Fig. 2: Place field alignment to spatial geometry persists over days.
a Example place cell maps from two simultaneously recorded cells on day 3 from electrophysiology (left) and calcium-imaging (right) recordings. b Schematic showing best match rotation (BMR) between trials. c Boxplots showing distribution of BMRs across days using data from electrophysiological (N = 6) and calcium-imaging (N = 5 animals, right) recordings. The BMRs were computed as the proportion of pairwise trial comparisons for which each rotation yielded the best match, averaged per animal. Using 3 × 4 repeated measures ANOVA a significant effect of rotations was found for electrophysiological (left) and calcium-imaging (right) groups. Post hoc Rom’s tests revealed that rotations at 0o and 180o were significantly higher than others (electrophysiology: 0o vs 90o: p < 0.0303; 0o vs 270o: p < 0.0303; 180o vs 90o: p < 0.0332; 180o vs 270o: p < 0.0332; calcium imaging: 0o vs 90o: p < 0.0023; 0o vs 270o: p < 0.0013; 180o vs 90o: p < 0.0005; 180o vs 270o: p < 0.0023), indicating that the place fields aligned to geometry. Additionally, 0o vs 180o was significant only in the electrophysiology group (p < 0.0303) and 90o vs 270o was significant only in the calcium imaging group (p < 0.0273). Boxplots are composed of a horizontal line (median), a box (upper and lower quartiles), whiskers (minimum and maximum values), and dots (individual data points). d Schematic of heading prediction method using the center-out measure. e Heading prediction accuracy. The orientation of the hippocampal map predicted heading on days 2 and 3 but not on day 1 (robust one sample t-test relative to chance; Day 1: p = 0.1680; Day 2: p < 0.0300; Day 3: p < 0.0280). Bar charts represent mean ± standard error of the mean (SEM), circles represent individual animal points (Day 1 N = 12, Day 2 and 3 N = 11 animals). Black dashed line represents chance level (50%). Asterisks (*) indicate a significance value set at 0.05. Source data are provided as a Source Data file.
