Examining the Simple and Complex Mismatch Negativity in Early Phase Psychosis

Abstract

Background: Using electroencephalography (EEG) to examine the simple mismatch negativity (MMN), a marker of auditory cortex function, has been of great interest in the exploration of biomarkers for psychotic illness. Despite many studies reporting MMN deficits in chronic schizophrenia, there are not reliable reports of MMN reductions in the early phase of the illness, suggesting the MMN is not a sensitive enough measure of vulnerability to be used as a biomarker. Recently, a more computationally complex measure of auditory cortex function (the complex mismatch negativity; cMMN) has been hypothesized to provide a more sensitive marker of illness vulnerability. Methods: The current study employed a novel dual rule cMMN paradigm to examine the cMMN in 14 individuals with early phase psychosis (EPP) and 15 healthy controls (HC). Additionally, the MMN response to five deviant types was also recorded using the optimal multi-feature MMN paradigm in 13 individuals with EPP and 17 HC. Demographic variables, symptom severity, and measures of functioning were all collected to explore relationships with the neurophysiological variables. Results: We found significant reductions of cMMN amplitudes at the site of maximal amplitude (Fz) in EPP (p = .017) with large effect sizes (Hedges’ g = 0.96). We also found a reduction of MMN amplitudes in response to duration deviants in the left frontal region (p = .036, Hedges’ g = 0.83). There were also correlations between more severe positive psychosis symptoms of auditory hallucinations (r = -0.582, p = .037) and unusual thought content/delusional ideas (r = -0.601, p = .030) and increased MMN amplitudes in response to duration deviants. Increased symptoms of unusual thought content/delusional ideas were also related to higher MMN amplitudes following gap (r = -0.558, p = .047) and location deviants (r = -0.590, p = .034). Positive psychosis symptoms of suspiciousness/persecutory ideas were related to a higher MMN in response to frequency deviants (r = -0.670, p = .012) and grandiose ideas were related to a higher MMN in response to gap deviants (r = -0.556, p = .049). Negative psychosis symptoms of social anhedonia were related to higher MMN amplitudes following duration (r = -0.790, p = .001) and frequency deviants (r = -0.625, p = .022), while reduced experience of emotions and self was related to higher MMN amplitudes following the location deviant (r = -0.603, p = .029). Additionally, there were significant correlations between higher cMMN amplitudes and more severe auditory hallucinations (r = -0.632, p = .015). Discussion: This study is an early step in the exploration of the cMMN as a biomarker for psychosis risk and provides evidence that the dual rule cMMN paradigm shows promise as a method for cMMN elicitation. Future studies should utilize this paradigm to examine the cMMN in a sample of high-risk individuals while employing a longitudinal design to determine the predictive capability of this measure.