Reinforcement Magnitude and Response Rate: A Bead Study

Introduction

This paper explores the phenomenon of reinforcer magnitude and its effect on response rate. It aims to determine whether a correlation exists between the magnitude of reinforcement and participants' rate of response, and further, to ascertain the nature of that relationship — specifically, whether a stronger reinforcer magnitude produces a longer response rate. Such a relationship would be causal in nature.

The primary learning principle studied in this research is the magnitude of reinforcers. There is an abundance of research indicating that reinforcers are effective mechanisms for promoting learning or sustaining participant engagement. Reinforcers are a means of emphasizing a particular type of behavior; they can include either rewards or deterrents such as punishments. The other primary concept examined here is response rate.

For the purpose of this paper, response rate is defined as the length of time a subject is willing to respond to instructions or a call to a particular behavior. The study seeks to determine whether response rate can be influenced by reinforcers and, specifically, by the magnitude of those reinforcers. As noted, a bevy of existing research demonstrates a correlation between reinforcer magnitude and response, though most of that research focuses precisely on the relationship between reinforcer magnitude and response rate.

Literature Review

Operant conditioning is a form of learning in which a stimulus determines whether a behavior increases or decreases. Learning is obtained through the effects of consequences. Numerous studies have been conducted on the effect of reinforcers, emphasizing the levels at which reinforcement is administered. These studies suggest that the magnitude of a reinforcer has a substantial impact on response rates (Trosclair-Lasserre, Lerman, Call, Addison, & Kodak, 2008). It could also be said that the magnitude of a reinforcer can lead to either an increase or the extinction of a behavior.

In order to study the effects of irregular and continuous reinforcement on pecking responses, a small group of pigeons were placed in a Skinner box pecking apparatus. They were treated under two different conditions in which the duration of food presentation varied from 2 to 5 seconds, with intervals presented in a random manner. The effect of reinforcer magnitude on response rate showed a significant difference across conditions: animals responded more frequently in the 5-second condition than in the 2-second condition, though a large degree of variability existed from subject to subject.

Reinforcement magnitude (RM) is the dimension in which a reinforcer is presented — its duration, quantity, or intensity. Previous studies suggest that RM has a substantial influence on how both humans and animals respond, and that it plays an important role in the treatment of problem behaviors. Nicole M. Trosclair-Lasserre conducted a study in which three out of four children with behavioral problems demonstrated higher response rates when the reinforcer — in this case a tangible item or social interaction — was presented for a longer period of time. Prior to that study, a preference assessment was conducted to identify what each child considered a reinforcer. Under a progressive ratio schedule, the majority of responses occurred during the long reinforcement condition rather than during the short or no-reinforcer conditions. Other researchers have reported similar findings, though alternative explanations and potential sources of bias must also be considered.

In a study conducted by Aaron J. Brownstein and Stanley S. Pliskoff (1968), three adult pigeons served as subjects. Two schedules of reinforcement were simultaneously available during training, with concurrent schedules programmed by a changeover-key procedure. The main purpose of this study was to determine the effects of relative reinforcement rate and changeover delay in response-independent concurrent schedules of reinforcement. Results showed no meaningful change in behavior across conditions.

Collectively, these studies demonstrate the importance of the reinforcement effect in behavioral research. Each study suggested that the dimension or magnitude of a reinforcer has a significant impact on the dependent variable under examination. It is clear that reinforcement is an effective approach to both behavior modification and learning. For instance, Jenkins and Clayton (1948) demonstrated that reinforcement magnitude has a highly significant effect on the response rate of pigeons. Their findings also suggest that superior learning performance or preference is obtained when a reinforcer is present for a longer period of time.

The effect of reinforcement quantity on response rates was the central focus of the present study. Multiple trials were conducted in order to obtain reliable results. The main purpose was to determine whether the magnitude of a reinforcer would influence the amount of time a participant spent stringing beads. Based on prior research, a strong positive correlation and a significant difference in results were expected, given that reinforcement magnitude is projected to have a measurable effect on response rate.

Methods

Participants were children between the ages of 6 and 11 recruited from a university-affiliated elementary school. Students who were enrolled in special education programs (commonly referred to as "resource" classes in many educational institutions) were excluded from the study. Participants were allowed to leave class individually for no more than 10 minutes at a time while taking part in the experiment.

The target behavior upon which the study was based was the response rate of participants. That response rate was measured in terms of how many beads students were able to string together, which was largely determined by the reinforcer magnitude. In this study, reinforcer magnitude was operationally defined as candy, which children received for successfully stringing beads in sets of 15.

The dimension of measurement included the successful completion of bead-stringing in sets of 15, as well as the number of candy pieces participants received for completing each such set. The target behavioral goal was for students to desire to continue stringing beads in order to receive additional candy.

Materials used in this study included string and a large quantity of beads for participants to string together. The reinforcing stimuli consisted of individual pieces of candy in the form of Jolly Ranchers. Response measurement was based on a fixed-ratio (FR) 15 schedule of reinforcement, in which students received a colored token for every 15 beads strung together. Token colors varied across trials and corresponded to different amounts of candy: red tokens were redeemable for one piece of candy, blue tokens for three pieces, and green tokens for six pieces.

The primary variable was the color of the token presented after a participant successfully strung together a set of 15 beads. Because token color determined the amount of candy received, this variable was instrumental in assessing whether the number of bead sets produced — which also directly corresponded to the number of trials participants chose to engage in — was related to the amount of candy they received.

The candy, delivered via tokens, functioned as the reinforcer. The color of the token — with green tokens, redeemable for six pieces of candy, representing the highest magnitude — served as the measure of reinforcer magnitude. The only other variable was the number of times participants wished to continue the trial. Participants could choose to stop at any time or to continue as long as they wished.