Three experiments were conducted to evaluate response allocation of stereotypy during free-operant and restricted-operant conditions. Five children with autism or related developmental disabilities participated in at least one and up to three of the experiments. In Experiment 1, the stereotypic response that emerged as most probable during a free-operant phase was restricted, and response allocation was again evaluated. The results for 3 participants showed that restricting the high-probability response was correlated with covarying reductions in a nontargeted stereotypy. In Experiment 2, the effect of environmental enrichment on response allocation was evaluated. One participant reallocated behavior to appropriate object manipulation, 1 participant showed no change in behavior, and a 3rd participant reallocated behavior to object manipulation only when the putative stimulus products of the object manipulation matched those of stereotypy. In Experiment 3, additional interventions were implemented to promote response reallocation. Results showed that both response restriction and reinforcement for object manipulation decreased stereotypy and increased object manipulation. Collectively, the results of these experiments point to a need for complex evaluations of interventions for stereotypy.