Maintenance of appropriate behavior in a token program

Following a baseline period, a token reinforcement program was instituted in three resource-room classes for very disruptive children where the children spent 1 hour per day. The only reinforcers used in the token program were those available to any classroom teacher, viz., free time in an activity area. The effects of the token program were evaluated in the three resource-room classes as well as in the homerooms of all students participating in the resource-room program. Following 4 months of the token program, a return to baseline was made in one resource classroom. As assessed by standard classroom observations, the token program was effective in reducing disruptive behavior in all three resource-room classes; and when it was withdrawn in one resource-room class, the disruptive behavior remained at a very low level in the resource room. Ratings by teachers in the homerooms showed a significant reduction in disruptive behavior, but recordings by homeroom observers showed no such reduction. Possible reasons for maintenance of low rates of disruptive behavior after withdrawal of tokens were discussed in terms of the type of reinforcers used.