Average reward as a determinant of Sperformance in differential conditioning* A total of 68 albino rats, distributed across three experiments, received differential reward magnitude training in a nonchoice brightness discrimination apparatus. The results showed that performance to a given amount of S-reward varied between groups that received the same average reward in the runway situation, and between groups that received the same stimulus-specific incentives but different amounts of average reward in the situation.

No abstract

Three groups of rats (C, D, and E) Previous runway differential conditioning studies (cf. Bower, 1961; Ludvigson & Gay, 1966) have shown that the speed with which a rat responds to astimulus associated with a specific reward magnitude is affected by responding to stimuli associated with areward of a different magnitude. Among such discrimination "contrast effects" (CEs), the observation of a negative S-CE, i.e., that the speeds of discrimination Ss in the alley associated with small reward (S-) are depressed relative to those of nondiscrimination Ss which always receive the identical small reward, is of particular interest in the present study.In studies demonstrating a negative S-CE, discrimination (D) and control (C) Ss receive identical treatment on half of the daily trials. Thus, Ss in both conditions receive a small reward (e.g., one pellet) in one discriminandum (e.g., black alley) on, for example, two of four daily trials. The differential treatment between groups is administered on the remaining two trials of the day. D conditions receive a larger (e.g., 12-pellet) reward contingent upon traversing a different discriminandum (e.g., white alley). C conditions receive two additional one-pellet runway trials, either in the black (Bower; 1961) or white (Ludvigson & Gay, 1966) alley. In these studies, the observation that D-condition speeds in the one-pellet discriminandum are depressed relative to those of C conditions may simply reflect the fact that Ss in the D conditions experience relatively large rewards while Ss in the C conditions do not. Altematively, this S-CE may reflect some more specific aspect of the differential treatment of D and C conditions, e.g., the administration of relatively large reward contingent upon runway traversal. The present study was concemed with whether experience with large reward, independent of the conditions under which this reward is obtained, is sufficient to depress runway speeds to a sm aller reward magnitude. METHOD The Ss were 24 experimentally naive, male albino rats approximately 100 days old at the beginning of the experiment.The runway apparatus consisted of a mid-grey 12-in. start box (SB) and two 42-in. alley-goal sections which were movable to permit alignment of either one with SB. One alley-goal section was painted black, the other white. The inner width of each section of the runway was 4 in., and the height 3% in. throughout. A transluscent Plexiglas top covered all units. An opaque door separated SB from the alleys and transluscent guillotine-type retrace dOOfS separated goal from Psydlon. Sei., 1969, V 01. 14 (1) alley sections. Photocell-c1ock circuitry provided traversal times over two consecutive 12-in. segments of the alley, beginning at the start dOOf.Fourteen days prior to the first day of training (Day 15), all Ss were placed on a 23-h food-deprivation schedule which was maintained throughout the study. On Days 5-14 Ss were taken from their individual horne cages and handled for about 2 min each. On Days 13 and 14, Ss were also placed in ind...

No abstract

EGGER, M. D. & MILLER, N. E. When is a reward reinforcing? An experimental study of the information hypothesis.