This experiment was designed to investigate the influence of pacing, problem-solving instructions, and hypothesis testing on the conditioning of plural verbal responses. Ss were assigned to a Paced or Nonpaced group. Within each of the above conditions, Ss were assigned to one of four instruction subgroups: Task, Nontask, Random, and Control. Ss in all groups were merely told to say words. E said “good” following plurals emitted by the Task and Nontask groups. The Task group was also instructed to make E say “good” as many times as possible. “Good” was presented randomly for the Random condition, and nothing was said by E during the control treatment. The results indicated a significant increase in the number of plurals emitted by the Task and Nontask groups over trials but no statistically significant pacing effects.