Fig. 6: Ketamine induces a long-term remapping following the acute decoherence period.

a Spatial tuning curve of an example cell split into 50 trial blocks. Solid lines represent mean and shaded regions represent SEM. b Trial-by-trial correlation matrices of spatial firing rates for three example sessions. Color indicates Pearson correlation between trials. c Spatial correlation matrix averaged across 30 recorded sessions (8 mice). d Correlation of each trial to the baseline spatial template (calculated from the 50 baseline trials, n = 3233 cells). Control epoch in magenta, and ketamine epoch in green. Solid lines represent mean and shaded regions represent SEM. e Correlation to the baseline spatial template for the first 15 min in the baseline (gray), control (magenta), and ketamine (green) conditions. Solid lines represent mean and shaded regions represent SEM. f Ketamine reduced spatial correlation to the baseline spatial template (two-sided Wilcoxon matched pairs signed rank test, Z = 35.35, p < 10⁻¹⁰, n = 3233 cells). Width represents the kernel probability density of the data at different values. The central mark of the boxplot indicates the median, the bottom and top edges of the box indicate the 25th and 75th percentiles, respectively, and the whiskers extend to the most extreme data points not considered outliers. g Distribution of difference between mean correlation of 25 trials preceding ketamine and mean correlation 25 trials following ketamine. h The logistic decoder was trained on spatially-binned firing rates from the 50 baseline trials, then tested on the control or ketamine epochs. i Example mouse position predictions from a control epoch trial (top) and a ketamine epoch trial (bottom). j The average root mean squared error (RMSE) of the logistic decoder is shown across all 30 sessions. Control epoch in magenta, ketamine epoch in green. Solid line is mean RMSE and shaded region is SEM. RMSE is higher following ketamine injection (two-sided Wilcoxon matched pairs signed rank test, Z = −4.64, p = 3.51 × 10⁻⁶, n = 30 sessions). Significant comparisons highlighted ****p < 0.001. Source data are provided as a Source data file.