Correlation and neural information coding fidelity and efficiency

Analysis of neural spike patterns shows that the firing patterns of neurons within an ensemble can be correlated. The question of how this interneuronal correlation affects neural coding is not well understood. We show that in a simulated neural population in which correlation is a result of convergent inputs, both coding ability and efficiency are increased. Coding efficiency is further increased when the population is composed of neurons with diverse responses rather than neurons with identical response characteristics. (C) 1999 Elsevier Science B.V. All rights reserved.