Review Article

Understanding Neural Population Coding: Information Theoretic Insights from the Auditory System

Figure 2

Stimulus information obtained from responses of auditory cortex neurons at different effective precisions. (a) Procedure to modify the effective response precision. We shuffled spikes (independently across trials and stimuli) across neighboring time bins. Illustration shows time bins with  ms. Effective precision was degraded to 4 ms by shuffling spikes in neighboring bins (), degraded to 6 ms by using three neighboring bins for shuffling () or degraded to 12 ms by shuffling spikes in all six bins of the pattern (). (b) Normalized stimulus information from responses to random chords. For each neuron, the absolute information values were normalized to the value at 1 ms precision. The median and 25th and 75th percentile across neurons are shown by boxplots. Information values from 1 to 6 ms were calculated by using  ms bins, whereby the effective precisions of 2, 3, and 6 ms were obtained by shuffling spikes in two, three, and six neighboring bins; values for 12, 24, and 48 ms precision were calculated by using bins of = 2, 4, and 8 ms, respectively, and by shuffling spikes in six neighboring bins. (c) like (b), except for responses to natural sounds. Figure redrawn from data presented in [19].
(a)
(b)
(c)