A sensory preconditioning analogue was employed to separate signaling effects from motivational effects observed in appetitive-to-aversive transfer of training. Rabbits received appetitive conditioning (tone-water pairings) of the jaw-movement response, followed by aversive conditioning of the nictitating membrane response, during which water delivery served as the CS for paraorbital shock. When the tone was subsequently presented, only subjects exposed to both sets of pairings demonstrated conditioned jaw movement and nictitating membrane responses. The occurrence of both responses to the tone is inconsistent with the action of reciprocal inhibition between motivational states. The results are interpreted in terms of multiple mediators for transferof-training paradigms.Many influential theories of instrumental behavior as-. sume that there exist aversive and appetitive motivational states in the central nervous system which act reciprocally to inhibit one another (e.g., Konorski, 1967; Rescorla & Solomon, 1967). Each motivational state is thought to be activated directly by an unconditioned stimulus (US) or indirectly by a conditioned stimulus (CS) that has been paired with the US. The most conclusive evidence for competing motivational states (see Dickinson & Pearce, 1977) comes from classical-classical transfer paradigms in which a CS is initially paired with a US from one motivational state and subsequently paired with a US from the other motivational state (Bromage & Scavio, 1978;Krank, 1985; Scavio, 1974Scavio, , 1975Scavio & Gormezano, 1980). For example, Scavio (1974Scavio ( , 1975 has shown that conditioning the rabbit's nictitating membrane response to a tone CS by pairing the CS with a shock US will retard the development of a jaw-movement response to that CS when it is subsequently paired with an oral injection of a water US. Since Scavio also showed that nictitating membrane and jaw-movement conditioned responses (CRs) were statistically independent, the retarded acquisition of the jaw-movement conditioned response could not be attributed to peripheral incompatibility of responses. Hence, Scavio's results support a theory in which central motivational states inhibit one another.In contrast, when USs from opposing motivational states are directly paired with one another, conditioning appears to occur rapidly (Asratyan, 1965(Asratyan, , 1980Beritoff, 1965;. Dearing & Dickinson, 1979;Gormezano & Tait, 1976; This research was supported by Grant A0312 from the Natural Sciences and Engineering Research Council of Canada to R. W. Tait. Requests for reprints should be addressed to R. W. Tait, Psychology Department, University of Manitoba, Winnipeg, Manitoba, Canada R3T 2N2. Tait, 1974). For example, Tait (1974) showed that when a water US is followed by a paraorbital shock US, the water becomes an effective CS for the aversively motivated nictitating membrane response. When the temporal order of the stimuli was reversed in a separate group of rabbits, forward conditioning of the appetitively motivated ja...

No abstract

Backward conditioning of (he rabbit's appetitive jaw movement (Kxp. 1) and aversive nictilatiniz iiicmhiaiic (bxp. 2) responses was evaluated with three different procedures: test trials, forward acquisition ol the same response, and crossinotivational acquisition ol the other response system. In both experiments, backward conditioning consisted ol three CS alone and 22 CS-US pairings at a backward inlerstiiuulus interval of 500 msec. Contrasting the backward group with explicitly unpaired and no-treatment controls revealed that the assessment procedures did not yield identical interpretations. In lixperimcnt 1. the test trials indicated that the backward pairings produced no associative effects, whereas the forward acquisition and crossinotivational acquisition tests suggested that excitatory backward acquisition was obtained. In Experiment 2. the test trials indicated that no associative effects were obtained: The forward acquisition test identified weak inhibitory conditioning, but the crossmolivational lest implicated excitatory conditioning. Reasons for the discrepancies in interpretation of the assessment proee dures were explored. 'This rcsc;in.h was supported by yiiinl AO.'I -from Ihc N.nur.il Sciences anil llnyinCLrinj; I