Lesions of the perirhinal cortex impair sensory preconditioning in rats

Abstract

The effects of lesions of the perirhinal cortex on the development of associations between two conditioned stimuli (CSs) were examined with a sensory preconditioning procedure. Rats were given either bilateral electrolytic lesions of the perirhinal cortex or control surgery. They were then given either paired or unpaired presentations of a light CS and a tone CS. All of the rats were then given eyeblink conditioning procedures that involved paired presentations of either the light or tone and a periorbital shock unconditioned stimulus (US). The rats were finally given a test session that consisted of unpaired presentations of the tone and light CSs. Sensory preconditioning was established in the control group, but not in the lesion group. The findings are consistent with the view that the perirhinal cortex is involved in forming associations between neutral stimuli (even in the absence of reinforcement).

Introduction

A great deal of progress has been made toward understanding the neurobiology of simple associative learning [1], [9], [11], [16], [19], [24], [25], [26], [29], [39], [42], [49]. Classical conditioning procedures can be used to establish simple associative learning by pairing a conditioned stimulus (CS) with an unconditioned stimulus (US). Sensory preconditioning is a special case of simple associative learning because it involves an initial pairing of two conditioned stimuli (CS1 and CS2 in phase 1) in the absence of an unconditioned stimulus, after which the subject is trained in traditional classical conditioning (CS1-US in phase 2) [5]. The presence of an association between CS1 and CS2 can only be inferred by post-training test presentations of CS2 (phase 3). Previous studies have shown that animals given initial CS1-CS2 pairings before CS1-US training respond to test presentations of CS2, whereas control groups given unpaired presentations of CS1 and CS2 do not [36], [37], [47], [48]. It is thought that the animals given paired training respond to CS2 by virtue of its association with the excitatory CS1 [5], [48]. The two conditioned stimuli often differ in modality (e.g. tone and light) and thus require the formation of a crossmodal association.

It is difficult to specifically distinguish between disruptions of encoding or retrieval because of their intimate relationship [38], [50]. However, the clearly delineated encoding and retrieval opportunities for a CS1-CS2 association in sensory preconditioning (i.e. phase 1 and phase 3, respectively) enable a more focused study of the underlying neural bases of sensory associations.

Previous studies of the neural mechanisms of sensory preconditioning demonstrated that lesions of non-specific association cortex [47], the fimbria [36], or field CA1 of the hippocampus [37] prevented the establishment of sensory preconditioning. Based on these studies, a reasonable conclusion is that the connectivity between the hippocampus and cerebral cortex must remain intact for sensory preconditioning to be established.

The cortical areas surrounding the hippocampus in the rat have been divided into postrhinal, perirhinal, and entorhinal cortices [8]. The perirhinal and postrhinal cortices both provide input to the entorhinal cortex. However, in the rat, the perirhinal cortex receives substantial input from both auditory and visual cortical areas, whereas the postrhinal cortex receives input primarily from visual cortex [7], [10]. This connectivity establishes the perirhinal cortex as a major source of polysensory information for the subiculum and hippocampus via substantial connections with the lateral entorhinal cortex [8], [32], [40]. Thus, the perirhinal efferents to the hippocampus may provide the necessary information for crossmodal sensory associations. There is precedence for the view that the hippocampus and parahippocampal region constitute functionally distinct components of a larger system [6], [12], [13], [14], [33], [43] or are components of overlapping, but separate systems [15], [20], [21], [22], [28], [30], [35], [51].

It has recently been proposed that the parahippocampal region binds the elements of paired associates, whereas the relations among the paired elements are dependent on hippocampal processing [12]. In context of sensory preconditioning, it is possible that the perirhinal cortex functions in the encoding or storage of the CS1-CS2 association, which is flexibly expressed by the paired group in the testing phase.

The current study used a sensory preconditioning procedure to examine the role of the perirhinal cortex in the development of associations between two conditioned stimuli. Rats were given either bilateral lesions of the perirhinal cortex or control surgery followed one week later by three phases of training. In Phase 1, half of the rats from each surgical group were given paired presentations of a tone CS and a light CS. The other half of the rats were given unpaired presentations of the two CSs. In Phase 2, all of the rats were given eyeblink conditioning procedures that involved paired presentations of either the light or tone and a periorbital shock US. In Phase 3, all of the rats were given a test session that consisted of unpaired presentations of the tone and light CSs (Table 1).