Choice With Uncertain Outcomes: Conditioned Reinforcement Effects

Abstract

Pigeons responded on concurrent chains with equal initial- and terminal-link durations. In all conditions, the terminal links of one chain ended reliably in reinforcement; the terminal links on the alternative chain ended in either food or blackout. In Experiment 1, the terminal-link stimuli were correlated with (signaled) the outcome, and the durations of the initial and terminal links were varied across conditions. Preference did not vary systematically across conditions. In Experiment 2, terminal-link durations were varied under different stimulus conditions. The initial links were variable-interval 80-s schedules. Preference for the reliable alternative was generally higher in unsignaled than in signaled conditions. Preference increased with terminal-link durations only in the unsignaled conditions. There were no consistent differences between conditions with and without a common signal for reinforcement on the two chains. In the first series of conditions in Experiment 3, a single response was required in the initial links, and the stimulus conditions during 50-s terminal links were varied. Preference for the reliable outcome approached 1.0 in unsignaled conditions and was considerably lower (below .50 for 3 of 5 subjects) in signaled conditions. In a final series of signaled conditions with relatively long terminal links, preference varied with duration of the initial links. The results extend previous findings and are discussed in terms of the delay reduction signaled by terminal-link stimuli.

Determinants of pigeons’ choice between certain and probabilistic outcomes

“…Overall, many of the results obtained fit well with both Mazur's (1989Mazur's ( , 1991 model and the local delayreduction approach proposed by Dunn and Spetch (1990). Mazur's model provides a more parsimonious account than that offered by the delay-reduction framework for the effect of varying BOTL duration in unsignaled procedures and the lack of effect of this same manipulation within signaled procedures.…”

“…Because between-subject variability is typically high, we cannot strongly argue against these predictions on the basis of the results presented here. It is noteworthy, however, that in each of the present experiments, as well as in those of Dunn and Spetch (1990), Kendall (1974Kendall ( , 1985, and Spetch et al (1990), mean choice proportions for the 50% side were above .5 in all signaled conditions with equal terminal links of 30 sec or longer. Furthermore, in a recent study (Belke, 1992;Belke & Spetch, 1994), 6 out of 8 pigeons displayed a strong preference for the 50% alternative in a variant of the present procedure in which they were forced to remain on the 50% side after they chose it until a food outcome occurred.…”

“…It may be that choice proportions reflect the joint contribution of the lower rate of primary reinforcement and the higher value of the 50% food terminal link, whereas rate of responding during the terminal links leading to food reflects only the latter. Dunn and Spetch (1990) and Spetch et al (1990) noted that results from the percentage reinforcement procedure were analogous to findings from serial autoshaping studies, in which responding to a stimulus that had an intermittent relationship to the occurrence of a stimulus associated with food was higher than responding to a stimulus that was always followed by a stimulus associated with food (Collins & Pearce, 1985;Collins, Young, Davies, & Pearce, 1983). However, Pearce and Collins (1987) used choice probes to test preference between stimuli that were intermittently or perfectly predictive of a stimulus associated with food, and found that their measures of preference did not mirror the difference in response rates found in serial autoshaping.…”

“…The contrast between the signaled and unsignaled results suggests instead a role played by the signals for food (i.e., the conditioned-reinforcing properties of the terminal-link stimuli). Along these lines, Dunn and Spetch (1990) and Spetch et al (1990) proposed a framework that borrows heavily from Fantino's (1969) delay-reduction hypothesis of conditioned reinforcement. This will be referred to as the "local delayreduction framework," both to distinguish it from the original hypothesis and to indicate that, as yet, it represents only an approach to understanding choice between certain and probabilistic outcomes, not a formalized model.…”

Designing Interventions That Include Delayed Reinforcement: Implications of Recent Laboratory Research

“…Indeed, unreliable reinforcement may be more typical in instructional and clinical situations than reliable reinforcement. Research has shown that nonhuman subjects are likely to prefer unreliable reinforcement to reliable reinforcement when conditioned reinforcers are presented during delay intervals preceding the availability of the unreliable reinforcement (Belke & Spetch, 1994;Dunn & Spetch, 1990). Little applied research has been conducted in this area, however.…”