Figure S6.
Group differences in neural effects, stratified by medication status, Related to Figures 5, 6, and 7
(A) Replay-associated ripple power in post-learning rest (peak power increase in ripple range, 120 – 150 Hz). (Left) Controls versus unmedicated patients t(38) = −2.24, p = 0.031. (Middle) Controls versus medicated patients t(38) = −1.83, p = 0.075. (Right) Medicated versus unmedicated patients t(24) = −0.41, p = 0.68. Unpaired t test, 2 tailed. Sample: controls n = 27, unmedicated patients n = 13, medicated patients n = 13. See Figure 5C for control versus patient result.
(B) Learning-induced increase in representational similarity from Stimulus Localizer to Position Probe, where representation effect is extracted at time of group peak effect after picture onset. (Left) Controls versus unmedicated patients, group ∗ representation interaction: F(1, 38) = 3.16, p = 0.048. (Middle) Controls versus medicated patients, group ∗ representation interaction: F(1, 40) = 2.30, p = 0.11. (Right) Medicated versus unmedicated patients, group ∗ representation interaction: F(1, 26) = 0.07, p = 0.93. Sample: controls n = 27, unmedicated patients n = 13, medicated patients n = 15. See Figure 7D for control versus patient result.
(C) Reactivation separation analysis in post-learning rest: temporal separation of structurally adjacent states (‘near’, [A-B]) versus states separated by a single intermediate state (‘far’, [A-C]), within high-coactivation epochs. (Left) Controls versus unmedicated patients, group ∗ distance interaction: F(1, 38) = 5.56, p = 0.024. (Middle) Controls versus medicated patients, group ∗ distance interaction: F(1, 39) = 12.52, p = 0.001. (Right) Medicated versus unmedicated patients, group ∗ distance interaction: F(1, 25) = 1.69, p = 0.21. Sample: controls n = 27, unmedicated patients n = 13, medicated patients n = 14. See Figure 6A for control versus patient result. All effects plotted as mean ± SEM.