Extended Data Fig. 9: Baseline licking across training in all animals.

a) Licking behavior showed in extended time before and after reward delivery to illustrate baseline intertrial licking behavior across the indicated training trials, for control (black), stimLick- (green), and stimLick+ (purple) mice. Grey numbered blocks indicate analysis epochs for panel (b). b) (top) Quantification of baseline licking at analysis epoch 1 (indicated at right of panel (a)) across training. (bottom) Quantification of baseline licking at analysis epoch 2. c) Mean baseline licking rate (final 300 trials of training) for all the experimental groups shown in Fig 1–6. No stimulation controls (white, n = 9), stimLick- (green, n = 6), stimLick+ (dark purple, n = 5), stim+Lick+ (light purple, n = 4). d) Licking behavior over the 3 s preceding uncued trials at the end of training (trials 600–800) for the best 4 and worst 4 performing mice displayed an insignificant trend towards more baseline licking in bad learners. e) (left) No correlation between baseline licking and final latency to collect reward (a measure of learned performance) for all mice (Pearson’s p = 0.46, n = 20). (right) No correlation between baseline licking and final latency to collect reward (a measure of learned performance) for only control mice that received no exogenous dopamine manipulations during training (corresponding to data from Fig 1–2, Pearson’s p = 0.9, n = 9). f) No correlation between baseline licking and initial NAc–DA reward responses for control mice (Pearson’s p = 0.75). All error bands represent +/− SEM around the mean.