Indeed, when compared across the experimental phases, cerebellar cTDCS significantly impaired motor adaptation of the proximal upper-limb compared to M1 stimulation, with a near-significant impairment compared to sham ( p = 0.056).
The mean difference in area under the curve between the sham and cerebellar group was 63.3 (degrees × bins), but this failed to reach statistical significance ( p = 0.056).
Post-hoc tests on AUC error (Bonferroni corrected for multiple comparisons) following a significant Task*Stimulation Group interaction; F(2, 86) = 15.33, p < 0.001, ηp 2 = 0.26, revealed this trend was significant between the cerebellar and M1 group ( p = 0.019) while a strong trend was found for the cerebellar and sham group difference ( p = 0.059).
Further analyses of Adapt 2 showed relatively inconclusive results, however there appears to be a clear trend of reduced error following cathodal cerebellar TDCS in the joystick task.