Frontal and superior temporal auditory processing abnormalities in schizophrenia

Abstract

Background: Although magnetoencephalography (MEG) studies show superior temporal gyrus (STG) auditory processing abnormalities in schizophrenia at 50 and 100 ms, EEG and corticography studies suggest involvement of additional brain areas (e.g., frontal areas) during this interval. Study goals were to identify 30 to 130 ms auditory encoding processes in schizophrenia (SZ) and healthy controls (HC) and group differences throughout the cortex.

Methods: The standard paired-click task was administered to 19 SZ and 21 HC subjects during MEG recording. Vector-based Spatial-temporal Analysis using L1-minimum-norm (VESTAL) provided 4D maps of activity from 30 to 130 ms. Within-group t-tests compared post-stimulus 50 ms and 100 ms activity to baseline. Between-group t-tests examined 50 and 100 ms group differences.

Results: Bilateral 50 and 100 ms STG activity was observed in both groups. HC had stronger bilateral 50 and 100 ms STG activity than SZ. In addition to the STG group difference, non-STG activity was also observed in both groups. For example, whereas HC had stronger left and right inferior frontal gyrus activity than SZ, SZ had stronger right superior frontal gyrus and left supramarginal gyrus activity than HC.

Conclusions: Less STG activity was observed in SZ than HC, indicating encoding problems in SZ. Yet auditory encoding abnormalities are not specific to STG, as group differences were observed in frontal and SMG areas. Thus, present findings indicate that individuals with SZ show abnormalities in multiple nodes of a concurrently activated auditory network.

Keywords: Auditory; DTI, diffusion tensor imaging; ECG, electrocardiogram; EEG, electroencephalography; EOG, electro-oculogram; ERF, event-related field; ERP, event-related potential; FDR, false discovery rates; Frontal cortex; HC, healthy controls; IFG, inferior frontal gyrus; ITG, inferior temporal gyrus; MEG; MEG, magnetoencephalography; PANSS, Positive and Negative Syndrome Scale; PFC, prefrontal cortex; S1, first click; S2, second click; SES, socioeconomic status; SFG, superior frontal gyrus; SMA, supplementary motor area; SMG, supramarginal gyrus; SSS, Signal Space Separation; STG, superior temporal gyrus; Schizophrenia; Superior temporal gyrus; VESTAL, Vector-based Spatio-temporal Analysis using L1-minimum norm; fMRI, functional magnetic resonance imaging; sMRI, structural magnetic resonance imaging.

Supplementary Fig. 1

Within-group VESTAL statistics for 50 (left panel) and 100 ms (right panel) S2 activity showing left and right STG (L-STG, R-STG) and frontal regions (L-Frontal, R-Frontal) activated areas for HC (top panel) and SZ (bottom panel) (thresholded at FDR q < 0.01).

Supplementary Fig. 2

Between-group analyses for 50 ms S2 activity. Activation clusters in yellow/red (thresholded at FDR q < 0.05) show stronger L-STG, R-STG, and frontal medial cortex (FMC) activity in HC than SZ (HC > SZ). Activation clusters (thresholded at FDR q < 0.05) in blue show stronger activity in left middle frontal gyrus (L-MFG), L-SMG, right lateral occipital cortex (R-LOC), right frontal pole (R-FP), and right postcentral gyrus (R-PG) in SZ than HC (SZ > HC).

Supplementary Fig. 3

Between-group analyses for 100 ms S2 activity. Activation clusters in yellow/red show stronger activity in L-STG, R-STG, and FMC in HC than SZ (HC > SZ). Activation clusters in blue show stronger activity in L-SMG, L-MFG, R-LOC, R-PG in SZ than HC (SZ > HC).

Fig. 1

VESTAL data processing stream. (A) T1-MRI from an individual subject; (B) MNI-152 Atlas space; (C) cortical mask from MNI-152; (D) cortical mask transferred back to the individual MRI space; (E) VESTAL source grid with cortical and subcortical regions. Gray triangles are BEM mesh for MEG forward calculation; (F) VESTAL source image of the subject's auditory response overlayed on the T1-MRI; (G) VESTAL activity transferred to the MNI-152 coordinates; (H) regional time-course from VESTAL results.

Fig. 2

Within-group VESTAL statistics for 50 (left panel) and 100 ms (right panel) activity showing significantly activated areas for HC (top panel) and SZ (bottom panel) (thresholded at FDR q < 0.01).

Fig. 3

Between-group analyses for 50 ms activity. Activation clusters in yellow/red (thresholded at FDR q < 0.05) show stronger activity in HC than SZ (HC > SZ). Activation clusters (thresholded at FDR q < 0.05) in blue show stronger activity in SZ than HC (SZ > HC). The effect sizes for L-STG, R-Frontal and R-SFG M50 measures are provided in Table 2.

Fig. 4

Between-group analyses for 100 ms activity. Activation clusters in yellow/red show stronger activity in HC than SZ (HC > SZ). Activation clusters in blue show stronger activity in SZ than HC (SZ > HC). The effect sizes for L-STG, R-Frontal and R-SFG M100 measures are provided in Table 2.