A quantitative analysis of the behavior maintained by delayed reinforcers

Abstract

Random-interval reinforcement was arranged for a sequence of pigeon first-key pecks followed by second-key pecks. First-key pecks, separated from reinforcers by delays that included number of second-key pecks and time, decreased in rate as delays increased. Delay functions, or gradients, were obtained in one experiment with reinforced sequences consisting of M first-key pecks followed by N second-key pecks (M + N = 16), in a second where required first-key pecks were held constant (M = 8), and in a third where minimum delay between most recent first-key pecks and reinforcers varied. In each, gradients were equally well fitted by exponential, hyperbolic and logarithmic functions. Performances were insensitive to reinforcer duration and functions were consistent across varied random-interval values. In one more experiment, time and number delays were independently varied using differential reinforcement of rate of second-key pecks. Delay gradients depended primarily on time rather than on number of second-key pecks. Thus, reinforcers have effects based on earlier responses, not just the ones that produced them, with the contribution of each response weighted by the time separating it from the reinforcer rather than by intervening behavior. Situations where unwanted responses (e.g., errors) often precede reinforced corrects can maintain them unless designed to avoid such effects of delay.

The distribution of response bout lengths and its sensitivity to differential reinforcement

“…That is, reinforcement of bouts of a particular length seems to differentially strengthen those bouts relative to bouts of other lengths. The hypothesis that reinforcement operates on bouts and not on individual responses is consistent with evidence that bout initiations increase with rate of reinforcement according to Herrnstein's () hyperbola (Shull, ; see also Hill et al, ), with evidence that reinforcement also operates on responses that precede the one that produces reinforcement (Catania, ; Catania, Reilly, Hand, Kehle, Valentine, & Shimoff, ), and, to some extent, with Killeen's (1994) notion of response‐reinforcement coupling. However, one aspect of the present results seems inconsistent with the bout‐length‐as‐response‐class hypothesis: If reinforcement of bouts is expressed in a single population of bout lengths centered near the mean of reinforced bout lengths, why would there be two populations of bout lengths when there is only one ratio requirement (one response in VI 120‐s, five responses in tandem VT 120‐s FR 5)?…”

A Critical Review of the Support for Variability as an Operant Dimension

“…Several later experiments have shown that variability, normally measured as the dispersion of different sequences, are higher when a variability contingency is in effect, than when it is not (e.g., Denney & Neuringer, 1998;Kong, McEwan, Bizo, & Foster, 2019;Miller & Neuringer, 2000). This finding is highly reliable, and the interpretation of "direct reinforcement of variability" and "variability as an operant dimension" has been widely accepted (e.g., Catania, 2013;Doughty & Galizio, 2015;Dracobly, Dozier, Briggs, & Juanico, 2017;Lee, Sturmey, & Fields, 2007;Locey & Rachlin, 2013;Odum, Ward, Barnes, & Burke, 2006;Rodriguez & Thompson, 2015;Stahlman & Blaisdell, 2011;Stokes, 1995;Stokes, Mechner & Balsam, 1999;Ward, Kynaston, Bailey, & Odum, 2008). Nevertheless, a few researchers (Barba, 2015;Epstein, 2014;Holth, 2012;Machado, 1989Machado, , 1992Machado, , 1993Machado, , 1997Machado & Tonneau, 2012;Marr, 2012) have raised the question of whether direct reinforcement of variability is the most satisfactory view or whether the variability in these experiments is more effectively or pragmatically considered as a derivative of other procedural characteristics.…”

Contributions of Peter B. Dews (1922‐2012) to the experimental analysis of behavior

“…How a reinforcer influences antecedent behavior quantitatively has recently been further documented in an extensive set of experiments by Catania et al (). The thoughtful planning of their approach is comparable to the IRT research by Dews.…”

“…My definition of IRT is the time period since the previous response ending with a response. By including a terminal response in the definition, one can speak of reinforcing IRTs without it being jargon.3 How a reinforcer influences antecedent behavior quantitatively has recently been further documented in an extensive set of experiments byCatania et al (2015). The thoughtful planning of their approach is comparable to the IRT research by Dews.…”