Association for Behavior Analysis International

Nevin and colleagues introduced behavioral momentum theory in 1983 as a quantitative framework for understanding the response strengthening effects of reinforcers on operant behavior (Nevin, Mandell, & Atak, 1983). Behavioral momentum theory asserts that operant behavior is comprised of two independent processes—response rates and resistance to change—by making the analogy to the independence of velocity and mass in classical physics. With its influential ideas persisting for over three decades, its development has not been resistant to change. Behavioral momentum theory continually has been refined to account for an impressive array of data extending beyond resistance to change, including preference, compliance, the partial reinforcement extinction effect, and, most recently, relapse. This symposium includes three presentations focusing on the application and extension of behavioral momentum theory to resistance to extinction and relapse using reinstatement procedures. The architect and primary developer of behavioral momentum theory, Tony Nevin, will serve as discussant.

Relapse of maladaptive behavior is a common pitfall in a range of clinical and applied problems. Behavior momentum theory is a useful framework for understanding what makes behavior more persistent when it is decreasing. More recent extensions have examined behavioral persistence when behavior is increasing, such as relapse. Our experiment explored how different baseline rates of response-dependent food impact reinstatement of extinguished behavior when reintroducing food into the context. Seven pigeons pecked keys for food on a multiple variable-interval 30-s variable-interval 120-s schedule. After baseline, 6 sessions of extinction was followed by 3 sessions of response-independent food deliveries. This same series of conditions was then repeated. Baseline response rates were higher and extinguished more slowly in the component with the higher rate of food delivery. Response rates recovered to near baseline levels during response-independent food deliveries. The greatest recovery, proportional to baseline rates, occurred in the component with the higher baseline rate of food delivery. These results replicate and extend those showing greater reinstatement with higher baseline rates of food delivery but lower baseline response rates. Baseline reinforcer rate, whether delivered response dependently or response independently, determines not only resistance to change with decreasing operations but also with increasing operations.

According to behavioural momentum theory, resistance to extinction and relapse depend on the relative rates of reinforcement associated with training stimulus contexts. Other studies suggest that relapse with reinstatement depends on the reinstating conditions being arranged within the training stimulus context. In the present two experiments with 17 pigeons, we assessed whether relapse was a function of training reinforcement rates when reinstating food presentations were arranged either within or external to the training contexts. The context arranging higher baseline reinforcement rates consistently produced greater resistance to extinction but reinstatement effects were mediated by the context in which reinstatement occurred. Specifically, training reinforcement rates also were positively related with the degree of reinstatement when the reinstating food was arranged within the training stimulus contexts and in alternative contexts associated with a history of reinforcement. When arranging the reinstating food in alternative contexts associated with extinction, reinstatement effects were smaller and unrelated to training reinforcement rates. These findings suggest that relapse depends on the history of reinforcement associated with the context in which the relapse-inducing event occurs.

Behavioral momentum theory asserts that reinforcing two responses in the same context will mutually increase resistance to extinction and relapse by increasing the overall reinforcement rate within that stimulus context. Indeed, previous findings reveal that relative resistance to extinction between two responses is only differential when trained separately (i.e., multiple schedules), but do not differ when trained concurrently. The present study explores this further by directly comparing relative resistance to extinction and relapse of target responding trained either separately or concurrently with an alternative response. In all 6 pigeons, concurrently training the target and alternative responses produced greater target resistance to extinction and relapse with food reinstatement compared to training the responses separately. Furthermore, when combining the separately trained contexts during extinction, or separating the concurrently trained contexts, resistance to extinction and relapse still reflected the reinforcement rates established in the original training contexts. These data support the assertions of behavioral momentum theory suggesting that the overall reinforcement rate during training is the primary determinant of resistance to extinction and relapse.