Generalization of Reinforcement Effects in a Token Program in the Home

Abstract

A behavioral management program was established in the home for a behavior problem child. When the program showed success for the targeted behaviors, observations were begun in an additional problem area: proper bedroom maintenance. This behavior was divided into four separate components. The mother instructed the child to carry out each of the component behaviors, and promised a reward for one, or promised no reward, depending on the experimental condition. As the amount of the reward varied, performance for all components varied, but no behavioral discrimination appeared among the reinforced and nonreinforced component behaviors. Compliance with the nonexperimental treatment program remained high throughout the study. Questionnaire testing showed precise verbal discrimination of the token reward contingency. The results show that the effects of token reinforcement programs may extend to behaviors other than those specifically targeted.