(A) Timeline of latency to enter assay. Mice were habituated to a chamber containing a dark burrow for 10 min 1 day prior to test. Mice with preference for burrow during habituation were included during test. (B) Schematic of latency to enter assay during test. Left: at the beginning of each trial, mice were confined to a corner opposite of a dark burrow with a transparent wall (holding zone). After 15 s, the transparent barrier is removed and mice can freely move about the chamber. The trial ends when mice enter the burrow or 60 s have passed. Blue light (473 nm, 5 ms, 20 Hz) is delivered in alternating trials. In light-on trials, blue light delivery begins 5 s prior to barrier removal. After burrow entry, mice can remain in the burrow for 10 s before being returned to the holding zone. (C) Bars represent average time spent in the burrow during a 10 min habituation (dlPAG: eYFP, n = 5; ChR2, n = 4; l/vlPAG-syn: eYFP, n = 5; ChR2, n = 5; l/vlPAG-CCK: eYFP, n = 17, ChR2, n = 14). (D, E) Example locomotion map of five light-off (left) and five light-on trials (right) in a l/vlPAG-CCK-eYFP mouse (D) and l/vlPAG-CCK-ChR2 mouse (E). (F) Individual dots represent mean latency during five light-on epochs subtracted by mean latency during five light-off epochs (ON minus OFF). Trials without entry were regarded as latency of 61 s. Light delivery increased latency to enter the burrow in l/vlPAG-syn-ChR2 mice compared to l/vlPAG-syn-eYFP mice (l/vlPAG-syn-eYFP, n = 5; l/vlPAG-syn-ChR2, n = 5; unpaired t-test, **p=0.0054). Light delivery reduced latency to enter the burrow in l/vlPAG-CCK-ChR2 mice compared to l/vlPAG-CCK-YFP mice (l/vlPAG-CCK-eYFP, n = 17; l/vlPAG-CCK-ChR2, n = 14; unpaired t-test, ***p=0.0003). Stimulation of the dlPAG did not affect latency (dlPAG-eYFP, n = 5; dlPAG-ChR2, n = 4). Errorbars: mean ± SEM.