Extending a Matching-Law Based Account of Extinction Bursts to Humans
摘要
Suspending the reinforcement contingency maintaining a response can sometimes lead to a temporary increase in the local response rate, duration, or intensity for that response. This is referred to as an extinction burst. Shahan (2022) recently developed a quantitative theory of extinction, based on the Temporally Weighted Matching Law (TWML), which posits that extinction bursts are governed by behavioral processes tied to choice and reallocation. We sought to evaluate the generality of this matching-based account of extinction bursts using behavioral data from six individuals with intellectual and developmental disabilities. Although the TWML-based model of extinction bursts did a fair job of accounting for general trends in both human-operant and clinically relevant human behavior obtained in varied settings, it performed less well under some conditions. We discuss how convergence settings and parameter estimation procedures influenced model fits, highlighting methodological considerations for applying the TWML account of extinction bursts outside of basic laboratory settings.