Aberrant auditory evoked responses in schizophrenia: evidence from single-trial analysis

The average N100 (a negative response occurring around 100 ms poststimulus) component of the auditory evoked potential (EP) has been recently used in the study of schizophrenia. Averaging, however, eliminates all temporal variability of the recorded signals and, therefore, hampers the exploration of the temporal dynamics underlying the generation of the N100 component. In this study, we analyzed EPs on a single-trial basis using an iterative independent component analysis procedure that is capable of extracting individual components out of an entire EP waveform. This approach allowed estimation of an N100 in each single trial and measurement of its morphological features such as polarity, which could be either negative (most frequently) or positive (less frequently). In the latter case, the N100 component was termed aberrant. We analyzed responses from 23 normal controls (NC) and 15 schizophrenia (SZ) patients in a paired stimulus paradigm, where a first stimulus S(1) was followed by a second one S(2) 0.5 s later. To compare N100 responses within and across the two subject groups, we defined a negative polarity index NPI as the percentage of single trials that had a negative polarity N100. Our results show significantly higher NPI values in NC compared to SZ, for both the S(1) and S(2) responses. Additionally, the difference in NPI values between the S(1) and S(2) responses was significant in NC but not in SZ. We conclude that both normal and schizophrenia subjects exhibit aberrant N100 responses, but these events are more frequent in the SZ patient group. The higher number of aberrant responses can explain the lower amplitude EPs typically observed in schizophrenia, and may be one of the factors contributing to sensory gating deficits consistently reported in these patients.