Sustaining Behavior with Conditioned Reinforcement as the Only Response-Produced Consequence

Abstract

Food mas intermittently presented to pigeons on a responseindependent basis. Concurrently, pecking at a key was intermittently reinforcecl with the presentation of stimuli which accompanied the food presentation, but not the food. Under these conditions, key-pecking was maintained indefinitely at moderate rates and control procedures demonstrated that the results could only be accounted for on the basis of conditioned reinforcement.Conditioned reinforcemenc is often employed as an explanatory concept in the interpretation of behavioral events and is usually assigned a major role in accounting for the maintenance of behavior which is temporally and/or topographically separated from unconditioned reinforcement. As a consequence, a large number of experimental studies have focused Lipon demonstrating and assessing the conditioned reinforcing function of stimuli. W i t h few exceptions, however, these studies have employed an extinction procedure as the experimental paradigm. I n general, extinction paradigms involve ( a ) pairing a neutral stimulus with an unconditioned reinforcer and ( b ) subsequently testing the ability of the originally neutral stimulus to reinforce behavior after [he unconditioned reinforcer has been removed from the experimental situation. These studies have generally demonstrated that a neutral stimulus can acquire the power to reinforce behavior by being paired with an unconditioned reinforcer [see reviews by Miller (1951), Myers (1958), and Kelleher and Gollub (1962]. They have also indicated, however, that this acquired reinforcing power diminishes rapidly in the absence of unconditioned reinforcement.The generality of the latter finding could lead one to question the emphasis placed upon the role of conditioned reinforcemenc in the maintenance of behavior. However, perhaps what should be questioned is the emphasis placed upon the role of the extinction paradigms themselves as the mechods of choice for studying conditioned reinforcement. The praccice of assessing the conditioned reinforcing function of stimuli in the absence of unconditioned reinforcement assures the elimination of the possible confounding influences of unconditioned reinforcement on behavior. The re~noval of unconditioned reinforce---

Autoshaping pigeons’ keypecking with a conditioned reinforcer

“…The records for the response component could best be characterized as grainy with occasional long pauses. This failure to obtain scalloping when using a conditioned reinforcer has also been reported by Zimmerman & Hanford (1966), Thomas (1969), and Thomas & Johanson (1970).…”

“…Other techniques which maintain behavior for long periods of time with conditioned reinforcers, such as the chain schedules used by Ferster & Skinner (1957), the concurrent free-food procedure of Zimmerman & Hanford (1966), the multiple schedule technique of Thomas & Johanson (1970), and higher order schedules (Kelleher, 1966), all necessitate the training of the Ss to respond for primary reinforcement and then switching them over to conditioned reinforcement. The result of using both conditioned and primary reinforcement is that the behavior maintained by the conditioned reinforcer may be affected by response induction, or possibly even by a contrast phenomena from the primary reinforced behavior.…”

EFFECTS OF CONDITIONED REINFORCEMENT FREQUENCY IN AN INTERMITTENT FREE‐FEEDING SITUATION1,2

“…Four adult, male, White Carneaux pigeons, three of which (76R, 77R, and 40B) were used in the Zimmerman and Hanford (1966) study, served. A fourth (23B) had a 1.5-yr history on several chained schedules of food reinforcement and a more recent five-month history on the single-key multiple-schedule procedure.…”

“…Zimmerman and Hanford (1966) recently described a singlekey procedure in which access to food was automatically provided via a variable-interval tape if no peck was emitted for at least 6 sec. Concurrently, pecking at the single (blue) key intermittently produced conditioned reinforcement (the 0.5-sec magazine cycle with the shutter in its resting position).…”

“…Infrequent magazine failures, which might have permitted access to food during or after these cycles, might have accounted, at least in part, for the long-term maintenance of behavior which produced the stimuli associated with food. To eliminate the possible contribution of such failures, the magazines were modified (Leckrone, Zimmerman, and Hanford, 1966). The aperture providing access to the grain was covered with a transparent lucite shutter.…”