AMOUNT OF REINFORCEMENT AND FREE‐OPERANT RESPONDING¹

Abstract

Several years ago, we began a series of studies which we hoped would allow us to specify some of the important parameters of food reinforcers. Our first aim was the development of a method which would allow investigation of these factors in the individual organism. Hav-ing the benefit of the observations of Jenkins and Clayton (1949) and Guttman (1953), we assumed that rate of response under an interval reinforcement schedule would provide a datum sensitive to differences in reinforcing stimuli. We used pigeons, because we hoped to separate such potentially important factors as amount of food per se and number of pecks required to obtain the food. Our early efforts were uniformly disappointing: birds were run under Fl or VI conditions, with a given amount of food per reinforcement (each amount with an associated key color), until their day-to-day rates were relatively stable;3 the amount of reinforcement was then changed and the process repeated. Comparisons of the final rates achieved with each amount failed to show the expected differences, even though both amount of food and amount of consummatory responding were varied over rather wide ranges. Differences were noted, however, with four types of procedures: (1) The very first minu,tes of a daily session, prior to the delivery of the first one or two reinforcements, frequently produced rates correlated with the amount of food which, in past sessions, had been paired with the present key color. The disciminative stimulus of key color thus seemed to be controlling the response rate at the start of a session. (2) After many sessions in which one amount of food was used on odd-numbered days and a different amount on evennumbered days, rate differences were obtained during an extinction session in which the key colors formerly paired with each amount of food were presented alternately for 2-minute periods.4 (3) Although no differences between two amount-of-reinforcement conditions were observed after stable Fl 2 behavior was reached, a shift to Fl 4 BO 2 produced clearcut differences which then gradually disappeared over four daily 2-hour sessions on this new schedule. (4) A procedure similar to the "probe-stimulus" technique (Ferster & Skinner, 1957) showed response rates during probes that were correlated with the amount of reinforcement previously paired with the stimulus now used as a probe. In this procedure, two amounts of reinforcement, each with a correlated key color, were presented in alternate daily training sessions until day-to-day variability was relatively low. At this point, no differences between average rates for the two reinforcement conditions were seen. On subsequent sessions, the key light associated with one amount of food was inserted for 2-minute periods while the other amount with its key light was used for the remainder of the session. Probing with the larger-amount stimulus yielded an immediate increase in response rate; probing with the smaller-amount light gave an immediate decrease in rate. These probes were repeated three times daily...

Rats don’t always respond faster for more food: The paradoxical incentive effect

“…Hatton and Shull (1983) found larger feedings to presage longer pausing than did smaller feedings on fixed-interval schedules-but only when they were intermixed, not blocked. These results are similar to those found by Keesey and Kling (1961) on variable-interval schedules, and by Perone and Courtney (1992) and Baron, Mikorski, and Schlund (1992) on ratio schedules. In the present study, pausing was longer after larger feedings (e.g., in all the phases of Experiment 3, median PRPs were 2.6 sec for the three-pellet conditions, 1.25 sec for the 2-pellet conditions, and 1.0 sec for the 1-pellet conditions).…”

Graded Differential Reinforcement: Response‐dependent Reinforcer Amount

“…In single-key, free-operant procedures, providing various amounts of the reinforcer seems to have little or transient effect on behavior (Catania, 1963;Keesey & Kling, 1961) fect (Lendenmann, Myers, & Fantino, 1982). Varying the amount of the reinforcer within a session may have the effect of decreasing rates immediately following delivery of increased amount (Staddon, 1970).…”

Effects of reinforcement magnitude during successive reinforcement-punishment cycles

“…The direct relationship obtained between maintained response rates and reinforcement magnitude suggests that insensitivity of operant response rates to shifts in reinforcement magnitude (e.g., Keesey & Kling, 1961;Catania, 1963) can be overcome by correlating magnitude shifts with other changes in experimental conditions. Repeated changes in experimental conditions,…”

“…However, reinforcement magnitude has been shown to affect single-manipulandum responding under stimulus probe, extinction or punishment procedures (Keesey & Kling, 1961;Shettleworth & Nevin, 1965;Ferraro, 1966). These findings may be generalized to suggest that non-contingent shifts in reinforcement magnitude will control free-operant response rates if correlated with any change in experimental conditions.…”