Remote State Representation in Early Psychosis

The purpose of the current study is to investigate computationally-informed precision treatments to improve two forms of state representation dysfunction observed in psychosis: 1) Abnormal perceptual inputs that impair state estimation; or, 2) Reduced state representation stability that affects cognitive control, working memory, and behavioral outputs. We will test the effects of two forms of cognitive training: visual perception training or visual cognitive control training in individuals with early psychosis. Participants will have the option to complete all training and assessments entirely remotely. We will recruit both young adults who have been diagnosed with a psychosis spectrum illness (such as schizophrenia) as well as individuals without a history of psychosis to participate in this study.

Early psychosis can manifest low-level perceptual deficits (such as an abnormal mismatch negativity response); these perceptual abnormalities are observed in ~60% of individuals, where they are predictive of more severe disability at 12 month follow-up, consistent with multiple studies showing that perceptual input abnormalities, when present, have a widespread deleterious downstream impact. Psychotic disorders can also manifest deficits in working memory, consistent with dysfunctional state representation stability, seen in ~80% of patients. Thus, psychosis is heterogeneous in its underlying information processing pathology and clinical course, indicating a critical unmet need for precision treatment approaches.

We will address this unmet need by investigating the behavioral and neurophysiologic effects of a brief course of either visual perception training (designed to improve state estimation processes at the perceptual input level) or visual cognitive control training (designed to enhance state representation stability of visual information), in individuals with psychotic disorders such as schizophrenia, schizoaffective disorder, and bipolar disorder with psychosis. Because study visits may be conducted remotely, participants will be drawn from a national sample. Our goal is not to perform a treatment efficacy study comparing these two interventions. Rather, we seek to use predictions derived from basic and computational neuroscience to test the effects of neuroplasticity-based precision treatments targeting two distinct contributing information processing pathologies in psychosis, with the goal of improving state representation processes and cognition.