, how can we best understand their condition? Is it a form of internal conflict? Is it the product of a troubled childhood and an internal battle between a traumatised, fragmented personality? Recognising, understanding and resolving these questions are imperative if we are to help people with this condition manage their lives better. Perceptual control theory is a self-regulatory framework that could offer a way forward…

Few provisions are necessary for basic survival – food, water and warmth are sufficient. However, a child who receives nothing but this is unlikely to thrive. Survival beyond the realms of mere existence demands the presence of a caregiver, and it is theorised that childhood survival instinct dictates that they must love and want to be loved by their parents (Putnam et al., 1986). So what happens when the parent is absent, or even abusive?

Abuse during childhood has been cited as a significant factor in the aetiology of dissociative identity disorder (DID) (Putnam et al., 1986), a severe form of dissociative pathology whereby the patient displays at least two distinct and enduring identities/personality states that recurrently ‘take control’ of the person’s behaviour (see box opposite). Although the figures should be treated with caution because of the questionable reliability of retrospective self-report data from this population (Kihlstrom, 2005), studies have reported that over 80 per cent of DID patients report physical or sexual abuse during childhood (Middleton & Butler, 1998).

Since the abuse is most commonly inflicted by a caregiver, the correlation between DID and child abuse is perhaps best understood in terms of conflict. Bowlby states that attachment behaviour is just one of four behaviours that are pre-programmed in the infant, each of which ‘contributes in its own specific way to survival’ (Bowlby, 1988, p.5). If the infant’s carer is subsequently abusive, this may cause an irresolvable conflict between the ‘pre-programmed’ survival behaviour to ‘attach’ and the will to escape trauma and abuse.

This article does not intend to provide a comprehensive review or comparison of the many theories and scepticisms of DID. Instead, I suggest how DID could be understood and treated within the framework of a comprehensive theory of human behaviour known as perceptual control theory (PCT: Powers, 1973).

Perceptual control theory
PCT is related to an early approach to understanding human behaviour called cybernetics. This regards people as essentially intricate control mechanisms who behave in a manner that allows them to maintain numerous intrinsic and/or essential variables within survivable limits (Ashby, 1952). On a basic level the manner in which humans adjust their behaviour to exert control can be compared to a thermostat that has a ‘goal state’ of achieving a room temperature of 21°C. If the temperature drops below 21°C, the thermostat adjusts its ‘behaviour’ to reduce the ‘error’ and increase the temperature, and vice versa if the temperature goes above 21°C. Essentially, both the thermostat and the person are similar in that they use behaviours as vehicles to control their predetermined perceptions or goals.

PCT itself is based on the notion of a ‘hierarchy’ of control systems, each one being given its ‘goal’ or ‘reference value’ by the control system above it. Systems throughout the hierarchy work in the same manner, by comparing the actual perception with the desired perception and then producing a behaviour to reduce the discrepancy between the two. For example, a driver will compare their actual position on the road with the desired position on the road and then produce a behaviour to reduce the ‘error’ (e.g. slightly adjusting the steering wheel). Systems further up the hierarchy control progressively more complex and abstract perceptions about oneself (e.g. ‘to be a good person’). High-order goals such as this have no direct correlative behaviour – there is no single behaviour that makes someone ‘good’ – and consequently, the goal is achieved by setting ‘sub-goals’ or ‘reference values’ for systems lower in the hierarchy. This ‘top-down’ approach to control eventually results in actual observable behaviour that aids a person in moving towards their ‘reference value’ and counteracting any environmental factor that moves them away from it.

Internal conflict in PCT and DID
The aspect of PCT that is perhaps most relevant in the discussion of DID, is within its conceptualisation of conflict. The complexity of the proposed ‘hierarchy of control’ predominantly allows numerous goals to be achieved simultaneously (Marken, 1990). Minor conflicts between control systems are a common occurrence, such as whether to splurge or save, indulge or be healthy. PCT states that minor conflicts such as this are quickly resolved through a process of ‘reorganisation’ and are unlikely to cause psychological distress. Reorganisation involves functionally reworking connections within the hierarchy; a random process that continues until error is reduced (Powers, 1989).

Conflict resolution becomes especially important – in the discussion of psychological distress and DID – when two high-order systems, positioned equally within the hierarchy, strive for incompatible goals. Moving towards one systems goal causes error in the other. In the case of the abused child, the perceptual goal of ‘being loved by her parents’ and ‘escaping trauma’ can be seen as hierarchically equal as both are essential for survival. The acute conflict that arises from the abuse is irresolvable; the child cannot control the abuse, escape the trauma or detach from the caregiver; so a physiological defence mechanism is employed. This allows her to ‘shift’ into an alternative state of consciousness and she dissociates (Emde et al., 1976). PCT states that when reorganisation of the system occurs but error persists, the hierarchy allows a ‘switch to alternative means of control’ (Powers, 2005). In the case of DID each control system is allowed to pursue its own goal for separate ‘alters’. Within PCT the compartmentalised alters would be managed by functionally distinct hierarchical control systems.

