Phenomenological models of delusions: concerns regarding the neglect of the role of emotional pain and intersubjectivity

Sass and Byrom's paper (1) describes one phenomenological and several neurocognitive models of delusions and then proposes a synthesis. The authors put forward a model in which alterations in self-experience interact with neurocognitive processes such that persons over-value information that should be ignored and withdraw “from practical, world-oriented activity in favor of self-referential processing”, ultimately arriving at convictions which may be profound barriers to wellness.

We support the integration of neuroscience and phenomenological models. These different approaches are rarely considered as complementary even though any comprehensive model of severe mental illness requires that both the subjective phenomena and neurocognitive processes involved be accounted for and reconciled. We also agree with the authors’ reflections that no single psychological, neurobiological or phenomenological model is likely to explain delusions. In this commentary, however, we will suggest that the authors’ synthesis neglects substantial bodies of knowledge about delusions, including their instability within the flow of daily life, temporal links between painful affects and the presence of delusions, and correlates with difficulties forming ideas about the thoughts and experiences of others. In what follows we will detail each of these points and suggest that ignoring this literature risks dehumanizing the dilemmas at the heart of delusional experience, something which could have deeply negative consequences for treatment.

To begin, we agree that understanding delusions primarily as miscalculations based on neurocognitive abnormalities does not match the evidence. We, however, see that the core problem with purely cognitivist explanations is that delusional beliefs are unstable and often emerge in particular intersubjective contexts. People with schizophrenia may be delusional during some but not other periods of the day, and may be delusional about certain issues but not others (2,3). The instability of delusions poses a problem for the cognitivist models, as well as for the phenomenological model offered in Sass and Byrom's paper. Why would delusional processes fluctuate so dramatically if they are a matter of the trait-like deficits proposed?

A related problem is that multiple studies suggest that fluctuations in delusions often follow alterations in emotional state. Threats to self-esteem and the emergence of clearly discernable forms of emotional pain have been found to predict the occurrence of positive symptoms in multiple studies (4,5). This would seem to suggest that delusions are not merely miscalculations based on neurocognitive deficits or the product of fundamental alterations in the sense of self, as the authors describe. Instead, if pain triggers delusional experience, it may be that delusions are in part attempts, albeit ineffective ones, by human beings to explain or communicate their pain to other human beings. This is also consistent with work from evolutionary psychiatry suggesting that paranoia may be a dysfunction of the basic threat detection system.

It has been suggested that persecutory delusions may emerge when persons do not feel that they belong within any group and so, in the absence of a sense of safeness and security, they feel constantly threatened and ultimately explain that in terms of imagined threats (6). Importantly though, delusions may at different times stem from very different motives. While delusions may express distress at some moments, at other times they may function as a means to ward off emotional connections with others. In other words, their oddness may be intentional and serve the purpose of remaining unknowable to prevent the emergence of pain (7).

Another literature that is not considered here concerns the difficulties many with psychosis have forming complex integrated ideas about the subjective experiences of other persons, sometimes referred to as theory of mind and metacognition (8,9). This would seem relevant in two senses. Persons unable to notice specific things about others or to appreciate the perspectives of other people might well adopt delusional stances as a fail-safe response to uncertainty. They would also seem less likely to be able to adjust their views based on what others think, as others’ views would be inaccessible.

Sass and Byrom do discuss a general withdrawal to a solipsistic state, but we suggest looking at something different: problems that occur when persons with psychosis fully try to focus or interact with others, something that seems essential for any coherent model of how delusions unfold as they do in the flow of daily life. Of note, alternative models not mentioned by Sass and Byrom do exist for understanding delusions and their place in the human condition. These include the work of Salvatore et al (10), who suggest that the experience of ontic threat interacts with real life experiences, cognitive biases and deficits in the ability to understand others to produce at least certain kinds of delusions.

In summary, Sass and Byrom's view seems to neglect well-established links of delusions with emotional pain and metacognition, and this carries several hazards. In our opinion, the risk is to cast delusions as exotic phenomena and therefore, mistakenly, as mental states which are incomprehensible rather than understandable forms of human experience. This may further result in positioning persons with delusions as beings removed from pain and the most central elements of human experience.

At the level of clinical practice, we see an even more significant danger. Sass and Byrom's model seems to us to risk relieving clinicians from the charge of understanding the humanity of their patients and could well result in clinicians standing at a distance from their patients and their suffering.