Extended Data Fig. 1: Mice rapidly learn new goal locations in a novel virtual reality spatial learning task.
a, Experimental timeline. Box, typical behaviour of a well-trained animal. Mean ± SEM change in lick rate (purple) and movement speed (yellow) of an example animal, averaged over trials (n = 80) in a single session in the familiar environment. b-d, Example animal’s speed distributions in AZ (black in familiar or green in novel) or NRZ (tan) and respective ROC curves over three days during no-VR baseline periods (b), familiar navigation (c), and novel navigation (d). e, Raw (top row) or percent change (bottom row) in AUC from Day 1, based on speed (left), lick latency (middle), or lick rate (right). LMM followed by Tukey correction (n = 14 familiar sessions and 13 novel sessions from 7 mice). f, Same as e for PVxAi32 mice with sub-selected trials with low- or high-intensity stimulation zones only (left, “L + H stim zones only”), or trials where stimulation duration did not exceed 5 s (right, “Short stim only”). LMM followed by Tukey correction (n = 15 goal and 15 sham stimulation sessions from 5 mice each). Data in e-f represent mean ± SEM. *P < 0.05; **P < 0.01; ***P < 0.005; ****P < 0.001.
