Systematic manipulation of individual events in a truly random control in rats.

Abstract

Truly random sequences of tone, conditioned stimuli (CSs), and shock, unconditioned stimuli (USs), were given to groups of rats; then conditioning to the OS was measured using a conditioned suppression procedure. In Experiment 1, eliminating chance CS-US pairings by systematically removing CSs that overlapped USs weakened conditioning monotonically as a function of the number of paired CSs removed. In Experiment 2, systematically delaying early chance pairings, while holding constant the number of CSs, USs, and pairings, produced a nonsignificant weakening of conditioning. In Experiment 3, delaying pairings again produced a nonsignificant weakening of conditioning when the early nonpaired events were CSs, but significantly weakened conditioning when the early nonpairings were USs. The data suggest that each chance pairing in the truly random control produces an increment in conditioning unless "blocked" by prior USs alone.The truly random control is a procedure frequently used to control for nonassociative factors in Pavlovian conditioning (e.g., Ayres & Quinsey, 1970;Bull & Overmier, 1968;Rescorla, 1968). It is characterized by the presentation of conditioned stimuli (CSs) and unconditioned stimuli (USs) randomly and independently of each other so that CS onsets provide no information about the occurrence of US onsets. Because there is no contingency between the CS and US in such a procedure, it was originally thought that the procedure should produce little or no conditioning (Rescorla, 1967). However, in several recent experiments the truly random control has apparently produced very powerful conditioned effects (Benedict & Ayres, 1972;Kremer, 1971;Kremer & Kamin, 1971;Quinsey, 1971). The most recent of these studies, moreover, showed that an important determinant of the conditioned 'This research was supported by Grant BG-36982 from the National Science Foundation and by a grant from the Research Council of the University of Massachusetts. We wish to thank William Matthews, who collected the data in Experiment 3, and Dan Anderson for his valuable comments on the manuscript. We also wish to thank Western Washington State College, Bellingham, Washington, for its support during the preparation of the manuscript.a Requests for reprints should be sent to John J.

Local context and the comparator hypothesis

“…In Experiment 1, a target CS was trained under conditions in which unsignaled USs were interspersed with target CS-US trials. Consistent with a variety of other behavioral observations under similar training conditions (e.g., Ayres et al, 1975;Durlach, 1983;Rescorla, 1968), a deficit in responding to the target CS was observed. Experiment 1 further revealed that this deficit in responding to a target CS could be attenuated in two different ways.…”

A transfer of control test for contextual associations

“…A variety of experiments (e.g., Ayres, Benedict, & Witcher, 1975;Keller, Ayres, & Mahoney, 1977;Rescorla, 1972) have shown that the neutrality of a CSI and, presumably, the associated contextual excitation hold only asymptotically. Preasymptotically, both theory and data indicate that the CSI may acquire temporary associative strength due to random CS-US pairings.…”

Conditioned inhibition training of the competing cue after compound conditioning does not reduce cue competition

“…In its molarity, comparator theory was similar to contingency theory (Rescorla, 1967), a theory that attempted to predict behavior based on molar contingencies or correlations between CSs and USs, correlations calculated over long spans of time including multiple sessions. That view, however, was challenged by the fact that different groups of rats could receive molar correlations of zero and yet respond to the CS or not at the time of testing depending on the molecular details of trial sequencing at the time of training (Ayres, Benedict, & Witcher, 1975;Benedict & Ayres, 1972). For explanation, such results appeared to require a theory such as the Rescorla-Wagner model, in which the molecular ordering of trial types was crucially important.…”