Figure 5.

Relation between noise and adaptation rate. A, D, G, Scatter plots of individual parameter estimates. Parameter estimates and 68% HDIs are shown for every subject as a dot with error bars. The black line is a linear regression of $B\left[s\right]$ onto ${\sigma }_{\eta }\left[s\right]$ and ${\sigma }_{ϵ}\left[s\right]$ for average ${\sigma }_{ϵ}\left[s\right]\left(\mathit{A}\right)$, a linear regression of $B\left[s\right]$ onto ${\sigma }_{\eta }\left[s\right]$ and ${\sigma }_{ϵ}\left[s\right]$ for average ${\sigma }_{\eta }\left[s\right]$ (D) and the correlation between $K\left[s\right]$ and $B\left[s\right]$ (G). B, E, H, Heatmaps of the parameter estimate distributions. The heatmaps illustrate the distribution of the parameter estimates for the entire population of 69 subjects. The intensity represents the percentage of samples in a specific range for (B) ${\sigma }_{\eta }\left[s\right]$ and $B\left[s\right]$ (B), ${\sigma }_{ϵ}\left[s\right]$ and $B\left[s\right]$ (E), and $K\left[s\right]$ and $B\left[s\right]$ (H). C, F, I, Effect size densities. The black line represents the probability density of the regression coefficient for $B\left[s\right]$ and ${\sigma }_{\eta }\left[s\right]$ (C), the regression coefficient for $B\left[s\right]$ and ${\sigma }_{ϵ}\left[s\right]$ (F), and the correlation coefficient for $B\left[s\right]$ and $K\left[s\right]$ (I). The green lines indicate the 95% HDIs. The red line shows the mode.