Spatial contiguity facilitates Pavlovian second-order conditioning.

Abstract

Two experiments investigated the effects of spatial contiguity upon the formation of second-order conditioning in pigeon subjects. Experiment 1 used an autoshaping procedure to pair two visual stimuli, S2 and S1, after S1 had previously been paired with food. The resulting second-order conditioning of S2 was superior when both stimuli appeared on the same response key within a trial, compared with their appearing on different keys. Experiment 2 found a similar importance of spatial contiguity between S2 and S1 in a conditioned suppression paradigm. In addition, consistently presenting S2 and S1 in the same spatial location produced superior conditioning compared with varying their spatial relation from trial to trial. The design of these experiments was such as to imply that spatial contiguity facilitates performance by improving the formation of association rather than by promoting stimulus generalization or pseudoconditioning. Moreover, the observation of a facilitative effect of spatial contiguity between S1 and S2 in two different paradigms that use qualitatively different unconditioned stimuli and evoke different responses implies some generality for these findings. Consequently, these results suggest that spatially contiguous stimuli are especially associable in Pavlovian conditioning paradigms.

Coevolution of learning and data-acquisition mechanisms: a model for cognitive evolution

“…In practice, however, the first problemwhat to learn-is equally significant. Psychologists and animal trainers are well aware of the fact that the stimulus to be learned must be close to the reward, in time or in space, in order to be learned [34][35][36][37]. In terms of our model, this closeness requirement is not merely a technical constraint on the neuronal system, but an adaptive part of the data-acquisition mechanism.…”

Evaluating conditioning of related and unrelated stimuli using a compound test

“…When the S 2 and S 1 occur at different locations, the processing of S 1 may be delayed, potentially reducing conditioning. Although Rescorla and Cunningham (1979) attempted to deal with this difficulty in a second experiment using fear conditioning, which promotes withdrawal from rather than approach to S 2 , the present experiments do not have that feature.…”

“…However, Rescorla and Cunningham (1979) failed to conduct the sort of compound test used in Experiments 1 and 2 of the present article. This permits a reinterpretation of their results in terms of competing tendencies at the time of test.…”

“…The earliest discussions of associative learning frequently saw it as promoted not only by temporal but also by spatial contiguity. As Rescorla and Cunningham (1979) pointed out, however, temporal and spatial contiguity play quite different roles in Pavlovian conditioning. The occurrence of two stimuli in temporal contiguity is one of the defining circumstances of Pavlovian condition- edness promotes associative learning.…”

“…Experiment 3 examined a case related to Experiment 2, in which the paired stimuli are similar by virtue of their occupying the same spatial location (Rescorla & Cunningham, 1979). In each case, the question was whether pairing related stimuli leads to greater associative change or instead builds on greater prior associative strengths.…”