Universidad Nacional de Educación a Distancia (UNED)
Project Title: Temporal organization of behavior in the inter-reinforcement interval: a 'superstitious' chaining hypothesis
When a reinforcer is delivered intermittently, some behaviors occur and increase despite not having a contingency relationship with reinforcement, these behaviors are often referred to as adjunctive or induced behavior. Recently, it has been proposed that adjunctive behavior is operant (Killen & Pellon, 2013) and maintained by delayed reinforcement, showing sensitivity to contingent delays (Pellón & Blackman, 1987) and to reinforcement by contingent hastening of reinforcement delivery (Alvarez, Ibias & Pellón, 2016).
Given that adjunctive behavior is consequence sensitive and considering that explicitly reinforced behavior and adjunctive behaviors occur sequentially in the inter-reinforcement interval (adjunctive behavior typically occurs early in the intervals between reinforcements, followed by behaviors such as lever pressing and magazine entries), It has been suggested that implicit chaining (“superstitious”) might occur between them (Pellon, 2004).
Fernando’s research, under the mentorship of Ricardo Pellon, aims to test some potential implications of this hypothesis. In particular, the assumed implicit or "superstitious" chaining that would happen between the behaviors that occur in the inter-reinforcement interval.
Several criteria have been taken into consideration as plausible evidence of "superstitious" chaining, such as reduction in responses that occur early in the interval when a later response is not available (as the later response, or stimuli signaling transition, provides reinforcement of the first link), effect that should be smaller or absent when a delay is imposed between the earlier and later responses (as the delay will reduce reinforcer efficacy of the later responses on the earlier ones). Additionally, we expect an increase in the later responses when the operandum of the earlier response is absent, effect that should be bigger when a delay is not imposed between the earlier and later responses (as response distributions will be superimposed and the absence of the first response will allow for sooner expression of the later response); thus, observing an asymmetric effect depending on which response operandum is unavailable and depending on the proximity between response distributions. For the hypothesis to be plausible, these criteria should be fulfilled when there are explicit chaining contingencies and when there are not. The SABA grant will allow to complete the experiments necessary to put to test the hypothesis and improve our understanding of reinforcement and adjunctive behavior, allowing for the integration of adjunctive behavior coherently within a behaviorist framework.
Back to Master's Thesis and Doctoral Dissertation Grants
