Uncertainty Reduction and Conditioned Reinforcement

Abstract

Pigeons were exposed to a single-key sequential choice procedure in which the choice phase was followed by either a multiple fixed-interval schedule of food reinforcement where different key-lights indicated which FI was in effect or by a mixed fixed-interval schedule of food reinforcement where the same key-light was present for either FI. In the choice phase a scrambled series of brief periods of blue and of green illumination of the key were presented. A single peck to blue produced the mult schedule; a single peck to green produced the mix schedule. The variable manipulated was the relative proportion of FI 10-sec. components in the mult and mix schedules. Pigeons pecked blue more frequently than green for proportions other than 0.00 and 1.00. Preference for blue was greater for the range of proportions from 0.03 to 0.50 than for the range from 0.50 to 0.97. Such asymmetry of preference is a problem for the information account of conditioned reinforcement.