Comparing Reinforcement Systems with Percentile Schedules

This study used percentile schedules of reinforcement – percentile schedule (K), percentile schedule with a dimension (K+) and a yoked control condition (C) – to examine the effects of giving extra reinforcement on the rate of tacting. Six undergraduate students participated in three schedules of reinforcement for fluency of correct tacting answers; they received reinforcement for meeting or exceeding criteria (K), received reinforcement for meeting and exceeding criteria and earned extra reinforcement when closer to the goal then the previous five answers (K+), and earned predetermined reinforcement through yoking for control (C). Celeration values were analyzed for three participants after participants engaged in low levels of errors. One participant had greater average celeration within timing values for the dimension of reinforcement schedule and yoked control and less celeration values for the reinforcement schedule without a dimension. One participant had similar celeration lines for the dimension of reinforcement schedule and the yoked condition than the reinforcement schedule without a dimension. One participant had greater celeration lines for the yoked condition, then the dimension of reinforcement condition, and lowest celeration value for the reinforcement schedule without a dimension. The remaining participants\u27 celeration values were not analyzed due to continued errors or low levels of tacting. Implications of the results and suggestions for further research are discussed