Academic journal article The Psychological Record

Stimulus Equivalence and the Blocking Effect

Academic journal article The Psychological Record

Stimulus Equivalence and the Blocking Effect

Article excerpt

Blocking is the effect used to describe the finding that when a conditioning history is established with one stimulus, that history may interfere with or block the conditioning of a second, redundant stimulus when the two stimuli are subsequently presented as a compound. The phenomenon was discovered in a conditioned suppression paradigm, in which it was shown that a stimulus failed to suppress responding when it appeared in a compound with an established conditioned suppressor, despite the contiguous relation between the stimulus compound and an unconditioned stimulus (Kamin, 1968, 1969). Blocking has since been investigated in a variety of classical conditioning procedures, including, for example, simple conditioning experiments (e.g., Bakal, Johnson, & Rescorla, 1974; Marchant & Moore, 1973; Martin & Kevey, 1991; Stickney & Donahoe, 1983), studies of odor aversion (e.g., Cheatle & Rudy, 1978), and further examinations of conditioned suppression (e.g., Dickinson, Hall, & Mackintosh, 1976; Kremer, Specht, & Allen, 1980; Uegaki, 1976). The effect has also been observed in operant preparations, in which the blocking of discriminative (Singh & Solman, 1990; von Saal & Jenkins, 1970; Wu & Solman, 1993) and reinforcement (Williams, 1975; Williams & Heyneman, 1982) control has been demonstrated. The blocking effect has profound implications for the study of learning, for it suggests that the temporal contiguity between stimuli may not always be sufficient for conditioning to occur.

The blocking effect has not been examined within the context of conditional discrimination learning, or more specifically, the study of stimulus equivalence. Several recent investigations of stimulus equivalence have employed sample stimuli consisting of two-element stimulus compounds, in which subjects were trained to match the stimulus compounds to unitary comparisons. When the elements of each compound were later separated during the test for equivalence, both elements were shown to have entered separately into equivalence relations with the other stimuli (Markham & Dougher, 1996; Stromer & Stromer, 1990a, 1990b). Emergent relations between the elements of the compounds themselves have also been observed (Markham & Dougher, 1996). These findings have engendered the speculation that under certain conditions, the contiguous arrangement of stimuli may be sufficient for equivalence classes to develop (Smeets & Barnes, 1997; Stromer, McIlvane, & Sema, 1993). Given this body of results, it seems reasonable to suspect that the blocking effect might also be observed in an equivalence preparation in which stimulus compounds are presented as sample stimuli during training. For example, if subjects first learn to symbolically relate sample stimulus A1 to comparison stimuli B1 and C1, that history is likely to prevent stimulus X1 from entering into equivalence relations with those same stimuli when it subsequently appears in a compound with stimulus A1. Such a demonstration may prove to be an important supplement to the identification of the conditions sufficient for the establishment of equivalence classes. Indeed, the blocking effect may be the result of conditions that are insufficient for some stimuli to become class members.

The reported experiment extends the blocking effect into the domain of stimulus equivalence. We sought to determine whether a prior history of matching unitary samples (A1, A2, and A3) to unitary comparisons (B1, B2, B3, C1, C2, and C3) would block the establishment of equivalence relations between redundant sample stimulus elements (X1, X2, and X3) and the other stimuli, when the "X" and "A" elements were later presented as compounds. In other words, when the "AX" compounds were separated during the test for equivalence, would accuracy on test trials for relations between the "A" stimuli and the "B" and "C" stimuli be substantially higher than accuracy on test trials for relations between the "X" stimuli and the "B" and "C" stimuli? …

Search by... Author
Show... All Results Primary Sources Peer-reviewed

Oops!

An unknown error has occurred. Please click the button below to reload the page. If the problem persists, please try again in a little while.