Memory formation and retrieval in auditory cortex

Abstract

One important aspect of memory is association of different modality components/inputs. Brain imaging and electrophysiological studies have shown that cortex of one modality can show responses to stimulus of another modality after intensive training. In the present study, we focused on the neocortex and explored how cross-modal associative memory was established, whether the established memory could be correlated with behavioral tasks and how the entorhinal cortex participated in the establishment and retrieval of this memory. We produced a conditioned association in rats between electrical stimulation of the auditory cortex and a visual stimulus that was followed by foot shock. The merits of using electrical stimulation include the fact that the site activated by electrical stimulation is more focused than that activated by a natural sound stimulus. Also the activated area in the auditory cortex, which reflects association with the visual stimulus, can be easily monitored by recording electrodes that attached to the stimulating electrode. In contrast to this, the cortical area that activated by the sound stimulus would likely have spread to a greater portion of the auditory cortex and beyond making monitoring difficult. The results showed the most direct evidence that auditory cortical neuron responded to the visual stimulus after the conditioning, and the responses to visual stimulus gradually reinforced when training trials increased. Control sites where no electrical stimulation was given showed no significant change, so the association was built specifically between visual stimulus and the sites in auditory cortex where electrical stimulation was given. To confirm the visuoauditory association behaviorally, we first paired the combined sound and light stimulus with foot shock. After conditioning, the sound stimulus was then associated with water reward, and the establishment and recall of the association was proven when the rat successfully secured the reward after the sound stimulus was replaced by the visual stimulus for the first time. The control group who did not have such association could not acquire water reward. To examine activities in the auditory cortex during above behavioral paradigm, electrical stimulation of the auditory cortex was used instead of the sound stimulus in a separated group of subjects. Similarly, the association between visual stimulus and electrical stimulation of auditory cortex was first built. Then electrical stimulation was used as cue to train the subjects to gain water reward, and when electrical stimulation was replaced by visual stimulus, the subjects could attain water reward in the first trial. Recording near the stimulation site in auditory cortex showed that when visual stimulus was given, neural activities around stimulation site increased, and such activities extinguished if the training processed. After fully extinction, responses to visual stimulus regained while visual stimulus and electrical stimulation of auditory cortex were associated again followed by foot shock. To explore the role of entorhinal cortex, two electrode arrays were implanted into both sides of auditory cortex respectively and a cannula into one side of entorhinal cortex. Formation of new associative memory in the auditory cortex with classical conditioning was bilaterally abolished when the unilateral entorhinal cortex was temporally inactivated, but returned if the entorhinal cortex was not inactivated. Retrieval of the established associative memory in the ipsilateral neocortex was affected by the inactivation of the unilateral entorhinal cortex, while the contralateral cortex was not affected, suggesting a less dependence of the hippocampal system in the retrieval than in the formation of associative memory. This study presents most direct physiological evidences of the establishment of memory traces in the auditory cortex of behavioral rats, showing the association between the auditory and visual modalities. The results filled the gap of the process that neurons of one modality respond to other modality stimulus in highly trained monkeys. The straightforward physiological results from behavioral subjects confirmed these findings and provided four further steps concerning the cross-modal associative memory. First, associative memory can be established at single neuron level after 20 trials of conditioning. Second, neuronal responses reflect the recalling of the established memory. Third, the associative memory could be confirmed behaviorally, and only 10 trials of conditioning were needed. Finally, the extinction process lasts for hundreds trials of testing in several days and reestablishment of the extincted memory happens after a brief reconditioning. The establishment of the associative memory in the auditory cortex needed the involvement of the entorhinal cortex. It was interesting to note that a unilateral inactivation of the entorhinal cortex affected the establishment of the association bilaterally in the cortex. That was also confirmed behaviorally as the animal showed no context learning. This result may reflect the bilateral projection of the entorhinal cortical neurons to the hippocampus. However, the retrieval of the memory was only affected in the ipsilateral cortex to the inactivated entorhinal cortex, suggesting a less dependence of the hippocampal system in the retrieval than in the formation of associative memory. The present results imply that the hippocampal system is necessary for the formation of associative memory in the auditory cortex from the beginning. Memory retrieval still needs the involvement of the hippocampal system at least at early stage within 10 days, though its dependence on the hippocampal system would be less than that of the formation of associative memory.