Abstract
Ss’ preferences when choosing between a pair of token rewards, one a known and the other an unknown value, were determined under two conditions, (1) the range of possible reward values was known from the outset of the experiment and (2) the range had to be learned. Analysis of choice probabilities and latencies indicated choices yielded almost optimal outcomes although there was some tendency to choose a non-optimal unknown alternative if it might possibly yield a large reward. Choices of known alternatives closer to the endpoints of the range of values were made more quickly even though choice probabilities were equal, and choices were made more slowly by Ss who had to learn the range of values.
Article PDF
Similar content being viewed by others
References
Katz, L. Effects of differential monetary gain and loss on sequential two-choice behavior. J. exp. Psychol., 1964, 67, 245–249.
Keller, L., Cole, M., Burke, C. J., & Estes, W. K. Reward and information values of trial outcomes in paired-associate learning. Psychol. Monogr., 1965, 79, No. 12 (Whole No. 605).
Myers, J. L., & Atkinson, R. C. Choice behavior and reward structure. J. math. Psychol., 1964, 1, 170–203.
Author information
Authors and Affiliations
Additional information
This work was carried out at the Institute for Mathematical Studies in the Social Sciences, Stanford University. Thanks are due to W. K. Estes who contributed the apparatus used in the study.
Rights and permissions
About this article
Cite this article
Cole, M., Calfee, R. & Keller, L. Reward structure and payoff preferences in human learning. Psychon Sci 5, 169–170 (1966). https://doi.org/10.3758/BF03328336
Published:
Issue Date:
DOI: https://doi.org/10.3758/BF03328336