P2 Visual Event-Related Potential Response Predicts Clinical Outcomes in Patients at Clinical High Risk for Psychosis - Transcript for SIRS 2025 Early Career Award
Jennifer Lepock
80%-90% of schizophrenia (SCZ) patients experience a period of less intense symptoms preceding psychosis with changes in mood, behavior, interests and thoughts (Addington & Heinssen, 2012) For example, whereas an individual with SCZ may have a fixed belief the FBI are spying on them, a Clinical High Risk (CHR) individual may have fleeting thoughts that they are being watched or followed by unknown persons, and still have insight that this may not be true. Our ability to predict SCZ based on CHR symptoms is still limited. A majority of CHR individuals will not go on to develop SCZ, but their psychiatric follow-up requires substantial time and cost for them and the health-care system, and carries risks of side effects and stigmatization. To avoid unnecessary treatment and target interventions to those most at risk, new tests are needed to predict which CHR patients are most likely to develop SCZ. Although only a minority of CHR patients develop psychosis, approximately half of those who do not do so continue to experience persistent symptoms. Moreover, approximately half of CHR patients who do not develop a psychotic disorder have poor social and/or occupational functioning at follow-up.
Neurophysiological indices are a potential tool for enhancing our ability to predict outcomes in CHR patients. These indices offer the advantage of being objective and rater-independent . Cognitive event-related brain potentials (ERPs) are a type of neurophysiological index that uses electroencephalography to non-invasively record voltage changes at the scalp associated with cognitive events such as stimuli or responses. A number of ERP waveforms have been found to be reliably reduced in schizophrenia, contributing to our knowledge about its underlying neurocognitive pathophysiology.
Some of these ERP waveforms have demonstrated promising predictive results in CHR patients.
ERP measure that could plausibly contribute to prognostic algorithms in CHR populations is the N400 waveform.
400 amplitudes are normally smaller (less negative) to meaningful stimuli that are more related to preceding context. This “N400 semantic priming effect” is attenuated in patients with schizophrenia, compared with control individuals. We have found CHR individuals to also have attenuated N400 semantic priming effects compared to controls, and these deficiencies predicted functioning at one and two year follow up.
There is evidence that prediction effects can be seen even earlier when a stimulus fits what is predicted, during the stage of early visual processing of word form reflected by the P2, during the 175-300 ms time window after stimulus onset.
In the present study, we aimed to examine whether the P2 amplitudes are abnormal in patients at clinical high risk (CHR). In CHR patients, P2 is an early indicator of using context to predict what comes next with regard to the orthographic form of predicted words, and deficits in this prediction may reflect increased risk for functional impairment.
