Fig. 3. Remapping is specific to entryway and persists across space and time.
a Schematic of trajectory segmentation by the second-most recent entryway into the compartment (Two-back), the most recent entryway, and exitway (left). Split-half correlations of population activity within the compartment when the data are partitioned according to various segmentations for initial CA1 (top) and CA3 (bottom) recordings. Only partitioning by entryway yielded reliable remapping (CA1: Wilcoxon signed-rank test versus 0, Two-back: Z = 1.4812, p = 0.139; Exitway: Z = 0.7082, p = 0.4822; Wilcoxon rank-sum test between conditions, Two-back versus Entryway: Z = 2.6437, p = 0.0082; Entryway versus Exitway: Z = 3.2658, p = 0.0011; Two-back versus Exitway: Z = 0.4510, p = 0.652; CA3: Wilcoxon signed-rank test versus 0, Two-back: Z = 0.9349, p = 0.350; Exitway: Z = 0.0187, p = 0.985; Wilcoxon rank-sum test between conditions, Two-back versus Entryway: Z = 4.5182, p = 6.24e−6; Entryway versus Exitway: Z = 5.0822, p = 3.73e–7; Two-back versus Exitway: Z = 1.0003, p = 0.317). b Median [PVs − PVd] computed separately at each location within the compartment for data recorded in CA1 and CA3. Significance markers indicated outcomes of uncorrected Wilcoxon signed-rank tests. c [PVs − PVd] when only including data from progressively later times since entering the compartment for data recorded in CA1 and CA3 (top). For example, values at 5 s were computed using only data recorded at least 5 s after the mouse entered the compartment. Computed at every 0.25 s increment. Shaded regions containing significance markers indicate all increments for which the outcome of an uncorrected Wilcoxon signed-rank test versus 0 is p < 0.05. Mean cumulative distance traveled since entering the compartment (bottom). Each gray line denotes a single session. All bar graphs reflect mean ± 1 SEM; p-values are uncorrected and two-sided. Data provided as Source data file. *p < 0.05, **p < 0.01, ***p < 0.001.