Fig. 2: Theta-synchronized stimulation increases hippocampal theta.

A Changes in endogenous theta over the course of repetitive theta-synchronized (TS) in a representative participant. Endogenous theta-band power was computed for the peri-stimulation interval using generalized eigenvalue decomposition to remove stimulation-evoked responses. Endogenous theta power significantly increased over the course of repetitive TS stimulation pulses (p < 0.05, FWE corrected, permutation test). Time-frequency pairs with non-significant changes are transparency masked. The vertical dashed line indicates the onset of stimulation. The scatter plot below depicts a significant example from this session (n = 1007 stimulation events, Pearson’s r = 0.10, p = 0.001, two-tailed), with each point denoting an individual TS stimulation event. B Decreases in endogenous theta power over the course of repetitive phase-blind (PB) stimulation. The participant and plotting convention both follow (A), with an example significant effect shown below (n = 1142 stimulation events, Pearson’s r = −0.08, p = 0.008, two-tailed). C Group mean changes in theta throughout the repetitive stimulation period (±SEM in shaded regions) for TS and PB stimulation. Stimulation pulses were grouped into deciles to account for variability in the number of pulses across subjects (TS stimulation: 40–85 pulses per decile, mean = 58 pulses, PB stimulation: 49–117 pulses per decile, mean = 77 pulses). Endogenous theta power was normalized to the first decile. Endogenous theta power increased during TS versus PB stimulation (n = 7 participants, linear mixed-effects model β = 0.08, 95% CI = [0.02, 0.14], F(1, 6.8) = 6.4, p = 0.04, ***). Source data are provided as a Source Data file.