P300 and slow wave: the effects of reaction time quartile.

Event-related potentials (ERPs) were recorded from 25 adolescents in a modification of the odd-ball paradigm. On alternate blocks subjects were required to detect either a missing stimulus or a change in pitch, each of which occurred 17% of the time and was embedded in a series of background tone pips (66% occurrence). The study was designed to assess the relationship between P300 and Slow Wave elicited by two infrequent targets which differed in the amount of temporal uncertainty (and thus, equivocation) associated with them. Principal components analyses (PCAs) were used in an attempt to reduce overlap among components. Stimulus-synchronized (SSA) and response-synchronized (RSA) averages were computed for correct trials only in association with the first (Q1) and fourth (Q4) reaction time (RT) quartiles. Measurement of the SSAs replicated the results of Roth, Ford and Kopell (1978): P300 amplitude was larger in Q1 than in Q4, while Slow Wave amplitude increased in Q4 relative to Q1. In the RSAs for Slow Wave, only the frontally negative aspect remained larger in Q4 than in Q1, while the parietally positive component did not differ between quartiles. The PCA basis waves showed that the major portions of P300 and Slow Wave followed response execution, thus precluding their involvement in the discrimination process per se. These results support the functional dissociation of P300 and Slow Wave and the functional independence of the frontal and parietal aspects of Slow Wave.