Preventing relapse after incentivized choice treatment: A laboratory model

Two experiments with rats examined relapse of an operant behavior that occurred after the behavior was suppressed by reinforcing (incentivizing) an alternative behavior. In the first phase, a target response (R1) was reinforced. In a treatment phase, R1 was still reinforced, but a new response (R2) was introduced and associated with a larger reinforcer. As in human contingency management treatments, incentivizing R2 this way was effective at suppressing R1. However, when R2's reinforcement was discontinued, there was a robust and immediate relapse to R1. Experiment 1 found that the strength of R1 during relapse testing was not different from that seen in a no treatment control. Experiment 2 found that relapse could nevertheless be reduced by presenting reinforcers not contingent on responding during the test. Either the reinforcer for R1 or the reinforcer for R2 (which were qualitatively different types of food pellets) were effective. The experiments introduce a laboratory method for studying relapse and how to prevent it after contingency management treatments, and suggest at least one treatment that discourages relapse. The incentivized choice paradigm differs from other models of relapse of operant behavior (e.g., resurgence, renewal, reinstatement) in that it does not focus on the return of behaviors that are inhibited by extinction.