Systematic review of visual illusions in schizophrenia.

Visual illusions have long been used as tools to investigate sensory-perceptual deficits in schizophrenia. Recent conflicting accounts have called into question the assumption of abnormal illusion perception in patients and, therefore, the validity of this approach. Here, we present a systematic review of the current evidence regarding visual illusion perception abnormalities in patients with schizophrenia. Relevant publications were identified by a systematic search of PubMed, Literatura LILACS, PsycINFO, Embase, Scopus, Cochrane Central Register of Controlled Trials (CENTRAL), IBECS, BIOSIS, and Web of Science. Forty-five studies were selected which included illusions classified as 'Motion illusions', 'Geometric-optical illusions', 'Illusory contours', 'Depth inversion illusion', and 'Non-specific'. There is concordant evidence of abnormal processing of illusions in patients for most categories, especially in facial Depth Inversion and Müller-Lyer illusions. There were significant methodological disparities and shortcomings, but risk of bias was overall low for individual studies. The usefulness of visual illusions as tools in clinical settings as well as in basic research may be contingent on significant methodological refinements.

The Perception of the Müller-Lyer Visual Illusion in Schizophrenics and Non-human Primates: A Translational Approach.

The Müller-Lyer Illusion (MLI) has been suggested as a potential marker for the perceptual impairments observed in schizophrenia patients. Along with some positive symptoms, these deficits are not easily modeled in rodent experiments, and novel animal models are warranted. Previously, MK-801 was shown to reduce susceptibility to MLI in monkeys, raising the prospects of an effective perception-based model. Here, we evaluate the translational feasibility of the MLI task under NMDA receptor blockage as a primate model for schizophrenia. In Experiment 1, eight capuchin monkeys (Sapajus spp.) were trained on a touchscreen MLI task. Upon reaching the learning criteria, the monkeys were given ketamine (0.3 mg/kg; i.m.) or saline on four consecutive days and then retested on the MLI task. In Experiment 2, eight chronic schizophrenia patients (and eight matching controls) were tested on the Brentano version of the MLI. Under saline treatment, monkeys were susceptible to MLI, similarly to healthy human participants. Repeated ketamine administrations, however, failed to improve their performance as previous results with MK-801 had shown. Schizophrenic patients, on the other hand, showed a higher susceptibility to MLI when compared to healthy controls. In light of the present and previous studies, the MLI task shows consistent results across monkeys and humans. In spite of potentially being an interesting translational model of schizophrenia, the MLI task warrants further refinement in non-human primates and a broader sample of schizophrenia subtypes.