Solving the tag assignment problem for neural networks with simulated assemblies of minicolumns

A computer simulation of cell assemblies in which groups of minicolumns behave as discrete representations was developed. A neurally plausible model which simulates both the parallel processes that may give rise to illusory conjunctions and the serial processes that may solve the tag-assignment problem in normal perception has been constructed. Each simulated minicolumn is a local population of neuronal elements that share intracolumnar activity and have similar intercolumnar connections. The model for the neuronal elements is a simplification of the Hodgkin-Huxley model. A Hebbian learning rule models the development of LTP-type connection strength between pairs of minicolumns that are simultaneously active. A simulated fovea moves over a two-dimensional stimulus space in saccade-like steps, with direction and extent determined by competitive integration of location detectors for off-fovea stimuli