Encoding and retrieval are differentially processed by the anterior cingulate and prelimbic cortices: a study based on trace eyeblink conditioning in the rabbit.

A growing body of literature suggests that structures along the midline of the prefrontal cortex (mPFC), including Brodmann's area 32 (prelimbic cortex) and area 24 (anterior cingulate cortex) in the rabbit play a role in retrieval of learned information. The present studies compared the effects of post-training lesions produced either immediately or 1-week following learning, to either prelimbic (area 32) or anterior cingulate (area 24) cortex on trace eyeblink (EB) conditioning. Further, because recent evidence suggests that the mPFC may play an even greater role in learning and memory when emotional arousal is low, these studies compared the effects of lesions in groups conditioned with either a relatively low-arousal corneal airpuff, or a more aversive periorbital eyeshock unconditioned stimulus (US). A total of six groups were tested, which received selective ibotenic acid or "sham" control lesions to either area 32 or 24, immediately or 1-week following asymptotic learning, and conditioned with an eyeshock US or an airpuff US. Results showed that the greatest lesion deficits were found when conditioning with the less aversive airpuff US. Further, lesions produced to area 32 one-week, but not immediately following learning, caused significant deficits in performance, while lesions produced to area 24 immediately, but not 1-week following learning, caused significant deficits in performance. These findings add to the body of evidence which shows that area 32 of the mPFC regulates retrieval, but not acquisition or storage of information, while area 24 mediates a less specific reacquisition process, but not permanent storage or retrieval of information during relearning of memories abolished by mPFC damage. These findings were, however, specific to those experiments in which the relatively non-aversive airpuff was the US.

Prefrontal control of trace eyeblink conditioning in rabbits: role in retrieval of the CR?

Rabbits (Oryctolagus cuniculus) were trained on a trace eyeblink (EB) conditioning task to a criterion of 10 consecutive EB conditioned responses (CRs). One week later, ibotenic acid or sham lesions were made in the mPFC centered on the prelimbic region (Brodmann's area 32) or the cingulate cortex (Brodmann's area 24). Following a 1-week postoperative recovery period, all animals were retrained for 4 consecutive days using the same parameters as during acquisition, given 1 week off, and retrained for another 4 days. Mean EB conditioning deficits in the group with area 32 lesions occurred on the first and second days of each retraining period. However, by the third and fourth days of retraining, these lesioned animals were performing at a level comparable to that of the sham group. Lesions of area 24 did not produce deficits at either retesting period. These findings were interpreted to indicate that area 32, but not area 24, is involved in retrieval processes, rather than consolidation or storage, in that the animals were impaired at both retesting times, but were able to relearn the task.

Ibotenic acid lesions to ventrolateral thalamic nuclei disrupts trace and delay eyeblink conditioning in rabbits.

Intact cerebellar structures (i.e., deep nuclei and perhaps cortex) are essential for acquisition of both simple delay and trace eyeblink (EB) conditioning. However, successful trace conditioning also requires intact cortico-limbic structures (i.e., hippocampus, medial thalamus, and medial prefrontal cortex, mPFC). A direct connection between the cerebellum and ventrolateral thalamic nuclei (VLTN) has been demonstrated in several species. Since VLTN projects to both premotor and prefrontal cortex, it may be an essential link in a cerebellar-thalamic-prefrontal circuit that provides the CNS substrate for acquisition of the trace EB CR. The current studies thus assessed the role of the VLTN on trace EB conditioning in New Zealand albino rabbits. We first verified afferent connections to the mPFC (Brodmann's area 32) from the VLTN, by injecting the retrograde tracer Flourogold(c) into area 32. Strong labeling in VLTN from terminal projections to mPFC were found. We next assessed the role of VLTN in trace eyeblink conditioning in animals that received either sham or ibotenic acid VLTN lesions. EB conditioning began with 10 consecutive daily sessions of trace conditioning, followed immediately by 4 days of extinction, and then 4 days of delay conditioning. VLTN lesions significantly impaired acquisition of both trace and delay conditioning, and impaired extinction. These findings, thus confirm the importance of the VLTN in a postulated cerebellar-thalamic-prefrontal circuit that underlies successful trace, as well as delay EB conditioning.