Extended Data Fig. 1: Distribution of Scheduled ECoG recordings across study cohort.
From: Hippocampal network activity forecasts epileptic seizures
(a) Significant non-uniform distribution of Scheduled ECoGs over the multidien phases of IEA (P = 5.21 × 10−32, one-sided Rayleigh test) across participants (n = 15). Given inhomogeneous sampling of IEA phase, we used surrogate statistics and data resampling to minimize sampling bias. Bold, mean; shading, SEM. (b) Distribution of the total number of scheduled ECoG recordings sampled at four times of the day across participants (n = 15; box plots show the maximum and minimum values (whiskers), median (center line) and the 25th to 75th percentiles (box limits)). We verified that ECoG recordings are not sampled unevenly across the different times of the day (One-way ANOVA, F = 0.003, P = 0.99) across participants.
