The error-related negativity (ERN) and post-error positivity (Pe) components of the event-related potential (ERP) are relatively stable over time. The current study further assessed the temporal reliability of ERN and Pe amplitudes for random samples of 2 to 14 trials per participant and the grand mean over a 2-week retest interval. In a replication of previous results, intraclass and zero-order correlations revealed moderate to good temporal stability for participants' (N=20) grand mean ERN and Pe component amplitudes. Adding trials increased test-retest reliabilities; however, the temporal stability of ERN and Pe amplitudes with 14 or fewer trials were modest at best and considerably lower than that for the grand means. Overall, data support the temporal stability of grand-mean ERN and Pe amplitudes and suggest that more than 14 trials are needed to include in ERN and Pe averages for adequate test-retest reliability.
Copyright © 2010 Society for Psychophysiological Research.