Figure 7
Comparison of category information conveyed by non-selective neurons computed for neural population and unit activities. The distribution of the selectivity index (A) is shown for all neurons. Arrow indicates the mean of selectivity index. Mean category information from 150 to 350 ms after stimulus onset calculated by CI in passive (B) and correct (C) trials are shown for the individual units and the neural population in non-selective neurons. Error bars represent the STD of bootstrap samples in population and SE in single neuron sub-plots. (D, E) Plots show the categorization index of the population of neurons as a function of the number of single units used for making the populations. Neurons were sorted based on SI, and sets of neural population were made by subsequently adding new units with increasing (D) and decreasing (E) SI values. Significance was computed with a bootstrap method to estimate 95%, 99%, and 99.9% confidence intervals. Asterisks indicate a significant difference between groups with different neural selectivity (*p < 0.05, **p < 0.01, ***p < 0.001).
