Figure 3

Withdrawal behavior and lick behavior become dissimilar near perceptual threshold compared to more salient stimuli. (A) Comparison of sensitivity, and (B) response bias as a function of time between lick versus withdrawal behaviors. These results show that sensitivity between behaviors is predominantly not significantly different within a session, while the criterion is significantly different between behaviors in the early and late stage of the session. This suggests that the decision to lick is influenced by higher-order processes that do not impact the reflex behavior. (C) Euclidean distance calculated as a function of stimulus intensity for lick versus withdrawal behavior. Green region indicates near-threshold intensities, while black dotted lines represent the Euclidean distance if response types were similar for 50% and 95% of all trials at any given intensity. Lick and reflex behaviors become dissimilar near threshold, which suggests that the behaviors are encoding different information. All results are shown as mean ± SEM, except for Euclidean distance, which is shown as mean ± 95% CI (n = 40), grey shaded regions and asterisks in a-c indicate significance at P < 0.025 using a two-sided Wilcoxon signed rank test. Data was plotted using MATLAB R2019a (https://www.mathworks.com). The figure was created using Adobe Illustrator CC 2019 (https://www.adobe.com).