The kind of prolonged conflict that characterises DID can be reduced through reorganisation in a variety of ways, both temporarily and more permanently. For example, a short-term reorganisation may involve suppressing one side of the conflict through a new method, such as distraction. Another pertinent example is where one goal is perceived only in imagination, thereby short-circuiting the lower levels of the hierarchy that interact with the world. This retreat into fantasy is another example of how individuals with DID can manage childhood trauma (Austrian, 2000). By imagining that the abuse is happening to someone else, the child can pursue their need to love the caregiver without the conflict of simultaneously hating them. Parallel to this, the child who is being abused can hate the caregiver and attempt to escape the trauma without needing to simultaneously love them. This is demonstrated by this testimony of a DID sufferer: ‘It was as if there were two of me – the one that was abused and another one that it had not happened to’ (Mollon, 1996, p.8).

Up to 86 per cent of DID patients express a child alter (Ross et al., 1990), who is often frightened and traumatised, perhaps because it is ‘them’ who have suffered the abuse. It has been reported that high numbers of DID patients suffer amnesia and can often be unaware of their abusive past until it is revealed –via a child alter – in therapy (Ross et al., 1989, 1990). However, claims of amnesia in DID patients may be questionable, as it is extremely rare to find cases of true amnesia that cannot be accounted for by organic factors, with the exception  of normal forgetting or infantile amnesia (Kihlstrom, 2005).

Aside from the ‘child’ alter, it has been suggested that the average DID patient expresses between 6.3 and 15.7 separate alters (North et al., 1993); although there have been reports of patients with very large numbers of alters who interact in a highly complex organisational ‘hierarchy’. Mollon (1996) has suggested that the complexity and size of the ‘system’ may be linearly related to the severity of abuse. Certain studies have disputed the reliability of the above data (e.g. Kihlstrom, 2005). However, for the sake of this article – a consideration of how PCT can offer suggestions to understand and treat DID – it is perhaps more important to focus on the actual presenting symptoms and testimonies of DID patients.

The notion of alters working within a hierarchy is highly significant within the discussion of PCT, which states that all behaviour is the result of the internal hierarchy of control. The defence mechanism that is said to allow the infant to dissociate is developmentally significant as it coincides with the natural development of the internal hierarchy of control. During infancy the child’s hierarchy develops due to the process of reorganisation, whereby she learns new solutions to conflict that were not previously possible (Vanderijt & Plooij, 2003). Higher-order systems develop for the purpose of regulating and balancing lower-order systems in the child. Extreme lower-order conflict at an early age may interrupt the normal maturation process of hierarchical control over pertinent perceptions, such as those involving eliciting signals of care and security from significant others. This makes the child less able to comfortably experience and solve conflicts in later life, as their repertoire of resolutions is underdeveloped. This is coherent with the clinical observation that individuals with DID cannot experience conflict – ‘contradictory feelings or perspectives are not typically contained within a single alter. Each alter represents
a particular feeling or perspective’ (Burton & Lane, 2001, p.303).

The concept of each emotion or perspective requiring its own individual ‘being’ may also help explain why more severe abuse correlates with a bigger and more complex system of alters. Particularly severe abuse may make the feeling of conflicting emotions especially unbearable for the DID patient, so they fragment further and further. Whilst a ‘normal’ person can comfortably feel ‘excited but worried’ or ‘sad but relieved’, the DID patient cannot manage such conflicting emotions. Evidence of the alters’ and the hosts’ inability to tolerate internal conflict has been provided by observations that they have ‘great difficulty tolerating any ambivalence’ (Burton & Lane, 2001).
The DID patients’ difficulty in tolerating ambivalence perhaps explains why alters predominantly convey one single – often extreme – personality trait, for example, the terrified ‘child’, the rage-fuelled ‘perpetrator’ (Ross, 1997) or the highly promiscuous woman (in the case of ‘Eve Black’: Thigpen & Cleckley, 1957). 

Recovery, treatment and effective reorganisation

PCT states that conflict resolution requires the intervention of a higher-order control system within the hierarchy. This is due to the ‘top-down’ nature of control: higher systems set the ‘reference value’ or ‘goals’ for the systems below. Therefore, in order to reduce conflict between two lower-order systems, the systems above them must be involved. Simply working with the lower-order systems will not produce lasting results because their ‘reference values’ (set by the level above) will still be conflicting (Powers, 1973, 2005).

