An Assessment of Token Value and Effectiveness

Abstract

Token systems are a commonly used treatment procedure to increase levels of a desired behavior (e.g., compliance) or decrease levels of an unwanted behavior (e.g., aggression) for a variety of different responses (Hackenberg, 2009). Token systems have been used with a variety of populations including, but not limited to, children diagnosed with disabilities, prisoners, and school children. Because token systems are commonly used, it is important to identify the reinforcing value of these systems to increase their effectiveness (Fiske et al., 2015). Fiske et al. (2015) used progressive-ratio schedules to compare the effects of paired tokens, primary reinforcers, and unpaired tokens on the number of responses completed by two children diagnosed with autism. The results showed for both participants that responding was variable during the paired token condition, stable during the primary reinforcement condition, and low during the unpaired token condition. The purpose of the current study was to conduct a systematic replication of Fiske et al. (2015) to extend the behavior-analytic literature on token reinforcement in applied settings. Specifically, we addressed several of the limitations of Fiske et al. (2015) by (a) including specific conditioning procedures and creating a controlled history of reinforcement with the token system, (b) including participants with and without disabilities to extend the generality of the findings, and (c) equating the magnitude of reinforcement across the primary and paired token conditions. The reinforcing value of tokens was evaluated by using progressive-ratio schedules to compare the efficacy of primary reinforcement, paired tokens, and unpaired tokens on the level of task completion.