Self-organization of grid fields under supervision of place cells in the model of neuron with associative plasticity (1503.07707v2)

Abstract: The grid cells (GCs) of the medial entorhinal cortex (MEC) and place cells (PCs) of the hippocampus are key elements of the brain network for the metric representation of space. Currently, any of the existing theoretical models can explain all aspects of GC-specific spatially selective patterns of activity. It is also not clear how these patterns are formed during the network development. On the other hand, it was previously shown that the network that can learn to extract high principal components from the activity of the place cells could provide the basis for GC-like activity patterns development. Supporting this hypothesis is the finding that PC activity remains spatially stable after the disruption of the GC firing patterns and that grid patterns almost disappear when hippocampal cells are deactivated. Development of the early PCs before GCs formation also supports the role of PCs as the spatial information providers to GCs. Here we have developed a new theoretical model of grid fields formation based on synaptic plasticity in synapses connecting PCs in hippocampus and neurons in entorhinal cortex. Learning of hexagonally symmetric fields in the model is due to complex action of several simple biologicaly motivated synaptic plasticity rules. These rules include associative synaptic plasticity rules similar to BCM rule and homeostatic plasticity rules which restrict synaptic weigths. In contrast to previously described models, this network could learn GC patterns after a very short experience of navigation in a novel environment. In conclusion we suggest that learning on the basis of simple and biologically plausible associative synaptic plasticity rules could contribute to the formation of grid fields in early development and to maintainence of normal GCs activity patterns in familiar environments.