Extended Data Fig. 11: Chronic pain impairs encoding of behaviours and states in ACC.

(a,e) Average Fisher decoding accuracy of behaviours (a) and pain states (e) in SNI mice in each session. (b,f) Left: Average Fisher decoding accuracy of behaviours (b) and pain states (f) in SNI (n = 9) and uninjured mice (n = 9) before and after SNI or anesthesia (Mixed effects model, Tukey correction: pLick, interaction = 0.0089, pState, interaction = 0.067). Right: Same as left at three weeks post-SNI or anesthesia before and after morphine (Mixed effects model, Tukey correction: pLick, interaction = 0.041, pState, injury = 0.025). (c,g) auROC of behaviour (c) or state (g) Fisher decoder in each SNI (left) or uninjured (right) animal and session in real and shuffled data (Mixed effects model, Tukey correction: pSNI and control, lick and state, real vs shuffle <0.0001). (d,h) Permutation test p-values within each animal testing if Fisher decoding accuracy for behaviour (d) or state (h) is significantly higher than chance in each session (n = 1000 shuffles). (i) Left: Single cell discriminability for lick in positive pLick neurons before and after SNI or anesthesia (Two-Way ANOVA, Tukey correction: pInteraction = 0.0044). Right: Same as left at three weeks post-SNI or anesthesia before and after morphine (Two-Way ANOVA, Tukey correction: pInteraction = 0.048). (j) Same as (g) for negative pLick neurons (Two-Way ANOVA, Tukey correction: pLeft, interaction = 0.0031, pRight, injury = 0.0016, pRight, treatment = 0.029). (k) Left: Fraction of positive pLick neurons before and after SNI or anesthesia (Mixed effects model, Tukey correction). Right: Same as left at three weeks post-SNI or anesthesia before and after morphine (Mixed effects model, Tukey correction: pInteraction = 0.0024). (l) Same as (k) for negative pLick neurons (Mixed effects model, Tukey correction: pRight, treatment = 0.046). (m,n) Average behaviour-evoked activity in neurons encoding that behaviour which increase (m) or decrease (n) their activity upon bout onset (z-score from −2 to −1 s before onset). From top to bottom: Still, walk, rear, groom, and lick-encoding neurons at still, walk, rear, groom, and lick onsets, respectively. ⋆ = p < 0.05. Dots in panels b, f, and i-l are mean; dots in panels c and g are individual animals; error bars are SEM. See Table S1 Rows 121–136 for statistics.