Investigating Concurrent and Longitudinal ERP-Symptom Relationships Among Risk for Psychosis

Keisha D Novak, Purdue University

Abstract

Cognitive impairments in schizophrenia (SZ) include abnormalities in executive function, attention, and semantic processing. Event-related potentials (ERPs) are used as neurophysiological measures of cognitive impairment that have been shown to map onto symptom dimensions of psychotic disorders, such as schizophrenia. While much research exists on schizophrenia, less is understood about the longitudinal relationships between ERPs and symptom dimensions among individuals at risk for psychosis. Of published work in risk samples, most have been crosssectional, leaving clinical inferences regarding longitudinal patterns non-specific. The current study aimed to bridge this gap by recording ERPs (P300, ERN, N400) across a battery of tasks within a single risk sample, and measured positive, negative, and disorganized symptom severity via the Multidimensional Schizotypy Scale (MSS). Participants exhibiting psychosis-risk were recruited from the community (N=60), and completed a baseline and 6-month follow-up assessment (n=29). The primary goal of the baseline assessment aimed to replicate ERP-symptom dimension relationships observed in the SZ literature. Effect sizes for P300-positive and ERNnegative relationships were observed to be in the same directionality as noted in the clinical SZ literature. While not statistically significant, the small effects suggest that P300 and ERN may be similarly effected by presence of positive and negative symptoms, respectively. By contrast, N400, however, was found to have an effect size directionality opposite to that reported in the literature. This finding is consistent with mixed presentation of disorganized symptoms in clinical SZ populations. The follow-up assessment aimed to examine the relationship of symptom dimensions over time in a single at-risk sample, and leveraged ERPs as potential prospective predictors of worsening of symptoms. As expected, baseline symptoms prospectively predicted corresponding symptoms at follow-up. However, only N400 amplitude at baseline correlated with disorganized symptoms at follow-up, and no ERP prospectively predicted corresponding symptom dimensions at follow-up. Overall, examining the relationship between multiple ERPs and symptom dimensions in a single sample and via a longitudinal design is a novel addition to the literature. Future research will be necessary to clarify the use of ERPs as neural biomarkers to identify and predict symptom severity over time, ultimately reducing subjectivity in clinical diagnosis and treatment.

Degree

Ph.D.

Advisors

Foti, Purdue University.

Subject Area

Pathology|Mental health|Language|Clinical psychology|Demography|Ethnic studies|Logic|Neurosciences|Psychology|Sociology

COinS