Fig. 4: Role of grouped spindles in motor skill consolidation and generalization.

a Relationship between the proportion of grouped spindles (unitless), extracted at scalp derivation C4 during the first 20 min of NREM2 sleep, and the magnitude of overnight skill consolidation (in %) on the learned motor sequence in the IMMO (n = 15) and CTRL (n = 14) groups. b Relationship between the proportion of grouped spindles (unitless), extracted at scalp derivation C4 during the first 20 min of NREM2 sleep, and the magnitude of inter-manual skill transfer (in %) in the IMMO (n = 15) and CTRL (n = 14) groups. c Relationship between the proportion of grouped spindles (unitless), extracted at scalp derivation C4 during the first 20 min of NREM2 sleep, and the magnitude of new-sequence skill transfer (in %) in the IMMO (n = 13) and CTRL (n = 14) groups. Individual data (colored circles) and trend lines are provided, along with Pearson’s r and associated p-values. Maps show correlations between the proportion of grouped spindles of the first 20 min of sleep computed at multiple scalp derivations with skill consolidation (d), inter-manual skill transfer (e) and new-sequence skill transfer (f) for the IMMO (right) and CTRL (left) groups. Colors represent Pearson’s correlation coefficient values, and white dots indicate significant-level p-values uncorrected for multiple comparisons. The red dot represents significant-level p-value after correction for multiple comparisons using the Benjamini–Hochberg procedure (Ntests = 30).