Alcoholics and controls did not differ in error type, accuracy, or RT. Both groups produced ERNs following incorrect responses, but these negative potentials were larger for the alcoholics than the controls. Following correct responses, alcoholics also produced larger negative amplitude CRNs than controls. These electrical potentials index conflict monitoring processes poised to trigger control mechanisms that implement corrective or compensatory actions (Botvinick et al. 2001). Compensation following an error has traditionally been observed as a tendency to slow down (Rabbitt 1966; Gehring et al. 1993), referred to as post-error slowing. In our task, both the alcoholics and the controls showed a trend toward slowed RTs following errors, thus providing evidence that conflict monitoring control mechanisms were engaged in both groups. In addition, larger negative CRN amplitudes in the alcoholics were associated with longer RTs for all trial types (error, correct, post-error, and post-correct trials). While the study did not involve a direct test of impulsivity, it could be hypothesized that impulsivity would be reflected in a speed/accuracy trade-off in the reaction time task with high levels of impulsivity being associated
