Fig. 5: Task-related neuronal population coding in the olfactory cortex and striatum.

a Configurations of the dual-site single-unit recordings. Custom-made tetrode arrays were implanted into the Tu and aPC. Mice performed the same stimulus-outcome-associated task in head-fixed configuration as in the scanner. b Population vectors were constructed by concatenating the binned spike counts of individual single units \(\left\{{U}_{1},\ldots ,{U}_{n}\right\}\) pooled across sessions by matching trial bin \({b}_{i}\), trial order \({{tr}}_{j}\), and trial type. c The Euclidean distance of Tu population vectors from baseline displayed an RP coding at CS and during the waiting period, as well as reward coding at US. Displayed mean ± SEM, 500 ms bins slid in steps of 125 ms. d Euclidean distance from baseline during CS (from 0 to 1 s) and waiting window (from 1 to 2.5 s). One-way ANOVA with Tukey post hoc comparisons performed separately in each window (n = 169 units). e Pearson correlation between Tu population vectors displayed as mean across pairs of trials ± SEM. The population responses to CS100 and CS50 became more similar during wait, suggesting a partial overlap in Tu units coding for the two stimuli. f First three principal components of the time-embedded averaged trajectories of Tu during different trial types. The arrow marks the direction of the temporal evolution. Reward delivery abruptly deflected the population trajectories of CS50→R and CS50→N. g, h Same as (c, d) for aPC. aPC population displayed a dominant response to all CS, but no monotonic RP coding (n = 486 units). i Same as (e) for aPC. The similarity between aPC population response to all stimulus onsets is consistent with a sensory detection signal. j Same as (f) for aPC. Also aPC population trajectories reflected an initial indiscriminate sensory detection signal. In the figure: * indicates P < 0.05 (see Supplementary Table 1 for exact P values and test details) and n indicates the number of units. Box plots: The bounds of the box represent 25th to 75th percentiles. The center indicates the median. The lower and upper whiskers represent the minimum and maximum values without outliers; outliers indicated by circles. Source data are provided as a Source Data file.