Fig. 7: Increased variance in grid cell orientation following long search paths in darkness. | Nature Neuroscience

Fig. 7: Increased variance in grid cell orientation following long search paths in darkness.

From: Grid cells accurately track movement during path integration-based navigation despite switching reference frames

Fig. 7

a,b, Examples of four lever-anchored grid cells in dark trials from different sessions. Trials are split by the median of the search path into a short search path (blue) and a long search path (red). a, First column shows the short (blue) and long (red) search paths of the animal in dark trials. Second column shows a firing rate histogram of lever-anchored grid cell as a function of the direction of the mouse from the lever center. The MVL and peak firing rate during the short search path (blue) and the long search path (red) are shown. Third column shows firing rate maps of the four example grid cells during random foraging. b, Trial matrix representing the neuron’s firing rate as a function of direction to the lever center for both short and long search paths. Each row in the matrix represents the firing rate of the neuron in one trial, adjusted by the average angle of trial matrix direction to the center. The further the peak firing rate is from the center, the greater the deviation in grid cell orientation. c, Trial matrices of the decoded directional error during long search paths’ beginning (left) and end (right) of dark trials. Two recording sessions are shown, one per row. The white line shows the mean decoded directional error across trials. The MVL of the distribution is shown. The smaller the decoded directional error per trial, the higher the MVL, d, Probability density of the decoded directional error during the beginning and end of long search paths (N = 24 sessions from seven mice, two-sided Kolmogorov–Smirnov Test, statistic = 0.264, P = 7.492 × 10−27); Data are presented as mean values ± s.e.m. e, Absolute decoded directional error as a function of search duration during dark trials (N = 24 sessions from seven mice, two-sided Friedman test, statistic = 31.30, P = 7.349 × 10−7). f, Directional precision when the mouse was at the lever after short or long search paths (N = 24 sessions from seven mice, two-sided Wilcoxon sign-rank test, P = 1.238 × 10−3). g, Directional drift when the mouse was at the lever after short and long search paths (N = 24 sessions, two-sided Wilcoxon sign-rank test, P = 2.393 × 10−4). Box plots show the median (center line), first and third quartiles (box bounds) and 1.5 times the interquartile range (whiskers). ***P < 0.001, **0.001 < P < 0.01.

Source data

Back to article page