Extended Data Fig. 7: Relationships between cortical DS, SPW and spindles.

a, Consistent with recent reports²⁹, most DSs were not expressed within +/− 50 ms of an SPW (median (IQR) 91.8% (90.1–93.2%), n = 29 sessions of 3 kinds from 16 subjects). b, Cross-correlograms of DS-ripple coupling during Wake, NREM sleep, IS and REM sleep. There was a small increase in the probability of ripples around the time of DSs, and a smaller increase in the probability of DSs around the time of ripples, also replicating²⁹. Shaded regions represent 95% confidence intervals for the mean (n = 29 sessions of 3 kinds from 16 subjects). c, Various spindle properties during early and late baseline NREM sleep, and early and late recovery NREM sleep. d, Spindle and DS detection rates in wake, NREM sleep, intermediate state (IS) and REM sleep. Although spindle rates were high in IS sleep relative to NREM sleep, DS rates were not. In c and d, black bars represent significant differences, assessed using general linear hypothesis tests on mixed effects models. Gray bars indicate that either a main effect or post hoc test was nonsignificant. All tests were two-sided and corrected for multiple comparisons (see Supplementary Table 1 for p-values and effect sizes, and Supplementary Information for confidence intervals; see Methods for details). Boxplots show medians and quartiles. Whiskers are drawn to the farthest datapoint within 1.5 IQRs from the nearest hinge. e, Little-to-no coupling of DS to spindles was observed. Right, normalized or ‘excess’ ripple probabilities, after accounting for covariation of event rates at timescales greater than 2 s (shift-predictor method, 1 s jitter, 100 surrogates each). Shaded regions represent 95% confidence intervals for the mean. f, Ripple probability was consistently modulated by spindle occurrence. g, No qualitative difference was observed in the spindle-ripple cross-correlogram profile either over or between baseline and recovery NREM sleep. Shaded regions represent 95% confidence intervals for the mean (n = 29 sessions of 3 kinds from 16 subjects) in e–g.