Sam Wilkinson

Accounting for the phenomenology and varieties of auditory verbal hallucination within a predictive processing framework.

Two challenges that face popular self-monitoring theories (SMTs) of auditory verbal hallucination (AVH) are that they cannot account for the auditory phenomenology of AVHs and that they cannot account for their variety. In this paper I show that both challenges can be met by adopting a predictive processing framework (PPF), and by viewing AVHs as arising from abnormalities in predictive processing. I show how, within the PPF, both the auditory phenomenology of AVHs, and three subtypes of AVH, can be accounted for.

The Representation of Agents in Auditory Verbal Hallucinations

Abstract Current models of auditory verbal hallucinations (AVHs) tend to focus on the mechanisms underlying their occurrence, but often fail to address the content of the auditory experience. In other words, they tend to ask why there are AVHs at all, instead of asking why, given that there are AVHs, they have the properties that they have. One such property, which has been largely overlooked and which we will focus on here, is why the voices are often experienced as coming from (or being the voices of) agents, and often specific, individualised agents. In this article, we argue not only that the representation of agents is important in accurately describing many cases of AVH, but also that deeper reflection on what is involved in the representation of agents has potentially vital consequences for our aetiological understanding of AVH, namely, for understanding how and why AVHs come about.

Rethinking Social Cognition in Light of Psychosis: Reciprocal Implications for Cognition and Psychopathology:

The positive symptoms of psychosis largely involve the experience of illusory social actors, and yet our current measures of social cognition, at best, only weakly predict their presence. We review evidence to suggest that the range of current approaches in social cognition is not sufficient to explain the fundamentally social nature of these experiences. We argue that social agent representation is an important organizing principle for understanding social cognition and that alterations in social agent representation may be a factor in the formation of delusions and hallucination in psychosis. We evaluate the feasibility of this approach in light of clinical and nonclinical studies, developmental research, cognitive anthropology, and comparative psychology. We conclude with recommendations for empirical testing of specific hypotheses and how studies of social cognition could more fully capture the extent of social reasoning and experience in both psychosis and more prosaic mental states.

Inner speech is not so simple : a commentary on Cho and Wu (2013).

We welcome Cho and Wu’s (1) suggestion that the study of auditory verbal hallucinations (AVHs) could be improved by contrasting and testing more explanatory models. However, we have some worries both about their criticisms of inner speech-based self-monitoring (ISS) models and whether their proposed spontaneous activation (SA) model is explanatory.

How anxiety induces verbal hallucinations

Highlights • Anxiety both triggers verbal hallucinations (VHs) and shapes their content.• A current model that accounts for this in terms of failed anticipation is critiqued.• An important subset of VHs arise from a particular anxious style of anticipation.• This anxious anticipation makes an experience of thinking more like one of perceiving.

The speaker behind the voice: therapeutic practice from the perspective of pragmatic theory.

Many attempts at understanding auditory verbal hallucinations have tried to explain why there is an auditory experience in the absence of an appropriate stimulus. We suggest that many instance of voice-hearing should be approached differently. More specifically, they could be viewed primarily as hallucinated acts of communication, rather than hallucinated sounds. We suggest that this change of perspective is reflected in, and helps to explain, the successes of two recent therapeutic techniques. These two techniques are: Relating Therapy for Voices and Avatar Therapy.

Delusions, dreams, and the nature of identification

Delusional misidentification is commonly understood as the product of an inference on the basis of evidence present in the subjects experience. For example, in the Capgras delusion, the patient sees someone who looks like a loved one, but who feels unfamiliar, so they infer that they must not be the loved one. I question this by presenting a distinction between “recognition” and “identification.” Identification does not always require recognition for its epistemic justification, nor does it need recognition for its psychological functioning. Judgments of identification are often the product of a non-inferential mechanism. Delusional misidentification arises as the product of this mechanism malfunctioning.

Levels and kinds of explanation: lessons from neuropsychiatry

I use an example from neuropsychiatry, namely delusional misidentification, to show a distinction between levels of explanation and kinds of explanation. Building on a pragmatic view of explanation, different kinds of explanation arise because we have different kinds of explanatory concerns. One important kind of explanatory concern involves asking a certain kind of “why” question. Answering such questions provides a personal explanation, namely, renders intelligible the beliefs and actions of other persons. I use contrasting theories of delusional misidentification to highlight how different facts about the phenomenon that is being explained impose constraints on the availability of personal explanation.

Predictive Processing and the Varieties of Psychological Trauma

A recently popular framework in the cognitive sciences takes the human nervous system to be a hierarchically arranged Bayesian prediction machine. In this paper, we examine psychological trauma through the lens of this framework. We suggest that this can help us to understand the nature of trauma, and the different effects that different kinds of trauma can have. We end by exploring synergies between our approach and current theories of PTSD, and gesture toward future directions.

