Psychology ETDs

Publication Date



A growing body of literature indicates that animals and children tend to prefer situations which provide for a greater number of choice alternatives. In the present work, the problem was approached through the investigation of preference for larger or smaller response classes in a free-operant choice procedure. The method of assessing preference in the choice situation was afforded through the well-documented matching law of concurrent reinforcement schedules, which generally states that the ratio of responses or time allocated to each of two independent and concurrently-available schedules will equal the ratio of the reinforcements obtained on those schedules. A systematic and asymmetric deviation from matching would indicate a response bias to one of the two schedule alternatives, in that a greater proportion of responses or time would be allocated to that schedule than the obtained reinforcement ratio would predict. Twelve pigeons were food deprived and run on a series of five concurrent variable-interval variable-interval schedules of grain reinforcement. The schedules were programmed by a changeover key method, and the alternative concurrent schedules (red-illuminated and green-illuminated) were each alternately available through pecks on a changeover key. Either schedule was programmed on one of two schedule keys with position variable over reinforcements. Response and time allocation ratios were taken upon stable performance at each of the five concurrent schedule values as a function of obtained reinforcement ratios, and these data yielded straight-line baseline matching functions for each bird. Similar procedures were followed in the experimental phase, with the exception that for a random six of the birds (Group E), a second illuminated and operative response key was added to the green-signalled schedule alternative, with the single red-illuminated key available as before in the concurrent schedule. Hence, in the experimental phase, the number of alternative responses capable of being reinforced was increased for the green-signalled schedule relative to the red. For the remaining six birds (Group C), only one of the two green-illuminated keys was operative, the position of which varied unpredictably on the two green keys over reinforcements. The latter group served to control for the effects of preference for number of illuminated keys. A comparison of straight lines fitted to response or time allocation data for the baseline and experimental phases revealed the following results. (1) An increase in variability (departure from linearity) was observed across the experimental phase conditions relative to the baseline phase. (2) Although nearly all of the slopes of the fitted lines were less than unity, termed undermatching in the literature, there were no systematic changes in slope between the baseline and experimental phases. This finding brings into question a discrimination hypothesis of undermatching that has been proposed in the literature. (3) A systematic response bias was found toward the single-key (red) schedule alternative in Group E as evidenced by a shift in the y-intercept of the experimental function relative to the baseline matching function across birds. The effect was more pronounced on the response allocation than time allocation measures, and no such bias was observed in Group C on either measure. This finding is contrary to the literature on preference for alternatives, but may be interpretable through the observation that variability of response location was controlled in the single-key schedule alternative, while strong idiosyncratic position preferences developed in the two-key alternative for Group E as compared to Group C. Hence, it is possible that it is variability of responding, rather than the availability of alternatives per se, that governs preference in such situations.

Degree Name


Level of Degree


Department Name


First Committee Member (Chair)

Douglas Peter Ferraro

Second Committee Member

Frank Anderson Logan

Third Committee Member

G. Robert Grice

Fourth Committee Member

Joseph Anthony Parsons



Document Type


Included in

Psychology Commons