Effects of Domain-Specific Noise on Visual Motion Processing in Schizophrenia

Abstract

Background: Visual perception impairments in schizophrenia stem from abnormal information processing. Information processing requires neural response to a stimulus (signal) against a backdrop of 1) random variation in baseline neural activity (internal noise) and sometimes irrelevant environmental stimulation (external noise). Filtering out noise is a critical aspect of information processing, and needs to be critically examined in schizophrenia. Methods: To understand how noise in the visual system constrains perceptual processing, we devised a novel paradigm to build in both signal and external noise on same visual stimulus. Here, instead of uniformed noise, domain-specific noise—variations in stimulus speed—was introduced to evaluate the performance of schizophrenia patients in speed discrimination. Each motion stimulus—a random dot pattern (RDP) comprising 200 moving dots—included a range of speeds, drawn individually from a Gaussian distribution for each dot. The task for patients (n = 26) and controls (n = 28) was to identify which of two stimuli moved faster based on their mean speeds. Findings: Patients exhibited deficient speed discrimination at baseline, in the absence of speed noise. Their speed discrimination was further degraded in the presence of low and medium levels of external noise. In the presence of a high levels of noise, degradation of patients' speed discrimination leveled-off, resulting in similar performance to controls. Conclusion: These domain-specific noise effects on speed discrimination provide direct evidence for the existence of heightened internal noise within a specific visual motion processing domain in schizophrenia.