Identifying the cognitive underpinnings of voice-hearing by comparing never, past and current voice-hearers.

Abstract

The current study aimed to compare specific cognitive profiles corresponding to auditory verbal hallucinations (AVH) status and elucidate which pattern of cognitive deficits may predict voice-hearing status. Clinical participants with schizophrenia spectrum disorders were partitioned into: (i) current voice-hearers (n=46), (ii) past voice-hearers (n=37) and (iii) never voice-hearers (n=40), and compared with 319 non-clinical controls. Cognitive assessment employed the MATRICS Consensus Cognitive Battery (MCCB), supplemented by the Delis-Kaplan Executive Function System (D-KEFS) Colour-Word Interference Test (Stroop) as a robust measure of executive function. On the Visual Learning domain, current and past voice-hearers had significantly poorer performance relative to never voice-hearers, who in turn had significantly poorer performance than non-clinical controls. Current and never voice-hearers had significantly poorer performance on the Social Cognition domain relative to non-clinical controls. Current voice-hearers also had significantly poorer performance on the Inhibition domain relative to non-clinical controls. Binary logistic regression revealed that Visual Learning was the only significant cognitive predictor of AVH presence. Visual learning, and potentially inhibition, may be viable therapeutic targets when addressing cognitive mechanisms associated with AVHs. Future research should focus on investigating additional cognitive mechanisms, employing diverse voice-hearing populations and embarking on related longitudinal studies.