Search

Place cell activity in the hippocampus depends in a complex way on the information from two major sensory sources: External (e.g., visual) cues and internal cues (self-motion, vestibular cues). Dissociation of the two inputs in the presence of the animal modifies the place fields in ways that depend on the magnitude of the difference: If the mismatch is small, place fields typically follow the external cues, but if the mismatch is large, their activity is determined mainly by the internal cues. The dynamics underlying the interaction between the external and internal cues in the hippocampus is not completely known, but is an important issue in understanding the hippocampus’ role in spatial learning and navigation. In this paper, we propose an attractor model of the hippocampus which captures the interaction between external and internal sensory inputs to the hippocampus. The proposed mechanism accounts for the difference seen in the experimental results in the case of a small versus a large discrepancy between the visual and idiothetic information.

It is well-known that identical sensory input under different perceptual, behavioral or contextual conditions can produce distinct patterns of activity in the place cells of the rodent hippocampus. However, the mechanisms underlying this have not been completely clarified. A recent experiment has shown that place cell activity on a 3-arm maze exhibits hysteresis as the maze is rotated with respect to distal cues. The apparent angular extent of a place field is greater when a maze arm rotates out of it than when it rotates back into the field. In this report, we present a simple attractor-based model of the hippocampus that reproduces this hysteresis phenomenon. The model allows us to make predictions about changes in the hysteresis effect as the animal becomes more familiar with the maze in several orientations. It also has implications for the place field remapping phenomenon seen in many hippocampal experiments.

Recently, it has been suggested that attractor networks may provide a mechanism for context-dependence in hippocampal place codes. We have proposed that context may be coded by “latent attractors” — mutually competitive and internally cooperative cell groups which channel the system's response to afferent stimuli. We have also argued that it is the disynaptically recurrent dentate gyrus–hilus (DGH) system which embodies these latent attractors in the hippocampus. Others have suggested the CA3 network — with its monosynaptic recurrence — is the site. While latent attractors (and, thus, context-dependent coding) can be implemented by a 1-layer or 2-layer recurrent network, we show that a 2-layer recurrent network can implement a sparse context-dependent place-code using latent attractors much more flexibly than a 1-layer network.

Primary neurons in all parts of the rodent hippocampus show location-specific patterns of activity called place fields, thus forming a distributed representation of place. Paradoxically, environments with very similar sensory cues can produce distinct place representations. The origin of this context dependence is not known. In this paper, we speculate that the dentate gyrus–hilus (DGH) system, with its disynaptic recurrent connectivity plays a central role in creating context-dependent place representations. We show that a simple system architecture can support distinct and consistent place representations for similar environments.