Role of the prefrontal--thalamic axis in classical conditioning.

The major conclusion to be drawn from the above-described research on the role of the PFCag in classical conditioning is obviously that it plays a primary and perhaps necessary role in the establishment of visceral cues associated with exposure to classical conditioning contingencies. Specifically, these visceral changes appear to be of an inhibitory character. This is significant, since we have postulated that inhibitory cardiac changes invariably accompany initial processing of sensory stimuli for informational value. Such visceral changes are thus not epiphenomena associated with other simultaneously occurring physiological events. A variety of lesion experiments implicate the PFCm as a central structure in this process, since damage to this area greatly attenuates, and in the case of hypothalamic knife cuts, completely eliminates learned bradycardia. Neuroanatomical tract-tracing experiments revealed that the PFCm and lag have direct projections to the NTS and DVM in the dorsomedial medulla and the nucleus ambiguous in the ventral medulla, all of which provide medullary output control of visceral activities. The nucleus ambiguous and DVM have been specifically implicated in vagal control in the rabbit (Ellenberger et al., 1983). Electrical stimulation of the PFCm provides additional evidence that this area of the brain participates in parasympathetic activities, including cardiac inhibition, since stimulation of the entire MD projection cortex, including the PFCm, produces HR decelerations accompanied by depressor responses. However, since lesions of the Iag produced relatively little effect on conditioned bradycardia, this part of the PFCag does not appear to play a major role in the development of conditioned bradycardia. Electrophysiological recording studies, including both multiple unit as well as extracellular single unit studies reinforce these conclusions. A short latency (40-180 msec) CS-evoked increase in MUA was recorded from cells in both the dorsomedial as well as central PFCm. The magnitude of these CS-evoked neuronal changes (a) was correlated with the magnitude of concomitantly occurring conditioned bradycardia; (b) was trial-related; (c) was not obtained in a similar pseudoconditioning group; and (d) declined to pretraining levels during subsequent experimental extinction. Similar, but not identical, CS-evoked changes in neuronal activity were recorded from MD. Although tone-evoked increases in MUA were also obtained from the Iag, this activity did not show the characteristics of associative learning. Single unit analysis also suggests the importance of the PFCm in elicitation of conditioned bradycardia.(ABSTRACT TRUNCATED AT 400 WORDS)