SU75. Neuromagnetic Mismatch Negativity in Individuals at Ultra-High Risk of Psychosis

摘要

>Background: Auditory mismatch negativity (MMN) is an event-related potential/field elicited automatically by violations of previously established auditory regularities (Näätänen et al, 2007). MMN amplitudes have been consistently found to be attenuated in both medicated and unmedicated schizophrenia (ScZ) patients (Umbricht and Krljes, 2005), potentially indicating aberrant predictive coding processing. Moreover, recent studies have found MMN deficits to be present already in individuals at ultra-high risk (UHR) for psychosis, indicating that MMN is compromised before psychosis onset and could represent a marker of risk for psychosis development. >Methods: The purpose of the current study is to investigate MEG recorded MMNm amplitude responses to both duration deviants and sound omissions in n = 50 UHR individuals who were screened with the Schizophrenia Proneness Instrument and the Comprehensive Assessment of At Risk Mental States. N = 25 healthy controls served as a comparison group. We employed an auditory oddball paradigm in which 3 different sequences of auditory stimuli were presented binaurally at ~70 dB (150 ms SOA, 700–100 ms ISI). The standard sequence contained 5 identical tones (80 ms) and the deviant sequence contained 4 identical tones and 1 duration deviant tone (40 ms). In addition, an omission sequence which contained only 4 identical tones was included to examine auditory predictions. Participants were instructed to focus their attention away from the sounds and to perform a simple visual detection task. >Results: We analyzed MMNm responses to unpredictable deviant and omitted sounds in sensor and source space. The linearly constrained minimum variance beamformer was used to localize sources of the MMNm responses and to compute artifact-free source-level time courses. Consistent with previous studies revealing MMN generators in auditory cortical areas, bilateral MMNm sources were localized in auditory cortices for both duration deviants and sound omissions. Our data show attenuated MMNm responses elicited by duration deviant sounds in UHR-individuals relative to control participants. >Conclusion: Overall, the study findings suggest that reduced MMNm amplitude is present before psychosis onset and reflects aberrant predictive mechanisms rather than deficient stimulus-specific adaptation. Clinical follow-up measurements will be obtained to determine whether MMNm amplitudes are a potential biomarker for predicting transition to psychosis.