Reward versus cost token systems: an analysis of the effects on students and teacher.

The effects of reward and cost token procedures on the social and academic behavior of two groups of elementary special-education students were assessed using a reversal design. Behavioral observations of three target subjects in each group revealed that both procedures were about equally effective in reducing rule violations and off-task behavior. Records kept on the daily arithmetic performance of all subjects showed that output doubled in both groups during the token phases, although accuracy remained unchanged. When students were allowed to choose either contingency, no pattern of preference was established. Small differences were found in teacher behavior: the reward procedure led to an increase in approval comments but cost procedures produced no changes in teacher behavior.