Paper
A Quest for the Generalized Conditioned Reinforcer
Published Dec 1, 1955 · E. L. Wike, H. McNamara
Psychological Reports
4
Citations
0
Influential Citations
Abstract
In recent years the problem domain relating to the development and behavioral consequences of learned rewards and drives has become of critical importance to contemporary learning theories (2 , 5) . The purpose of the present studies is to determine the effects of a type of learned reward which Skinner ( 7 ) has termed a "generalized conditioned reinforcer." Such a token reward is established when a pattern of cues is correlated repeatedly with more than one type of primary reinforcement (7, p. 77), while a simple learned reward is conditioned by being associated with one kind of drive reduction. It has been hypothesized that generalized conditioned reinforcers are more potent sources of acquired reward than simple conditioned reinforcing cues, that their effectiveness is less dependent upon the momentary state of the organism (7, p. 77) , and that they possess greater resistance to experimental extinction ( 1 7 ~ . 96) . In brief, the investigation of these proposals regarding the generalized conditioned reinforcer necessitates two steps: first, a cue-conditioning stage; and second, a test stage in which the effectiveness of the learned rewards is assessed. In the first stage of Experiment I a distinctive set of cues, a gray goal box, is present when hungry and thirsty rats ( H T group) eat wet mash. Control Ss undergo the same feeding experiences except that one group (H group) is hungry but not thirsty, and the other ( T group) is thirsty but not hungry. Accordingly, in the H T condition two drives are simultaneously reduced in the presence of the goal box cues, while in the H and T groups only one drive is diminished. In view of the known interdependence between states of hunger and thirst it is quite possible that the control animals are motivated to a low degree in the case of the "satiated" drive and therefore experience multiple drive reduction, but to a lesser extent than Ss in the H T group. In the second stage of the experiment the acquired reward value of the goal box must be determined. This is done by placing the empty goal box at the end of an alley straightaway and observing the speed with which the runway is traversed. If a generalized conditioned reinforcer acquires greater reward, then we would anticipate that the H T Ss would have shorter running times than Ss in the H and T groups. One obvious limitation in the design of Experiment I should be pointed
Generalized conditioned reinforcers may be more effective and resistant to experimental extinction than simple conditioned reinforcing cues.
Full text analysis coming soon...