Brain event-related potentials predict individual differences in inhibitory control.

Area of Science:

• Cognitive Neuroscience
• Neurophysiology
• Psychology

Background:

• Stop-signal reaction time (SSRT) measures inhibitory control, with research often focusing on late event-related potential (ERP) components like P300.
• Evidence for N200 involvement is mixed, and the role of early ERPs (<200 ms) in SSRT remains under-explored.

Purpose of the Study:

• To investigate the contribution of both early and late ERP components, across successful and failed stop trials, to individual differences in SSRT using machine learning.
• To identify spatio-temporal ERP features predictive of SSRT.

Main Methods:

• Machine learning models were applied to high-resolution ERP data (64 electrodes, 4 ms resolution) from 148 participants during stop-signal tasks.
• Models analyzed both successful and failed stop trials to predict measured SSRT.
• Findings were validated on an independent dataset of 97 participants.

Main Results:

• A predictive model incorporating data from both successful and failed stops achieved significant cross-validated prediction of SSRT (r=0.32).
• Predictive ERP features included spatio-temporal activity in right frontal (N200 overlap) and frontocentral (P300 overlap) areas, as well as early ERP activity (<200 ms).
• The right lateralized N200 and P300 (peak and slope) were significant predictors, and results were reproducible across datasets.

Conclusions:

• ERPs from failed stop trials contain relevant information for predicting SSRT.
• Both early (<200 ms) and late ERP components contribute to individual differences in inhibitory control.
• The findings underscore the complexity of the neural mechanisms underlying SSRT and suggest novel electrophysiological markers.