Therapeutic intervention, based on the principles of PCT, would involve a series of stages. Firstly, the person’s awareness of their internal conflict would be increased (Powers, 1998). Following this, the therapist helps to shift the client’s awareness to a higher level within the hierarchy. In the DID patient each alter
can be seen as analogous to a control system within the hierarchy. For this reason therapeutic intervention must be aimed above this ‘level’ of internal conflict.
From a therapeutic viewpoint the therapist should facilitate communication between alters (i.e. allowing systems within the hierarchy to acknowledge one another and increase awareness) in order to address their conflicting interests – for example, the child alter may want to reach out to the therapist for attention whilst the ‘protector’ blocks the ‘child’ in order to prevent her being re-traumatised (Mollon, 1996).

PCT states that all psychological distress is simply a different expression of the same underlying process – conflict (Powers, 1973). If all psychological distress results from the same underlying cause, then it would follow that it can all be ‘treated’ in the same manner. ‘Method of levels’ (MOL) is a non-directive psychotherapeutic approach to conflict resolution (Carey, 2006). It is currently used by a small number of psychologists in the UK, Australia and the US, and has recently been considered as a form of cognitive therapy (Carey, 2008).

There is evidence to support MOL in the treatment of a wide range of presenting problems (see Carey, 2008). However, its emphasis on conflict resolution is perhaps especially significant and valuable in the treatment of DID. PCT states that conflict resolution first requires awareness of the conflict. This awareness must then be shifted to higher levels within the hierarchy in order for it to be resolved. MOL capitalises on the ‘disruptions’ that occur in everyday speech; for example, a pause, a look away, a shift in posture or an evaluative comment: ‘That sounds ridiculous’ (Carey, 2008). These disruptions are said  to represent a shift in awareness to a level higher in the hierarchy. The MOL therapist facilitates this ‘shift’ by orientating the patient to consider why they just paused or to think about what just went through their mind (Carey, 2006). By orientating the client to ‘talk about what they are talking about’, or ‘think about their thoughts’, the MOL therapist acts as a vehicle to shift the patients awareness to a higher level; crucial for the possibility of conflict resolution (Carey, 2006).

MOL is a client-centred approach that helps individuals to begin to describe the processes of control, conflict and reorganisation they are experiencing. A client with DID would therefore be encouraged to talk about whatever current problem was on their mind, and the questioning style would help them to articulate the conflicting goals in relation to this problem. It is very likely that DID patients will present their alters within the session, in which case the therapist would allow the client to focus on the shifts in awareness that occur and allow the client to develop a higher-level system that observes and manages the shifts in alters that occur in relation to the higher-order goals that the client may present – such as ‘living an ordinary life’ or ‘managing relationships’. This proposal is clearly hypothetical at this stage, until MOL  is explored with this population.

Regaining interpersonal control
The controversy which surrounds the diagnosis of DID cannot be ignored; despite decades of research there is still no consensus as to cause, treatment or even whether it really ‘exists’. However, the factor that is perhaps more difficult to dispute is the disabling impact of the symptoms of DID. Whether memories and alters recovered in therapy are real or manufactured should perhaps not be the therapist’s main priority. If a patient is presenting with dissociative symptoms then it could be argued that they still require treatment regardless of whether their memories are based on actual past events – their perception of their experiences is most important in terms  of their levels of distress and functioning.

The principles of PCT may help explain the development of DID, providing suggestions about how dissociation may occur and why alters may develop. Furthermore, the client-centred focus of MOL allows the therapist to focus on the current and presenting problem, which lends itself to treating the numerous alters as and when they ‘appear’ during therapy.

Considering the large number of perceptual experiences that fall under separate control systems, it may seem surprising that more people do not develop DID. Modern lifestyles encourage performing numerous roles simultaneously (e.g. mother, wife and employee) and this ability to portray multiple self-states may be functional, if regulated in a flexible and controlled way. By portraying different personalities as a child, the DID patients effectively protect themselves from harm. However, this becomes problematic when it no longer suits the reality of the adult world. It is therefore the job of the therapist to allow the patient to regain intrapersonal control and effectively live within an adult community.

It is clear that further research and understanding is required if we are to fully understand the underlying causes, symptomology and most appropriate therapeutic approach to DID. PCT offers some interesting suggestions about how and why alters are developed, as well as some principles upon which a therapeutic intervention could be implemented, in the form of MOL. Further research should perhaps focus on the responses to MOL by DID patients themselves, assessing whether PCT’s theory of raising awareness of conflict and working towards resolution and integration is effective in this population.

