Impaired early visual response modulations to spatial information in chronic schizophrenia

Abstract

Early visual processing stages have been demonstrated to be impaired in schizophrenia patients and their first-degree relatives. The amplitude and topography of the P1 component of the visual evoked potential (VEP) are both affected; the latter of which indicates alterations in active brain networks between populations. At least two issues remain unresolved. First, the specificity of this deficit (and suitability as an endophenotype) has yet to be established, with evidence for impaired P1 responses in other clinical populations. Second, it remains unknown whether schizophrenia patients exhibit intact functional modulation of the P1 VEP component; an aspect that may assist in distinguishing effects specific to schizophrenia. We applied electrical neuroimaging analyses to VEPs from chronic schizophrenia patients and healthy controls in response to variation in the parafoveal spatial extent of stimuli. Healthy controls demonstrated robust modulation of the VEP strength and topography as a function of the spatial extent of stimuli during the P1 component. By contrast, no such modulations were evident at early latencies in the responses from patients with schizophrenia. Source estimations localized these deficits to the left precuneus and medial inferior parietal cortex. These findings provide insights on potential underlying low-level impairments in schizophrenia.

Copyright © 2011 Elsevier Ireland Ltd. All rights reserved.

Figures

Figure 1

Schematic of the relative eccentricity of the stimuli. For simplicity, only the stimulus orientations forming an illusory contour shape are shown. The wide eccentricity is shown in red. The narrow eccentricity is shown in blue.

Figure 2

Exemplar group-average VEPs from healthy controls (panel a) and chronic patients with schizophrenia (panel b) at a parietal midline scalp site (Pz). In these plots the solid lines represent the illusory contour (IC) stimulus condition and the dotted lines the non-illusory contour (NC) stimulus condition. Red lines refer to the wide stimulus eccentricity and blue to the narrow stimulus eccentricity. Panel c displays the results of a time-point by time-point non-parametric F-test on the VEP at electrode Pz (alpha ≤0.05; temporal criterion of at least 6 contiguous time-points).

Figure 3

Group-average global field power (GFP) waveforms from healthy controls (panel a) and chronic patients with schizophrenia (panel b). Conventions for the plots are identical to those in Figure 2. Panel c displays the results of a time-point by time-point non-parametric F-test on the GFP (alpha ≤0.05; temporal criterion of at least 6 contiguous time-points). Panels d and e display the post-hoc analyses on each group using the same criteria as in panel c.

Figure 4

Panel a displays the results of the topographic cluster analysis of the group-averaged VEPs from both patients and controls. Over the initial ∼250ms post-stimulus period, 4 topographies were identified in the group-averaged dataset. Two of these were observed over the 54-108ms post-stimulus period and were in turn used for single-subject fitting (see Materials and Methods for details). Panel b displays the results of the single-subject fitting analysis wherein both of the two topographies identified over the 54-108ms was spatially correlated with the instantaneous VEP. The bar graph shows the mean amount of time (over the 54-108ms post-stimulus interval) labeled with each topography (s.d. indicated) as a function of each cohort and stimulus eccentricity (i.e. data were collapsed across stimulus conditions for display purposes).

Figure 5

Source estimation analyses. Panel a depicts the loci within the source estimation space that exhibited a significant group * stimulus eccentricity interaction (non-parametric F-test). This test was performed after first averaging the VEPs from each condition and group as a function of time (54-108ms) to obtain a single vector per participant and condition that was in turn submitted to the source estimation and statistically analyzed. Two clusters were identified, which for clarity are colored orange and green. Panels b and c display the mean (s.d. indicated) scalar values within the cluster and across subjects and stimulus conditions. The coordinates shown are according to the Talairach and Tournoux (1988) atlas. Brodmann areas (BA) are likewise indicated. In both clusters, controls exhibited a significant difference between wide and narrow stimulus eccentricities, whereas patients did not.