Working memory predicts presence of auditory verbal hallucinations in schizophrenia and bipolar disorder with psychosis

Abstract

Objective: The recent dramatic increase in research investigating auditory verbal hallucinations (AVHs) has broadened the former narrow focus on schizophrenia to incorporate additional populations that experience these symptoms. However, an understanding of potential shared mechanisms remains elusive. Based on theories suggesting a failure of top-down cognitive control, we aimed to compare the relationship between AVHs and cognition in two categorical diagnoses of psychosis, schizophrenia and psychotic bipolar disorder.

Method: A total of 124 adults aged 21-60 participated, of whom 76 had present-state psychosis (schizophrenia, n = 53; bipolar disorder with psychosis, n = 23), and 48 were non-clinical controls. Diagnosis and hallucination presence was determined using the Structured Clinical Diagnostic Interview for DSM-IV TR. AVHs severity was assessed using the Positive and Negative Syndrome Scale. Participants also completed the MATRICS cognitive battery.

Results: The bipolar disorder with psychosis group performed better than the schizophrenia group for cognitive domains of Processing speed, Attention, Working memory (WM), and Visual memory. Hierarchical binary logistic regression found that WM significantly predicted presence of AVHs in both psychotic groups, but diagnosis did not significantly increase the predictive value of the model. A hierarchical multiple linear regression found that schizophrenia diagnosis was the only significant predictor of hallucination severity.

Conclusions: The findings of this study-the first, to our knowledge, to compare the relationship between AVHs and MATRICS domains across schizophrenia and bipolar disorder with psychosis-support theories that deficits in WM underly the genesis of AVHs. WM potentially represents a shared mechanism of AVHs across diagnoses, supporting dimensional classifications of these psychotic disorders. However, non-cognitive factors predictive of hallucination severity may be specific to schizophrenia.

Keywords: Hallucinations; MATRICS consensus cognitive battery; psychosis; working memory.