The present study investigated the reliability of CNV distraction and rebou
nd effects, and their relation with reaction time. Twenty-four subjects wer
e presented three blocks of trials: (1) a control block-a fixed foreperiod
reaction time task consisting of a flash-tone-key press sequence; (2) a div
ided-attention block-randomly intermixed trials with and without a short-te
rm memory task (three visually-presented letters) in the S1-S2 interval (50
% of each): and (3) a second control block. In trials with the short-term m
emory task, subjects recalled the letters after the key press to the tone.
Compared to the control block, CNV amplitudes during trials with letters we
re significantly smaller and reaction times to S2 were significantly slower
(distraction effect). In contrast, CNV amplitudes during trials with no-le
tters were significantly larger (CNV rebound), but the reaction times were
again significantly slower. This dissociation of CNV rebound and reaction t
ime could provide an objective neurophysiological tool to probe attention f
unctions in both normal and clinical populations. (C) 1998 Elsevier Science
B.V. All rights reserved.