Visual perception as a window onto prediction anomalies in schizophrenia

点击翻译按钮获取中文摘要

Abstract What are the aberrant brain processes that lead to symptoms of schizophrenia—to the distressing and disabling hallucinations, delusions, disorganization, and loss of motivation? Answering this question is central to developing targeted and effective treatments. Prominent mechanistic accounts of schizophrenia hinge on the notion of the brain as a “prediction machine” that maintains and updates a mental model of the probabilistic structure of the environment, which it uses in concert with sensory input to make sense of the world. Schizophrenia has been associated with an abnormality in this interpretative process, leading to abnormal perceptions, beliefs, and behaviors. What is currently lacking, however, is an empirical basis for delineating the specific nature of prediction abnormalities in schizophrenia, at the level of both behavior and brain. The general aim of the current project is to understand how visual perception is influenced by experience-based predictions in the schizophrenia. The visual system is a uniquely suited system for understanding prediction abnormalities in clinical populations for several reasons. First, robust behavioral paradigms can quantify the influence of predictions on visual perception. Second, parallel neurophysiology and neuroimaging work provides a basis for interpretation at the level of neuronal populations. Understanding these abnormalities in vision, then, may provide a scalable framework for understanding symptoms more generally. Furthermore, visual distortions are observed in schizophrenia before illness onset, and they relate to important clinical factors. Understanding prediction in the visual system can help explain specific clinical phenomenology. The current project proposes to investigate the influence of prior experience on visual processing by measuring visual aftereffects and their neural concomitants. Aftereffects are illusory perception of the “opposite” that arises after prolonged viewing of a image. Characterizing visual aftereffects in the schizophrenia spectrum can provide important insights into the computational and biological underpinnings of abnormal prediction. The substantial existing literature on the neural and computational origins of aftereffects means that different aftereffects can reveal at what level of the sensory hierarchy, and in which specific component processes, prediction abnormalities emerge. Specific study goals are to 1) characterize visual aftereffects in the schizophrenia spectrum; 2) determine whether they are present across illness stages and in individuals at-risk for the disorder; 3) evaluate the clinical relevance of altered visual aftereffects; and 4) to measure neuroimaging concomitants of altered aftereffects and link them to computational model components. Visual aftereffects can provide a tool with which to empirically test a canonical computational mechanism as the basis for both altered visual experience and symptom genesis generally, namely an inappropriate use of prior experience in generating the contents of consciousness.