|The Place of the Human Subject in the 21st Century Operant Laboratory
|Sunday, May 29, 2016
|4:00 PM–4:50 PM
|Zurich D, Swissotel
|Area: EAB/TPC; Domain: Translational
|Chair: Shrinidhi Subramaniam (West Virginia University)
|Discussant: Michael Perone (West Virginia University)
|CE Instructor: Shrinidhi Subramaniam, M.S.
Baron and Perone (1982) attributed what they described as a "paucity of laboratory research on human operant behavior" to several misconceptions about constraints on experimental research with humans. The intervening decades have seen great expansion of the number and variety of publications reporting experimental analysis of human behavior. Has human operant-conditioning research acquired the technical depth and topical breadth to address Baron and Perone's concerns? Through two empirical presentations, this symposium will provide a snapshot of contemporary laboratory human operant conditioning research. In two talks, speakers will present schedule-based work testing the generality of basic behavioral principles as they have been applied to experimental analysis of delay discounting and interval timing. Vanderveldt and Green will discuss the magnitude effect in delay discounting, a robust effect in nonhuman research that has been more elusive in human subjects. Subramaniam and Kyonka will present results of peak-interval experiments with humans in which screen color-interval correlations differed across conditions. Discussant Michael Perone will present his opinion of the quantity and quality of experimental human operant research today.
|Keyword(s): delay discounting, human operant, interval timing, reinforcement schedules
|Delay Discounting in Human and Nonhuman Animals: Some Similarities and Differences
|ARIANA MAE VANDERVELDT (Washington University in St. Louis), Leonard Green (Washington University)
|Abstract: Delay discounting is apparent when the subjective value of a reward decreases with the time until it receipt. Both human and nonhuman animals show discounting, which is well described by a hyperbolic function. Despite the pervasiveness of delay discounting across species, it has been claimed that animals discount delayed reinforcers orders of magnitude more steeply than do humans (i.e., animals are much more impulsive) and do not show a magnitude effect (i.e., animals do not discount larger delayed amounts proportionally more steeply than smaller delayed amounts). Whether these are true species differences or due to procedural differences will be discussed. For example, we employed a procedure in which humans discounted a real liquid reward that was consumed on each choice trial, a procedure more similar to that used with animals, rather than the more typical procedure in which humans discount monetary and/or hypothetical rewards. We find that humans now discount the delayed reward on the order of seconds. Other studies will be discussed in which comparisons between discounting by humans and animals will be presented.
Beyond Face Validity: Peak Procedures With Humans
|SHRINIDHI SUBRAMANIAM (West Virginia University), Elizabeth Kyonka (West Virginia University)
The peak-interval (PI) procedure is a well-established and widely used way of researching interval timing in laboratory animals, but has been used much less with human subjects. In a version introduced by Guilhardi and colleagues (2010), a target moved across a computer monitor at several velocities and participants fired shots at the center of the screen. The target was masked during 12-s PI trials. The distribution of shots was similar to PI response gradients from nonhumans. We used this PI procedure to isolate the relation between timing and misinformation. The correlation between screen color and interval duration (i.e., time to point availability) varied across conditions in Experiment 1 and groups in Experiment 2. Experiment 1 resulted in order effects that were prevented using a group design. When screen color was perfectly correlated with interval duration, PI distributions peaked at the target times. At intermediate correlations, differences between distributions from different screen colors were smaller, and when the correlation was 0, distributions converged. Color-interval correlation affected response strategy by altering the stimulus control. Regard for strong, experimentally imposed contingencies above face validity can overcome difficulty with complex human subjects in the operant laboratory.