The acquisition and reversal of a position habit as a function of incentive magnitude.

Effect of reward magnitude, percentage of reinforcement, and training method on acquisition and reversal in a T maze.

“…In particular, noncorrection 5s made more correct choices in acquisition, but did not run faster, for large than for small reward. This argues against Pubols' (1961) suggestion that the effect of magnitude on choices is mediated by the effect on speeds.…”

Acquisition and reversal of a spatial response as a function of sucrose concentration.

“…The interaction suggests that a significant portion of the stimulus complex to which the spatial response is associated arises as feedback from the anticipatory goal response, so that with changes in the goal response there is a greater generalization decrement of the now erroneous response. Because of the major procedural differences between the Wike and Farrow experiment and the present one, it is likely that the previous failure to observe the effect (Pubols, 1961) was due not to procedural differences but to reinforcement magnitudes being not sufficiently disparate.…”

“…It has been concluded (Pubols, 1961) that reinforcement magnitude affects only concurrent performance and not subsequent reversal performance in spatial-discrimination learning. However, Wike and Farrow (1962) demonstrated an indirect effect of acquisition magnitude on reversal performance, which was, namely, that 5s for which the magnitude changed from acquisition to reversal learned more rapidly than 5s for which the magnitude remained constant.…”

Effect of a ready signal on the latency of voluntary responses in eyelid conditioning.

“…The interaction suggests that a significant portion of the stimulus complex to which the spatial response is associated arises as feedback from the anticipatory goal response, so that with changes in the goal response there is a greater generalization decrement of the now erroneous response. Because of the major procedural differences between the Wike and Farrow experiment and the present one, it is likely that the previous failure to observe the effect (Pubols, 1961) was due not to procedural differences but to reinforcement magnitudes being not sufficiently disparate. 60 trials with a ready signal and 60 trials without a ready signal were presented in a quasi-random sequence to each of 20 Ss.…”