Adam S. Bristol

OVERSHADOWING AND LATENT INHIBITION COUNTERACT EACH OTHER : SUPPORT FOR THE COMPARATOR HYPOTHESIS

In 4 conditioned lick suppression experiments with rats, the combined effects of latent inhibition treatment followed by overshadowing treatment were assessed as a test of the comparator hypothesiss (R.R. Miller & L.D. Matzel, 1988) explanations of overshadowing and latent inhibition. Experiments 1 and 2 confirmed the prediction of the comparator hypothesis that combined latent inhibition and overshadowing treatments attenuate the response deficit produced by either treatment alone. Furthermore, consistent with the comparator hypothesis, posttraining changes in the associative status of the putative comparator stimulus altered responding to the target conditioned stimulus (Experiment 3), and switching contexts between latent inhibition and overshadowing treatments (Experiment 4) eliminated the interaction between the latent inhibition and overshadowing treatments.

Biological significance attenuates overshadowing, relative validity, and degraded contingency effects

Miller and Matute (1996) showed that blocking is attenuated when the blocked conditioned stimulus (CS) is “biologically significant” (i.e., when the CS has the potential to elicit vigorous responding of any kind). To the extent that blocking is representative of cue competition, this finding suggests that biological significance protects CSs against cue competition effects in general. In the present experiments, we tested this possibility by examining the influence of biological significance of CSs on other examples of cue competition, namely, overshadowing, the relative stimulus validity effect, and the degraded contingency effect in rats. In Experiment 1, we found that intense auditory stimuli induced transient unconditioned lick suppression, thereby indicating that intense sounds were of high inherent biological significance. In Experiment 2A, we found that cues with high inherent biological significance were protected from overshadowing. In Experiment 2B, this finding was extended to cues with high acquired biological significance, which was obtained through prior pairings with a reinforcer of the valence opposite to that used in the overshadowing treatment. In Experiments 3 and 4, we found that cues with high inherent biological significance attenuated the relative validity effect and the degraded contingency effect, respectively. These results lend support to the view that biological significance (inherent and acquired) protects stimuli from cue competition effects, a finding that is problematic for many contemporary theories of learning.