Figure 5. Serotonin increases stimulus detectability for receding but not for looming motion. A, Object position (lateral distance to the animal’s skin surface) as a function of time. The neural responses during looming (dashed gray box) and receding (middle solid gray box) gray boxes were compared to the baseline activity while the object was stationary and located far away from the animal (right gray box). B, Population-averaged burst fractions before (black) and after (red) serotonin application during looming (left), receding (middle), and baseline (right). Stars indicate statistical significance using a paired t test (looming: t(12) = −2.70, p = 0.03, n = 13; receding: t(12) = −3.35, p = 0.009, n = 13; baseline: t(12) = −3.08, p = 0.01, n = 13). C, Top, spike count distributions obtained during baseline (gray) and during receding stimulation (black or red) before (left) and after (right) serotonin application for an example ELL pyramidal cell. Best-fit Gaussian curves are superimposed on each distribution. Bottom, ROC curves from this same example neuron before (black) and after (red) serotonin application. D, Same as C but for looming motion. E, Population-averaged values for the auROC before (black) and after (red) serotonin application for receding motion. The horizontal brown line indicates the chance level. A significant increase was observed after serotonin application (signed-rank test, p = 0.0007, n = 13). F, Same as E but for looming motion. No significant change was observed (t test, t(12) = 1.08, p = 0.301, n = 13).