Theta-gamma phase-amplitude coupling in psychosis and healthy controls: Predicting working memory capacity across different tasks

Abstract

Theta-gamma phase-amplitude coupling has been proposed as a possible mechanism underlying working memory. Here, we assess whether cortical theta-gamma phase-amplitude coupling is related to working memory deficits in patients with psychosis, and whether any group differences observed in this neural measure are specifically related to group differences in working memory performance or instead reflect more general differences in brain dynamics. To address these issues, we collected 61-channel EEG data from 32 psychiatric patients with psychosis (including patients with schizophrenia, schizoaffective disorder, and bipolar disorder with psychotic features) and 35 age-matched healthy controls, and calculated the Phase Locking Value, a measure of theta-gamma phase-amplitude coupling across three tasks: a working memory task (Active WM+), a visual search task (Active WM−), and a passive viewing task (Passive WM−). We found significant theta-gamma phase-amplitude coupling across all tasks, including during a pre-stimulus baseline period. No significant differences between the PSY and HC groups were found, although the two groups did show some small differences in the specific patterns of coupling. However, surprisingly, we found that frequency dynamics in coupling strongly and significantly predicted individual differences in working memory capacity in both groups. This was true even when the coupling was assessed during a passive viewing paradigm that did not involve a working memory component. Taken together, these findings suggest that cortical theta-gamma phase-amplitude coupling may reflect behaviorally-relevant differences in structural or functional connections that persist regardless of stimuli, task or time window in both those with and without psychosis.