Recent electrophysiological studies have shown that electrical brain stimulation (EBS) can produce lasting increases in synaptic efficacy, defined as the quantitative relationship between pre- and postsynaptic activity. However, the behavioural significance of this long-term potentiation (LTP) has yet to be demonstrated clearly. The principal objective of this thesis was to determine whether increased synaptic efficacy, produced by high-frequency EBS, enhances behavioural responses to a fixed amount of presynaptic activity. Following identification of EBS parameters capable of monitoring synaptic efficacy for long periods without producing LTP, a paradigm was developed in which a single-pulse of EBS in the perforant path (PP) acquired stimulus control over the temporal pattern of operant responses. In this paradigm, postsynaptic evoked potentials in the dentate gyrus (DG) produced by the PP EBS were used to monitor synaptic efficacy on every trial of conditioning and also to measure one component of the neural activity generated by the EBS stimulus controlling the operant responses. In Experiment 3, high-frequency stimulation of the PP produced- LTP at the PP-DG synaptic interface and facilitated subsequent acquisition of stimulus control by the single pulse PP EBS. This effect could not have been due to sensitization or to the stimulus properties of the high-frequency trains. The final experiment confirmed the importance of the EBS-evoked activity in the DG to the stimulus control by the PP EBS. The rate of acquisition was directly related to the magnitude of the evoked potentials and two-stage bilateral lesions of the PP in a specific sequence reduced the probability of behavioural responses to the EBS. Taken together, these results indicate that the behavioural consequence of excitatory neural activity can be enhanced by an increase in synaptic efficacy. As such, they provide considerable support for the validity of LTP as a model of neural changes subserving learning and for physiological theories of memory based on modifications in the- strength of synaptic connections. / Arts, Faculty of / Psychology, Department of / Graduate

No description available.

No description available.

No description available.

The aim of these experiments was to investigate learning and memory extensively in two nonverbal domains (olfactory and visual), and to determine similarities and differences in the function of the neural substrates that subserve these modalities. Two complementary methodological approaches were taken: (1) examination of learning and retention in patients with resection from left (LR) or right (RR) temporal lobe, and (2) study of brain function via Positron Emission Tomography (PET) of healthy subjects during memory processing. / Two parallel recognition tests were developed (one olfactory, one visual) that examined memory at three stages: following a single exposure to test stimuli, after four exposures, and following a 24hr delay interval. In the olfactory patient study, LR and RR groups performed significantly worse than the healthy control subjects, with no difference between the patient groups; thus suggesting a lack of hemispheric superiority for this task. The PET study of healthy individuals supported the bilateral participation of piriform cortex during olfactory recognition. The results from these two studies, along with findings from animal work, suggest that the piriform cortices may play a role in odor memory processing, not simply in perception. / On the face memory task, LR and RR patients showed different results. Only RR patients were impaired, while LR patients did not perform differently from controls. This unique face learning paradigm was sensitive to right temporal lobe damage, and correctly classified patients by side of resection with a sensitivity rate of 82% and specificity rate of 79%, suggesting its possible utility as a clinical tool. PET face memory findings indicated greater participation of fusiform regions during long-term recognition, and greater right prefrontal activity during short-term recognition, when these conditions are directly compared to each other. / Finally, PET was used to study the same healthy subjects performing parallel odor and face working-memory tasks, focusing on regions previously shown to be important for working memory. Results revealed similar regions of activation in dorsolateral prefrontal cortex in the two modalities. This indicates an overlap in the brain regions that process olfactory and visual information when the same cognitive manipulations are being carried out online.

The brain areas required for latent learning in the rat are not currently understood. Previous tasks used to assess latent learning, defined as the acquisition of neutral information that does not immediately influence behavior, have shared characteristics that prevented their use to determine the neurobiology of latent learning. This thesis describes a new task called the Latent Cue Preference (LCP) task, derived from the Conditioned Cue Preference (CCP) task that has been successfully used to determine the brain areas required for conditioning in the rat and other animals. In the LCP task, water deprived rats alternately drink a salt solution in one distinctive compartment of a CCP box apparatus and water in the other compartment over 8 days (training trials). They are then given a choice between the two compartments with no solutions present (preference test). The results of the behavioral experiments showed that this training results in two parallel forms of learning: (1) latent learning of an association between salt and salt-paired compartment cues, and (2) conditioning to water-paired compartment cues. Latent learning itself involved two components: (1) the latent association between salt and salt-paired cues, and (2) motivational information about salt deprivation used to retrieve the latent association, and used to compete with the conditioning to water-paired cues. In addition, the findings showed that latent learning and conditioning involve different neural circuits. Latent learning required an intact cortical-to-hippocampus circuit via the entorhinal cortex, while conditioning required an intact subcortical-to-hippocampus circuit via the fimbria-fornix. The acquisition and storage of the latent association depended on an intact entorhinal cortex/dorsal hippocampus circuit, while the use of motivational information to retrieve the association recruited the ventral hippocampus. Conditioning, on the other hand, required an intact fimbria-fornix, lateral amygdala, and hippocampus. These findings provide new knowledge to the field of learning and memory research, and allowed an update of the current Multiple Memory Systems model.

