Figure 5
From: Feature selectivity is stable in primary visual cortex across a range of spatial frequencies
Noise correlation is stable at long distances. (A) Similarity of pairwise signal and noise correlation between sessions. Each circle indicates an animal and the bar indicates the mean similarity across animals. (B) Distribution of pairwise noise correlation in Session 1. Negative noise correlations are not shown. 25% of all neuronal pairs had noise correlations larger than 0.1. Only these pairs were further analyzed to examine the relationship with distance to remove the effects of neurons with small magnitude noise correlations. (C) Relationship between noise correlation and distance between neurons. Pairwise noise correlation slightly decreased as a function of distance (Pearson’s correlation r = −0.098, p = 4.20e-3, n = 852 pairs). (D) Relationship between the difference in noise correlation (difference between sessions) and distance between neurons. The difference in noise correlation was not correlated with distance (Pearson’s correlation r = −0.027, p = 0.436). (E) Noise correlation difference for an example reference neuron. Neurons with noise correlations larger than or equal to 0.1 in Session1 are shown in white. All other neurons are shown in gray. The colors of the lines indicate the stability of noise correlation between Session1 and 2 measured as 1 − ∆ noise correlation, where ∆ noise correlation is the absolute magnitude of the change in noise correlation. Note, the pair of neurons that are most unstable (red line) are relatively close to one another. Four neurons did not have a measurable noise correlation with the reference neuron and are not shown.
