Fig. 3: Individual dual-function ACC neurons temporally integrate outcome information to the next-run value representation.

a Trial-averaged activity of all neurons in reward omission (RO, left) and Hit (right) trials. White vertical lines denote reward delivery time. (n = 549 neurons from 6 mice). Each row represents one neuron. b Mean population responses of all neurons during 1-s window after reward omission or delivery during RO and Hit trials. Inset: Trial-averaged Ca²⁺ response (mean ± s.e.m.). Dashed vertical lines denote reward delivery time. c Mean population responses of identified outcome monitoring neurons (marked by light orange vertical bars adjacent to heat maps in a, similar to b). d Trial-averaged go cue-evoked response of all neurons in the early (T1 + T2, left) and late (T3, right) periods. White vertical lines denote the stimulus onset. e Mean population responses of all neurons during 1-s window after cue onset in early and late periods. Inset: Trial-averaged Ca²⁺ response (mean ± s.e.m.). Dashed vertical lines denote the stimulus onset. f Mean population responses of identified value updating neurons (marked by light blue bars adjacent to heat maps in d, similar to e). g Venn diagram of the number of neurons identified to perform outcome monitoring (light orange) and value updating (light blue) functions. h Accuracy of decoding outcome (left) and period (right) using a subpopulation of 200 neurons randomly selected from all (n = 549) neurons or all but without dual-function (n = 509) neurons. *P < 0.05, ***P < 10⁻³. Data analyzed by (b, e n = 549 neurons; c n = 179 neurons; f n = 118 neurons, from 6 mice) two-sided Wilcoxon signed-rank test, or (h: n = 1,000 bootstrap iterations) two-sided Wilcoxon rank-sum test. In box plot: center line, median; box limits, upper and lower quartiles; whiskers, 1.5 × interquartile range. Statistical details are presented in Supplementary Table 1. Source data are provided as a Source data file.