Extended Data Fig. 9: Unpaired aversive conditioning after preconditioning did alter dmPFC representations.
From: Prefrontal encoding of an internal model for emotional inference
(a) Schematic shows the unpaired aversive conditioning component of the modified sensory preconditioning paradigm. (b) Percentage of time spent freezing (‘freezing duration’, y-axis) during the CS period of each trial during aversive conditioning (AC) and recall for the paired-AC (dashed) and unpaired-AC (black) groups. (c) Median freezing duration (calculated for rats in each group) for the paired and unpaired groups during auditory (left) and visual CS periods (right) at recall. Dots represent individual animal values (filled = male, open = female). Mann-Whitney U test (two-sided); **(left): r = 16, p = 0.0043; **(right): r = 15, p = 0.0043; (d) Left: peri-event time histogram showing population averaged auditory-evoked calcium responses (lines) and SEM (shaded) of activated neurons during preconditioning (PC, grey) and recall (red) in the unpaired-AC groups. Dashed lines represent the onset and offset of sensory stimuli. Right: Box plots showing median (center), 25%/75% interquartile range (bounds of box) and extreme data points (whiskers) for auditory stimulus evoked calcium responses from peri-event time histograms in the left panel. Mann-Whitney U test (one-sided); n.s.: z = 0.25, r = 73698, p = 0.40; (e) Similar to d, but for calcium responses during visual CSs. Mann-Whitney U test (one-sided); n.s.: z = 1.13, r = 87535, p = 0.13; (f) The Euclidean angular distance (θ) of auditory and visual stimulus evoked population vectors for the unpaired-AC group during preconditioning (left) and recall (right). (g) The distribution of angular distances between auditory (A) and visual (V) stimulus-evoked population vectors for the unpaired-AC group during preconditioning by bootstrap resampling. Angular distance during Recall is within the 95% CI. (h) Euclidean angular distance of auditory stimulus and shock-evoked neural population vectors for the AC-paired (left) and AC-unpaired (right) groups during recall. (i) The distribution of angular distances between auditory CS during Recall and shock US during AC drawn from the AC-unpaired by bootstrap resampling. Angular distance of the AC-paired group (red line) falls outside of the 95% confidence interval (dashed lines). (j, k) Similar to h, i, but for the population vectors of visual CSs and US. (l) Left: Peri-event time histogram showing auditory-evoked calcium responses at recall for all (black), auditory-only (red) and co-responsive (Co, purple) cells classified during preconditioning in the AC-unpaired groups. Right: Box plots showing median (center), 25%/75% interquartile range (bounds of box) and extreme data points (whiskers) for sensory stimulus evoked calcium responses from cell classes. Kruskal-Wallis test (two-sided), n.s.: X2 (2,303) = 3.24, p = 0.20; (m) Similar to l, but for visual stimulus-evoked calcium responses at recall. Kruskal-Wallis test (two-sided), n.s.: X2 (2, 323) = 0.92, p = 0.63; (n) Left: Peri-event time histogram showing calcium responses to visual stimuli (20 s time-period within dotted vertical lines) and shocks (denoted by green rectangle) during unpaired aversive conditioning for cells which were identified as auditory-only, visual-only, co-responsive and others (non-significantly responsive and inhibited cells) during recall in the AC-Unpaired group. Right: Box plots showing median (center), 25%/75% interquartile range (bounds of box) and extreme data points (whiskers) for averaged shock-evoked calcium responses (from traces in the left panel). Kruskal-Wallis test (two-sided), n.s.: X2 (3, 327) = 5.61, p = 0.13. (o) Similar to n, but for cells which were identified as auditory-only, visual-only, co-responsive and others after preconditioning. Kruskal-Wallis test (two-sided), n.s.: X2 (3, 282) = 0.83, p = 0.84.
