ARTICLE REVIEW

Article Review

Article Review

Introduction

The article under critique is “Imaging learning and memory: Classical conditioning” by Bernard G. Schreurs published in The Journal of Psychology as well as in The Anatomical Record. For the past 50 years, psychologists have wrestled with questions regarding the relationship between conscious awareness and human conditioned behavior. A recent proposal that the hippocampus mediates awareness during trace conditioning has extended the awareness-conditioning debate to the neuroscience arena. In the following commentary, we raise specific theoretical and methodological issues regarding the Clark and Squire study and place their finding into a broader context.

Discussion and Analysis

Simple classical (Pavlovian) conditioning is a form of associative learning that has been classified as implicit or nondeclarative, meaning that conscious recollection and the integrity of the hippocampus are not required for successful performance. As the processing demands of the experimental task increase in complexity, however, the hippocampus appears to make a critical contribution to conditioned learning.

Specifying the conditions where the hippocampus becomes essential to conditioned behavior has remained an evasive but important scientific question, one that may help us to understand the role of the hippocampus in other learning tasks and the mechanisms underlying various forms of conditioning. In a recent eyelid conditioning study, Clark and Squire (1998) found that amnesic patients exhibited intact delay discrimination conditioning but impaired trace discrimination conditioning. In the delay paradigm, subjects were given one auditory conditioned stimulus (the CS1) that coterminated with an airpuff to the eye (the unconditioned stimulus, or US), whereas another auditory cue (the CS2) was explicitly unreinforced. Subjects showed greater conditioned responses (CRs) to the CS1 than the CS2, thereby indicating intact differential learning.

The trace paradigm was similar to the delay paradigm, except that a 1-s interval separated the offset of the CS1 from the onset of the US. In this case, the amnesic patients did not exhibit differential conditioning. In the control group, Clark and Squire reported that some subjects conditioned successfully and others did not. Using a postexperimental questionnaire as a probe of verbal awareness of the CS-US contingency and other aspects of the training session, Clark and Squire found a correlation between the control subjects who scored high on the declarative memory questionnaire and those that exhibited differential trace conditioning. The declarative memory questionnaire, however, was uncorrelated to performance on the delay discrimination paradigm.

As expected, amnesic patients were impaired in their declarative knowledge of both tasks. Classical conditioning used to be viewed as a type of learning that involves the acquisition of elicited responses (i.e., responses, like the defensive eye blink, that are preceded reliably by an identifiable eliciting stimulus and that are experienced phenomenologically as automatic or reflexive). Similarly, instrumental (operant) conditioning was regarded as a type of learning that involves the acquisition of emitted responses (i.e., responses, like a wink of the eye, that can occur in the absence of reliable or well-defined antecedent stimuli and are experienced as voluntary) (Davey, 2004).

An implicit assumption of these old definitions was that what is acquired ...