Differential conditioning of the rabbit's nictitating membrane response to serial compound stimuli.

Abstract

Three experiments were conducted to determine the time course and contents of CS representations through an examination of differential conditioning of the rabbit's nictitating membrane response to two serial compounds. One compound (A-X+) was always paired with the unconditioned stimulus, and the other (B-X-) was always presented alone. All three experiments entailed manipulation of the interstimulus interval between the initial distinctive element of each compound (A and B) and the second, shared element (X). The joint results revealed that (a) conditioned response acquisition to the initial elements depended on the presence of X in the A-X+ compound; (b) differentiation between A and B appeared across interstimulus intervals up to 4,600 ms; and (c) conditional control over responding following A and B appeared at interstimulus intervals of at least 4,600 ms and perhaps up to 12,600 ms. The results are discussed with respect to mechanisms of occasion setting, generalization, and configuration.

Temporal discrimination using different feature-target intervals in classical conditioning of the rabbit’s nictitating membrane response

“…The appearance of CRs to the onset of the feature cue in both groups is consistent with the results of previous rabbit NM studies that have entailed the presentation of two conditioned stimuli (CSs) in sequence prior to the US (A®X+; Kehoe, Gibbs, Garcia, & Gormezano, 1979;Kehoe, Marshall-Goodell, & Gormezano, 1987;Kehoe et al, 2000;Schreurs, Kehoe, & Gormezano, 1993). The primary mechanism for the initial acquisition of CRs to a feature cue appears to be higher order conditioning between the A and the X stimuli (Gibbs, Cool, Land, Kehoe, & Gormezano, 1991;Kehoe, Feyer, & Moses, 1981).…”

“…Real-time models assume that the cascade of stimulus elements diminishes over time, an assumption that real-time models share with older models that depict the temporal encoding of a stimulus as a decaying stimulus trace (Gormezano & Kehoe, 1981;Hull, 1943;Kehoe et al, 1987;Pavlov, 1927, p. 39;Sutton & Barto, 1981, 1990 or a short-term memory (Konorski & Lawicka, 1959;Riley, Cook, & Lamb, 1981;Roberts & Grant, 1976). For example, Buhusi and Schmajuk (1999) assume that the elements immediately following stimulus onset have a large magnitude but a short duration, whereas later elements are progressively smaller but have longer durations.…”

“…Configural theories focus on feature cues in the direct elicitation of CRs (e.g., James & Wagner, 1980;Kehoe et al, 1987). These theories assume that the feature cue is integrated with the target stimulus into a joint representation through a process of immediate perceptual fusion (Gluck & Bower, 1988;Pearce, 1987Pearce, , 1994Rescorla, 1973;Whitlow & Wagner, 1972) or through a learned synthesis (Kehoe, 1998;Konorski, 1967, pp.…”

Summation and configuration between and within sensory modalities in classical conditioning of the rabbit

“…First, there is only about 50 % generalization from tones to white noise (Kehoe & Napier, 1991). Second, the same tone and white noise used in this experiment have been used successfully in differential conditioning (Kehoe, Marshall-Goodell, & Gormezano, 1987). …”

A neural network model of adaptively timed reinforcement learning and hippocampal dynamics

“…In secondary conditioning, another conditioned stimulus C S2 is paired with C S1 until it, too, gains reinforcing properties. Various experiments have shown that the conditioned response to CS 2 can be adaptively timed (Gormezano and Kehoe, 1984;Kehoe, Marshall-Goodell, and Gormezano, 1987). Indeed, Gormezano and Kehoe (1984) claimed that, in their experimental paradigm, "first-and second-order conditioning follow the same laws" (p. 314), although they also acknowledged that some variables may differentially effect first-order and secondorder conditioning in other paradigms.…”