Understanding complex comorbid conditions of intellectual developmental disorder (IDD) is an NICHD priority.
Severe problem behavior (e.g., self-injurious behavior, aggression) of children with IDD is prevalent, potentially
dangerous, and negatively impacts social integration and quality of life. Function-based differential reinforcement of
alternative behavior (DRA) interventions reduce such behavior effectively, but treatment relapse often occurs when a
caregiver does not deliver reinforcement for the alternative behavior. Such relapse is known as resurgence. The
efficacy of DRA interventions in reducing severe problem behavior has long been theorized to be related to the same
basic processes that govern choice between concurrently available sources of reinforcement. However, quantitative
theories of such choice behavior (e.g., the matching law) have not previously been applicable to resurgence following
DRA because they could not account for situations in which one or both behaviors no longer produced reinforcement.
This project represents a collaboration between a basic scientist/theorist and clinical scientists with expertise in DRA
treatments to evaluate and translate the predictions of a novel choice-based theory of resurgence. This innovative
theory is built from two fundamental and well-established principles of reinforcement: (a) individuals allocate
proportionally more responding to responses that produce proportionally more reinforcement (i.e., the matching law);
and (b) the value of reinforcement decreases hyperbolically as it recedes further into the past. By combining these
two principles into a mathematical model, the theory accounts for resurgence in situations in which the only existing
quantitative theory of resurgence (behavioral momentum theory) has failed. The choice-based theory generates
novel predictions about the precise conditions under which resurgence might be expected to occur and how
treatments might be altered to reduce resurgence (e.g., extending treatment duration, dynamically adjusting schedule
thinning). This project empirically examines these and other novel predictions about variables of high potential clinical
relevance, but for which clear data are lacking (e.g., effects of different response efforts for the target and alternative
responses) and directly translates insights from the theory and basic research with animals in order to improve
treatment of problem behavior in the clinic and reduce treatment relapse. Our overarching hypothesis is that DRA
interventions informed by this novel theory will be less susceptible to resurgence of problem behavior: (a) during
periods when reinforcers are unavailable; (b) when reinforcer availability diminishes during reinforcement schedule
thinning; and (c) when other response or reinforcement parameters change, such as increases in response effort.
Thus, this project seeks to fill existing gaps in basic/theoretical knowledge about resurgence and represents a rare
close collaboration between basic and clinical scientists to better align basic research and clinical practice.