The suppression of a rewarded response receiving delayed punishment was examined when a stimulus, previously paired in conditioning trials with punishment, was introduced briefly with the response. Three groups of 20 alcoholics were employed. All received preliminary stimulation by a tone and punishment (shock), but only Group E received these stimuli in a conditioning paradigm. An instrumental response was subsequently trained in all Ss under immediate reward (money). When performance reached criterion, delayed punishment also was administered for this response. During these punished trials, the tone occurred briefly, immediately following the rewarded response for Groups E and C1. The remaining group (C2) received no tone on these trials. The response was more quickly suppressed in E than in C groups, and the two C groups did not differ in response suppression. The evidence was interpreted in terms of classical conditioning principles, and some practical implications of this finding were considered.