The role of the dorsal hippocampus, entorhinal cortex, amygdala and fimbria fornix in spatial discrimination was investigated using temporary inactivation and lesioning methods. Spatial learning was tested in a conditioned cue preference (CCP) paradigm involving three phases of behavioral testing on an 8-arm radial maze. In the first phase (pre-exposure) rats were given unreinforced pre-exposure trials in which they were free to move on two adjacent arms of the maze on three consecutive days. Rats were then alternately confined to the ends of the arms for eight days (training), one arm that contained food (Paired-arm) and one that did not (Unpaired-arm). The rats were then given a choice between the two arms with no food present. Only when given unreinforced pre-exposure trials did rats spend more time in the Paired-arm than in the Unpaired-arm, a CCP. Rats with muscimol induced inactivation of the dorsal hippocampus during unreinforced pre-exposure acquired a CCP for the Paired-arm but were impaired with hippocampal inactivation during training or testing. Inactivation of the entorhinal cortex resulted in impairment in all phases of the paradigm. Inactivation of the dorsal hippocampus in the Unpaired but not Paired-arm only resulted in impairment. The effects of dorsal hippocampus inactivation in either the Paired or Unpaired arms were reversed in rats with combined amygdala lesions\dorsal hippocampus inactivation. Rats with fimbria fornix\entorhinal cortex disconnections during pre-exposure were also impaired. These results reveal that spatial learning may rely on the interactions between the hippocampus, entorhinal cortex, amygdala, and fimbria fornix and that the dogma that the hippocampus mediates all forms of spatial learning requires revision.

No description available.

N-methyl-D-aspartate (NMDA) receptor activation is believed necessary for certain types of learning. The present experiments investigated the effects of the NMDA antagonist, MK-801, on spatial learning and memory in rats. Experiment 1 tested the effects of MK-801 on the acquisition and retention of a water maze task. MK-801 produced a performance, but not a spatial learning deficit. Experiment 2 tested the effects of MK-801 on the acquisition and retention of a radial arm maze task (RAM). MK-801 had no effect on initial acquisition and retention, but impaired subsequent reversal learning when the pattern of rewarded and unrewarded arms was reversed. Experiment 3 investigated the effects of MK-801 on RAM reversal learning in rats previously trained on the initial task in the absence of drugs. MK-801 produced a dose dependent impairment on reversal learning. These results are consistent with one interpretation that MK-801 impairs the ability to suppress interference from previously learned information.

Declarative learning entails the internalization of facts and events. This type of learning depends on the integrity of the hippocampal system. In rodents, spatial learning is studied as a model of declarative learning. In this thesis, electrophysiological and behavioral experiments assessed the role of NMDA receptors in synaptic plasticity and rats' spatial learning and memory. Primed burst potentiation (PBP), a form of synaptic strengthening, was studied in freely-behaving rats treated with NMDA receptor antagonists. The impairments caused by the antagonists correlated with those observed in behavioral studies. The results support the idea that NMDA receptors in the hippocampal system mediate the internalization of the contents and organization of new environmental information, and show that the receptors are not relevant for spatial working memory or performance once a representation of the environment is stable. The results also suggest that stable spatial representations resemble multiple relations of events and do not correspond to topographical maps of an environment. As learning proceeds, representations are activated by smaller subsets of environmental cues, which eventually become sufficient for effective navigation. The representations thus are encoded as relationships of stimuli that share similarities or that are unique to a particular event. The organization of novel information is given through NMDA receptor-mediated synaptic plasticity. This plasticity mechanism could resemble a process similar to the synaptic changes observed during PBP.