The self and schizophrenia: some open issues

The construct of schizophrenia, as depicted by the DSM-IV and the proposed DSM-5 diagnostic criteria, lacks a distinctive and typical clinical core. A variety of symptoms and signs are listed, but what links these disparate clinical aspects together remains unclear. We wondered elsewhere 1 whether there is actually a Gestalt in the schizophrenic syndrome, that the operational approach fails to grasp, or whether the Gestalt presupposed by the psychiatric tradition was simply an illusion, that the operational approach unveils.

Josef Parnas’ piece appearing in this issue of the journal 2 articulates the view that the psychopathological core of schizophrenia, which confers a Gestalt on this syndrome, is an alteration of the basic, prereflective sense of self. This “basic tone of selfhood”, “granted in the brain by a continuous source of internally generated input” 3, normally accounts for the subjective experience of agency, coherence, unity, temporal identity and demarcation 4, and is accompanied by a prereflective sense of immersion in the world 2,5. Its “trait alteration” generates the various clinical manifestations of schizophrenia. The current distinction between positive, negative and disorganization symptoms may appear in this light superficial and misleading: for instance, Schneider’s first rank symptoms, usually regarded as “positive” (i.e., involving the presence of experiences which are normally absent), may need to be reconceptualized as reflecting “the absence of something normally present – the sense of ownership or intentional control” 5.

This view, grounded in the phenomenological tradition 6,7, but consistent with classical descriptions 8,9,10 and several psychoanalytic conceptualizations (e.g., 11,12) of schizophrenia, crops out in the ICD-10 definition of the disorder (“The disturbance involves the most basic functions that give the normal person a feeling of individuality, uniqueness, and self-direction”), as well as in the texts of the DSM-III and DSM-III-R (“The sense of self that gives the normal person a feeling of individuality, uniqueness, and self-direction is frequently disturbed in schizophrenia”), while there is no trace of it in either the DSM-IV or the DSM-5 proposal.

The model, although appealing, raises several conceptual issues that might be the focus of further reflection and research. These issues are primarily related to the multiple notions of self – as many as twenty-one according to Strawson 13 – being used in literature.

First, an antinomy seems to emerge (e.g., 14) between the “trait” self-disorder described by Parnas, developing well before the onset of psychosis 2,4, and the disturbed sense of self portrayed by some psychosocial approaches (e.g., 15,16), which “results from” the illness, with which the person struggles actively, and from which it is possible to recover 17. It seems obvious that the prereflective level of self-experience delineated by Parnas is more basic than the level of “self as a narrative construction” 18 referred to in the above psychosocial literature. However, it is a fact that Parnas’ approach mainly focuses on what happens before the onset of schizophrenia, while those psychosocial approaches lay an emphasis on what happens after. Follow-up studies are needed to explore the course over time of the disorder of self-experience described by Parnas (in particular, to verify its persistence when psychosis remits).

Second, whether the concept of self endorsed by the phenomenological tradition – “autonomous, free and in control” 19 – can be generalized outside Western cultural contexts is questionable. It has been argued that the development of the individual self is unavoidably influenced by cultural meaning systems (the “collective self”) and that in other societies the pattern of self-disorder in schizophrenia may be different from that described in Western cultures 19. Cross-cultural studies focusing on this specific issue are currently lacking.

Third, the concept of a stable core self has been questioned by dialogical psychology (e.g., 20), according to which self-experience emerges from the dialogue of several “self-facets”, and impairment of that experience in schizophrenia may result from a difficulty to sustain this dialogue in interpersonal situations (e.g., 21). Again, different levels of self-experience are likely to be involved here, with the “dialogical self” being more akin to the “narrative self” mentioned above – an “open-ended construction, which is under constant revision” 18 – than to the basic sense of selfhood described by Parnas. Nevertheless, the interpersonal dynamics postulated by the dialogical approach – with intimate interpersonal contact further destructuring a vulnerable self (e.g., 22) – may be also relevant to the level of self-experience referred to by Parnas (e.g., 23).

In addition to these conceptual issues, Parnas’ model raises some practical concerns, which again might be addressed by further research.

First, while proposing and trying to validate a core Gestalt of schizophrenia, we should be aware of the risk to revive an “atmospheric” diagnosis of the disorder, possibly very reliable in the hands of super-experts, but dangerously volatile in ordinary clinical practice. This concern becomes even deeper if the approach is going to be adopted in the very sensitive area of early diagnosis of psychosis. Instruments for a systematic assessment of disorders of self-experience have been developed, which have shown a satisfactory interrater reliability in research settings (e.g., 24,25). However, the feasibility and reliability of these assessments when transferred to ordinary clinical conditions remain to be explored (as well as the way clinicians’ psychopathological competence can be upgraded in order to match this challenge).

Second, the diagnostic specificity for schizophrenia of the described self-disorder requires empirical support. Schneider’s first rank symptoms, which are typical symptoms evolving from disordered self-experience, have been widely reported to occur also outside the schizophrenic spectrum (e.g., 26). Indeed, one of the proposed changes in the DSM-5 diagnostic criteria for schizophrenia is the reduction of the emphasis on these symptoms, since “no unique diagnostic specificity for these characteristic symptoms in comparison to others has been identified” (www.dsm5.org). Parnas et al 27 were able to document in a research setting that anomalies of self-awareness discriminated significantly between patients with schizophrenia and psychotic bipolar illness, but this finding requires replication, and its generalizability to ordinary clinical contexts needs to be tested.

The third, and most significant, concern is that regarding therapeutic interventions. Is the postulated core self-disorder amenable to any of the currently available treatments? Are basic disturbances of self-experience (as opposed to structured delusions) the real target of antipsychotic treatment, or are we talking about an essentially unmodifiable constitutional deficit 14? Can cognitive-behavioural techniques be updated in the light of the self-disorder model, or is the level of their action not sufficiently “deep” to impact on the postulated psychopathological core? Should other psychotherapeutic approaches, including psychodynamic ones (e.g., 12,28), be developed or revived? Is the disorder of self-experience related to neurocognitive deficits, and is there any role for cognitive remediation? All this needs to be explored at the research level. Otherwise, there may be the risk to foster a new therapeutic pessimism, just at a time when an orientation toward recovery is being advocated for mental health services.

I believe the impact of Parnas’ intriguing approach will crucially depend upon the extent to which the above issues will be convincingly addressed.