Neural substrates of visual processing and object recognition deficits in schizophrenia.

摘要

Mounting evidence has shown that patients with schizophrenia have preferential deficits of the magnocellular versus the parvocellular visual system. Experiment 1 examined this deficit in schizophrenia patients utilizing an electrophysiological paradigm. Patients showed preferential magnocellular deficits in electrophysiological response indicative of impaired contrast gain (response amplification at low contrast) and contrast gain control (inhibition of responses at high contrast), which are used preferentially by this pathway to optimize responses. Patients also displayed deficits in psychophysical contrast sensitivity, further showing deficient contrast gain in the magnocellular pathway. These electrophysiological and psychophysical deficits were associated with neuropsychological and emotion processing deficits, which predicted functional outcome.;Experiment 2 utilized functional magnetic resonance imaging (fMRI) to examine the neural underpinnings of the paradigms used in Experiment 1. fMRI responses to magnocellular- and parvocellular-biased contrast stimuli from the electrophysiological paradigm showed that contrast gain (i.e., signal amplification) was related to increases in volume of relatively weak occipital activation, while contrast gain control (i.e., signal inhibition) was related to strong a occipital activation over a smaller volume. Inhibitory contrast gain control was also linked to negative parafoveal activation, which was less apparent for patients. fMRI responses to a contrast sensitivity procedure showed reduced volume of occipital activation to low spatial frequency (LSF), but not high spatial frequency (HSF), stimuli for patients, indicating a general deficit in activation volume for LSF stimuli which are preferentially processed by the magnocellular system.;Experiment 3 examined consequences of magnocellular dysfunction for object recognition in schizophrenia. Patients showed deficits in fMRI activation to LSF object stimuli over a widespread cortical network, indicating a loss of early-stage low resolution object information. Patients instead showed an increase in activation to HSF object stimuli in some areas, suggesting compensation for LSF deficits with HSF information. Together, these three experiments further elucidated the neural substrates of preferential magnocellular deficits in schizophrenia, and demonstrated that such deficits may propagate to higher cognitive processes such as object recognition.