A commentary on: Affective coding: the emotional dimension of agency

We welcome Gentsch and Synofziks (2014) review of the role of affect in modulating a persons sense of agency (SoA). The review is timely and synthesizes a great deal of evidence. However, we feel that their claims concerning the role of affect in modulating a persons SoA could be usefully extended to the study of auditory hallucinations (AH), one unusual experience not discussed in their review. In this commentary, we describe recent findings that suggest that one way in which negative affect plays a role in the development of AH is by reducing the SoA associated with self-generated cognitions and suggest that the insula may play an important role in mediating the effect of affect on SoA. Cognitive models of AH (e.g., Bentall, 1990; Waters et al., 2012) suggest that they occur when a cognition is misattributed to an external, non-self-source. Consistent with this claim, people who experience AH tend to show a bias toward accepting the presence of a verbal stimulus on tasks designed to measure reality discrimination (i.e., they tend to confuse internal, self-generated events for external, non-self-generated events, but not vice versa; Brookwell et al., 2013). This is often demonstrated using an auditory signal detection task (SDT), in which participants must try to detect speech in an ambiguous auditory stimulus (typically white noise). On trials where the speech is absent, participants have the opportunity to make a false alarm; that is, to report that speech was present in the white noise, when it was not. Presumably, when participants make a false alarm, they have mistaken their internal representation of the speech for the externally presented, “real” speech. Two recent studies have demonstrated a causal relation between increases in negative affect and weaker reality discrimination. Smailes et al. (2014) reported that participants who performed an auditory SDT after completing a negative mood induction (recalling an unpleasant autobiographical memory) made more false alarms, but not more hits, than did participants who performed the task after completing a neutral mood induction. Similarly, Hoskin et al. (2014), despite employing a different mood induction and a modified SDT, reported that participants were more likely to make false alarms during a condition in which they were exposed to a stressor than during a control condition. Thus, both studies showed that when participants experienced negative affect, they were more likely to misattribute internal, self-generated cognitions to an external source. These findings are consistent with data from studies that have shown that negative affect tends to precede the onset of AH in the daily lives of psychosis patients (Nayani and David, 1996; Delespaul et al., 2002), and suggest that negative affect is associated with the onset of AH, at least in part, because it modulates SoA for cognition. Gentsch and Synofzik propose that three stages of agency processing—prospective, immediate, and retrospective—exist. First person accounts (e.g., Romme et al., 2009; Scholtus and Blanke, 2012) suggest that people who experience AH do not go through a deliberative process to determine whether an unusual auditory percept was self-generated, or was a result of an external, non-self agent. Instead, an AH is experienced “in the moment” as something that was not self-generated. Thus, in terms of Gentsch and Synofziks proposed stages, it seems likely that the effect of negative emotions on a persons reality discrimination abilities would correspond to either prospective or immediate affective coding, rather than retrospective affective coding. That is, negative affect may reduce a persons SoA over cognition by interfering with action planning or with the generation of an accurate sensory outcome representation of a cognition. The most prominent cognitive models of AH suggest that a forward model system acts to predict the sensory outcomes of motor commands, and that dysfunction at one of a number of comparators can lead to a lack of agency over self-generated actions (Jones and Fernyhough, 2007). Interference with either action planning or accurate prediction could therefore lead to external misattributions of self-generated processes. A potential avenue for research would be to investigate whether negative affect can modulate, for example, the sensory attenuation which is associated with successful prediction via forward modeling. Gentsch and Synofziks review only briefly discusses the brain regions that may be involved in mediating the effects of negative affect on SoA. At two points, however, they cite evidence suggesting that damage to/atypical activity in the insula can lead to disorders of SoA. We concur that the insula is a good candidate for mediating the effects of negative affect on SoA. This is because, in addition to the evidence cited by Gentsch and Synofzik, a number of studies have shown (a) that different agency experiences are associated with changes in insula activity (e.g., Farrer and Frith, 2002; Farrer et al., 2003), (b) that insula activity is atypical in people who report AH (Wylie and Tregellas, 2010) and (c) that increases in negative affect are associated with changes in insula activity (Phan et al., 2002; Harrison et al., 2008). Research that examines whether negative affect brings about reductions in a persons SoA over their cognitions through modulation of activity in the insula is required. Demonstrating that affective problems may play an important role in the development of AH is important for a number of reasons. First, affect-induced changes in the SoA a person has over their cognitions can help to explain why AH are typically not experienced constantly (this issue is sometimes raised as a problem for cognitive models of AH; Gallagher, 2004). Seconds, and perhaps more importantly, it opens up the possibility of novel therapeutic interventions. While the primary focus of such interventions may be on ameliorating the affective problems reported by people who hear voices, they may indirectly reduce the frequency of AH by preventing negative affect-induced modulations of a persons SoA over their cognitions.