Rebecca Johnson is with the Department of Psychological Medicine at Imperial College London
[email protected]

Box: Dissociative Identity Disorder

Dissociative identity disorder
Dissociative identity disorder (DID), or multiple personality disorder as it was once known, is described as the most severe form of dissociative pathology (Kluft et al., 1988), whereby the patient displays at least two distinct and enduring identities/personality states that recurrently ‘take control’ of the person’s behaviour. These disturbances must not be accountable for by the use of substances (e.g. blackouts or erratic behaviour due to alcohol) and in the case of children, must go beyond the realms of imaginary playmates (DSM-IV criteria, APA, 1994).
To the observer, the DID patient seems to have several ‘people’ living within one body, each with different personalities, memories, life histories, ages and even gender. Occasionally, these ‘alters’ have no knowledge of one another despite their close proximity, with each living under the illusion that they are the sole personality. However, more commonly, each alter is aware of the other, although some have a more complex understanding than others (Mollon, 1996).
The validity of DID as a psychological disorder is disputed, with some stating that it is more a product of ‘enthusiastic clinicians’ (Piper, 1995) and ambiguous diagnostic criteria than a separate disorder (Kilhstrom, 1995). Furthermore, clinicians who do subscribe to the validity of DID do not necessarily agree on the aetiology, diagnostic criteria or the most effective course of treatment.


American Psychiatric Association (1994). Diagnostic and statistical manual of mental disorders (4th edn). Author: Washington, DC.Ashby, W.R. (1952). Design for a brain. New York: Wiley
Austrian, S.G. (2000). Mental disorders, medications, and clinical social work. New York: Columbia University Press.
Bowlby, J. (1988). A secure base: Clinical applications of attachment theory. London: Routledge.
Burton, N. & Lane, R.C. (2001). The relational treatment of dissociative identity disorder. Clinical Psychology Review, 21(2), 301–320.
Carey, T.A. (2006). The method of levels: How to do psychotherapy without getting in the way. Hayward, CA: Living Control Systems.
Carey, T.A. (2008). Perceptual control theory and method of levels:
Further contributions to a transdiagnostic perspective. International Journal of Cognitive Theory, 1(3), 237–255.
Emde, R.N., Gaensbauer, T.J. & Harmon, R.J. (1976). Emotional expression in infancy: A biobehavioral study (Psychological Issues Vol. 10, No. 1; Monograph 37). New York: International Universities Press.
Kihlstrom, J.F. (2005). Dissociative disorders. Annual Review of Clinical Psychology, 1, 227–253.
Kluft. R.P., Steinberg, M. & Spitzer, R.L. (1988). DSM-III revisions in the dissociative disorders: An
exploration of the derivation and rationale. Dissociation, 1, 39–46.
Marken, R.S. (1990). A science of purpose. American Behavioral Scientist, 34(6), 6–13.
Middleton, W. & Butler, J. (1998). Dissociative identity disorder: An Australian series. Australian and New Zealand Journal of Psychiatry, 32(6), 794–804.
Mollon, P. (1996). Multiple selves,
multiple voices: Working with trauma, violation and dissociation. Chichester: Wiley.
North, C.S., Ryall, J.E.M., Ricci, D.A. & Wetzel, R.D. (1993). Multiple personalities, multiple disorders, psychiatric classification and media influence. New York: Oxford University Press.
Piper A. (1995). A skeptical look at multiple personality disorder. In L.M. Cohen, J.N. Berzoff & M.R. Elin (Eds.) Dissociative identity disorder: Theoretical and treatment controversies (pp.135–173). New York: Jason Aronson.
Powers, W.T. (1973). Behavior: The control of perception. New York: Wiley.
Powers, W.T. (1989). Random-walk chemotaxis: Trial and error as a control process. Behavioural Neuroscience, 103(6), 1348–1355.
Powers, W.T. (1998). Making sense of behavior: The meaning of control. New Canaan, CT: Benchmark.
Powers, W.T. (2005). Behavior: The control of perception (2nd edn). New Canaan, CT: Benchmark.
Putnam, F.W., Guroff, J.J., Silberman, E.K. et al. (1986). The clinical phenomenology of multiple personality disorder: Review of 100 recent cases. Journal of Clinical Psychiatry, 47, 285–293.
Ross, C.A., Herber, S., Norton, G.R. & Anderson, G. (1989). Differences between multiple personality disorder and other diagnostic groups on structured interview. Journal of Nervous Mental Disease, 177, 487–491.Ross, C.A., Miller, S.D., Reagor, P. et
al. (1990). The dissociative
disorders interview schedule: A structured interview. Dissociation, 2, 169–189.
Ross, C.A. (1997). Dissociative identity disorder. New York: Wiley.
Thigpen, C.H. & Cleckley, H.M. (1957). The three faces of Eve. London: Secker and Warburg.
Vanderijt, H. & Plooij, F. (2003). The wonder weeks. Living Control Systems Publishing.

BPS Members can discuss this article

Already a member? Or Create an account

Not a member? Find out about becoming a member or subscriber