highly significantp < 0.0001
sham differences were highly significant when tested separately, p < 0.0001; for rhythmic: d = 2.14; for random: d = 1.79).
sham differences were highly significant when tested separately, p < 0.0001; for rhythmic: d = 2.14; for random: d = 1.79).
Finally, there was a trend towards significance for the 3-way interaction including the quadratic component (F(1,13) = 3.54, p = 0.061), which could be explained by a larger active vs. sham decrease of the (negative) quadratic coefficient for rhythmic than for sham stimulation (p < 0.05; CI = [−0.26 −0.03]; d = 0.76; B U[−0.10 0.24] = 2.26; both active vs. sham differences being significant when tested separately, p < 0.01; for rhythmic: d = 1.34; for random: d = 1.14).
When all participants were included in the analysis, the Pearson’s correlation coefficient between these two measures failed to reach significance (R 2 = 0.06; p = 0.39; df = 12; p bootstrap = 0.90).
Despite the fact that this interaction was only close to significance, a direct comparison is of high interest for testing our a priori hypothesis that rhythmic and random effects on perception might be different 10 , 12 , 27 .
The spatial and frequency signatures of these effects appear potentially distinct; however, such differences did not reach statistical significance when rhythmic and random TMS were directly compared.