Early-stage visual processing and cortical amplification deficits in schizophrenia

Background: Patients with schizophrenia show deficits in early-stage visual processing, potentially reflecting dysfunction of the magnocellular visual pathway. The magnocellular system operates normally in a nonlinear amplification mode mediated by glutamatergic (N-methyl-D-aspartate) receptors. Investigating magnocellular dysfunction in schizophrenia therefore permits evaluation of underlying etiologic hypotheses. Objectives: To evaluate magnocellular dysfunction in schizophrenia, relative to known neurochemical and neuroanatomical substrates, and to examine relationships between electrophysiological and behavioral measures of visual pathway dysfunction and relationships with higher cognitive deficits. Design, Setting, and Participants: Between-group study at an inpatient state psychiatric hospital and outpatient county psychiatric facilities. Thirty-three patients met DSM-IV criteria for schizophrenia or schizoaffective disorder, and 21 nonpsychiatric volunteers of similar ages composed the control group. Main Outcome Measures: (1) Magnocellular and parvocellular evoked potentials, analyzed using nonlinear (Michaelis-Menten) and linear contrast gain approaches; (2) behavioral contrast sensitivity measures; (3) white matter integrity; (4) visual and nonvisual neuropsychological measures, and (5) clinical symptom and community functioning measures. Results: Patients generated evoked potentials that were significantly reduced in response to magnocellularbiased, but not parvocellular-biased, stimuli (P = .001). Michaelis-Menten analyses demonstrated reduced contrast gain of the magnocellular system (P = .001). Patients showed decreased contrast sensitivity to magnocellular-biased stimuli (P < .001). Evoked potential deficits were significantly related to decreased white matter integrity in the optic radiations (P < .03). Evoked potential deficits predicted impaired contrast sensitivity (P = .002), which was in turn related to deficits in complex visual processing (P <= .04). Both evoked potential (P <= .04) and contrast sensitivity (P = .01) measures significantly predicted community functioning. Conclusions: These findings confirm the existence of early-stage visual processing dysfunction in schizophrenia and provide the first evidence that such deficits are due to decreased nonlinear signal amplification, consistent with glutamatergic theories. Neuroimaging studies support the hypothesis of dysfunction within low-level visual pathways involving thalamocortical radiations. Deficits in early-stage visual processing significantly predict higher cognitive deficits.