A paper I wrote with Michael Brown, published in Neural Computation in July 2006.

Summary for lay people:

The brain encodes information in which neurons fire electrical impulses when. These electrical impulses are called spikes. The meaning of a spike from a neuron is what world events cause the neuron to spike in that context. Such meaning is called the tuning curve of the neuron. A tuning curve is said to be wide if the neuron responds to many stimuli (you might call it promiscuous), and narrow if the neuron is very picky. Zhang wrote a paper with a leading theoretical neuroscientist, Terry Sejnowski, giving a simple formula for how narrow or wide tuning curves ought to be to be optimal in the sense of carrying the most information. We showed that the formula did not really apply in the real world due to a tacit simplifying assumption of theirs, and extended the analysis to a somewhat more realistic scenario.

Abstract:

The efficiency of neuronal encoding in sensory and motor systems has

been proposed as a first principle governing response properties within

the central nervous system. We present a continuation of a theoretical

study presented by Zhang and Sejnowski, where the influence of neuronal

tuning properties on encoding accuracy is analyzed using information

theory. When a finite stimulus space is considered, we show that

the encoding accuracy improves with narrow tuning for one- and twodimensional

stimuli. For three dimensions and higher, there is an optimal

tuning width.

The full text of the paper