The functions of amygdala and hippocampus in conditioned cue preference learning /

The experiments in this thesis examined the roles of stimulus configuration on conditioned cue preference (CCP) learning by asking what information is processed and by which neural substrates. Results from Experiments 1 and 2 showed that lesions of the lateral nucleus of the amygdala (LNA) but not of fimbria-fornix (FF) impaired CCP learning when the cues paired with food during training were distinct from those not paired with food in either of two different apparatuses. In Experiments 3 and 4 LNA lesions increased the size of the CCP when the cues paired with food and no food were ambiguous in two different apparatuses. Learning the ambiguous cue CCP required at least one session of unreinforced pre-exposure to the cues and was eliminated by FF lesions. In the last series of experiments, a latent learning effect of unreinforced pre-exposure on ambiguous cue CCP learning on the radial maze was found in normal animals that received at least 3 sessions of unreinforced pre-exposure. FF lesions made before, but not after, pre-exposure eliminated the latent learning effect. Hippocampus lesions made either before or after pre-exposure eliminated the CCP learning. Taken together, the results are consistent with the hypothesis that distinct cue CCP learning is based on conditioned approach responses to cues paired with food, mediated by a neural system that includes the LNA. The results also suggest that ambiguous cue CCP learning takes place in two phases. First spatial learning occurs during unreinforced pre-exposure, a process that requires an intact FF. Subsequently, information about the location of the reinforcer is added to the spatial information during the reinforced training trials by a process of "reconsolidation". An intact hippocampus is required for this process. The implications of these results and interpretations for latent learning and latent inhibition are considered