Noise in synaptic inputs reduces the reliability of neural responses. However, when noise is strongly correlated between inhibitory and excitatory synapses, fluctuations in inhibitory synaptic input may cancel corresponding fluctuations in excitatory synaptic input. Strongly correlated noise may arise if the excitatory and inhibitory pathways originate from a common source. The result would be to reduce variability in a neuron’s spike output and increase the reliability of the signal.

Jon Cafaro’s and Fred Rieke’s new paper “Noise correlations improve response fidelity and stimulus encoding” (published December 5, 2010 in Nature) reports on experiments that 1) test if converging excitatory and inhibitory input covary and 2) determine the impact of correlated excitatory and inhibitory input on neural output.

To measure the covariation of excitatory and inhibitory synaptic input required Dr. Cafaro and Dr. Rieke to measure the two conductances as near to simultaneously as possible. The results showed that excitatory and inhibitory input to retinal ganglion cells do indeed covary. Then they recorded from directionally selective retinal ganglion cells while stimulating the retina with a bar of light moved in different directions. The results showed that the signal processing underlying directional selectivity depends on covariation of excitatory and inhibitory synaptic inputs, which resulted in greater direction selectivity.


Other related blog posts:

Information that Moves from the Eyeball to the Brain

Correlated Response Fluctuations Between Cortical Neurons Rare

Information that Moves from the Eyeball to the Brain Continued

A Question of Synchrony, Correlation or Active Decorrelation among Brain Cell Responses

Signal Timing and Reliable Signal Transmission in the Brain