Multichannel electroencephalographic assessment of auditory evoked response suppression in schizophrenia

Abstract.

Reduced auditory evoked response (AER) suppression in a paired-stimulus paradigm (where suppression equals the difference between S1 and S2 amplitudes divided by S1 amplitude) may index genetic liability for schizophrenia. The present report is a multiple-channel electroencephalographic (EEG) study of AER suppression among 20 normal and 20 schizophrenia subjects. The typical paired-stimulus paradigm was used to evoke time-locked AERs. AER responses were scored at P50 and N100 in the time domain using both single (Cz) and multichannel data (after reduction using principal components analysis, PCA), and were scored for information in the gamma (20–50 Hz) and low-frequency (1–20 Hz) ranges using multichannel information (also after PCA). The time domain analyses demonstrated that schizophrenia patients differ from normal in amplitude of response to the first, but not to the second, stimulus for both P50 and N100. The frequency domain data demonstrated that schizophrenia patients differed from normal on amplitude of the low-frequency response (LFR) to the first, but not to the second, stimulus. The groups did not differ significantly on amplitudes of the gamma-band responses. Group separations were largest for the multichannel N100 and LFR data, with the LFR demonstrating a modestly better risk ratio for differentiating schizophrenia from normal subjects. The present results suggest two novel differences from previous AER suppression studies: (1) S1 amplitudes largely determine differences between normal and schizophrenia groups on AER suppression, and (2) frequency domain analyses may provide important complimentary information when studying AERs in schizophrenia.