Choice and Negative Reinforcement: The Effects of Amount, Delay and Probability
AuthorBonow, Jennifer A.
AdvisorGhezzi, Patrick M.
AltmetricsView Usage Statistics
In behavior analysis, self-control is defined as selecting a large, delayed reinforcer over a small, immediate (or less delayed) reinforcer while selecting the small-immediate reinforcer is termed impulsive (Ainslie, 1974; Rachlin & Green, 1972). This definition highlights the importance of the delay to, and magnitude of, reinforcement in choice making behavior. These two variables have been studied extensively in the self-control paradigm. A distinct but theoretically related area of research, discounting, investigates the interaction between magnitude and probability of reinforcement. Even though results of discounting studies suggest that the probability of reinforcement plays a large role in choice behavior (e.g., Rachlin, Logue Gibbon, & Frankel, 1986; Rachlin, Raineri, & Cross, 1991), there has been little investigation of this factor in the self-control research. The current study aimed to investigate the effect of probabilities of reinforcement of less than 1 on the choice-making behavior of adults. The first experiment examined selections between uncertain immediate and certain delayed reinforcers and the second experiment examined selections between uncertain large and certain small magnitude reinforcers. These experiments were done using negative reinforcement (the removal of a loud white-noise) during a preferred task (watching a DVD movie), as it has been demonstrated that humans are more likely to respond impulsively in studies employing negative, rather than positive, reinforcers (e.g., MacAleese, 2009; Navarick, 1982; Solnick, Kannenberg, Eckerman, &Waller, 1980). The data show that when probability of reinforcement is held constant, adult humans prefer immediate (Experiment 1) and large magnitude (Experiment 2) reinforcers and that reducing the probability of reinforcement can induce preference shifts in humans.