Oscillatory modulations in theta and beta band dissociate two types of responses in auditory condensation task
Successful performance in cognitive tasks relies on the deployment of cognitive control mechanisms including non-specific regulation of motor threshold and modulation of attentional processes specific to the task. Failure of each of the two mechanisms may lead to different kind of responses, post-response adaptations and, importantly, distinctive behavior correlates. In particular, slow correct and erroneous responses could be associated with attentional lapses and high levels of uncertainty; whereas fast responses are probably associated with low levels of uncertainty, with occasional errors caused by a failure to withhold action impulses. In the present study, we aimed to ask whether modulations of oscillatory activity indicate two different types of responses. We recorded EEG while subjects performed auditory version of two-choice condensation task, which is highly demanding for sustained attention, but implies no inhibition of prepotent responses. We found enhanced frontal midline theta (FMT) power in the pre-response time window on both correct and erroneous responses. Error-related FMT power increase was observed in post-response and post-feedback time windows. Frontal beta band oscillatory activity was enhanced exclusively after positive feedback presentation. Trial-by-trial correlation analysis between the FMT power and the response time (RT) revealed: (1) positive correlation before the response for both correct and erroneous trials, (2) negative correlation after the erroneous responses. Post-feedback frontal beta-band power positively correlated with the RT. Thus, slower errors were characterized by higher FMT power before the response and higher frontal beta power after positive feedback presentation, suggesting stronger cognitive effort and intensive feedback processing of more informative feedback, respectively, in the situation of high uncertainty. On the contrary, higher FMT power after error commission was associated with lower RT’s, implying instantaneous internal error detection in the case of low outcome uncertainty. Taken together, our results suggest that single-trial RT may serve as a valid measure for distinguishing slow trials with high levels of decision uncertainty due to attentional lapses and fast erroneous trials with preserved attentional processes and motor threshold failures. The study was implemented in the framework of The Basic Research Program at the National Research University Higher School of Economics in 2017.