Fig 1.
(a) Experimental setup. Participants were shown videoclips of correct and incorrect actions related to the production of clay coils, while a TMS-EEG co-registration was performed. A TMS-induced MEP from a representative participant is shown in the upper box, while an EEG recording is shown in the lower box. (b) A TMS figure-of-eight coil was placed over the participant’s left M1 to investigate corticospinal excitability during action observation. (c) TMS‐induced MEPs from the FDI and OP hand muscles were recorded through pairs of surface electrodes. (d) A TMS-compatible elastic cap with 64 Ag/AgCl electrodes was used to record cortical electrical activity. (e) Timeline. A fixation cross appeared for 1500 ms, followed by videoclips presentation for 1557 ms. Timing of single-pulse TMS stimulation was set at 759 ms for all conditions (i.e., 165 ms after the onset of the erroneous actions).
Fig 2.
The figure shows the GFP time-locked to the onset of the video clip (A) and to the delivery of the TMS pulse (B). The topoplots show the EEG signal averaged in the time-windows of interest (GFP > 3 SD) across participants. Specifically, Topoplot C shows the grand average EEG in the 478 ms to 682 ms time-window locked to clip onset. The cluster of channels used for inferential analysis (PO7, PO3, PO4, PO8, O1, Oz, O2) is indicated by black dots. Topoplots D and E show the grand average EEG in the 74 ms to 230 ms time-window following the TMS pulse. The two clusters of channels used for inferential analysis (Fz, F2, F4, FC1, FC2, FC4, Cz, C2 in D, and FT7, T7, C5, TP7, CP5, CP3, P7, P5 in E) are indicated by black dots.
Fig 3.
Interact plot for the LMM of first dorsal interosseus’MEPs.
The figure shows the estimated marginal means of the normalised MEP amplitude for the GROUP x SESSION x TECHNIQUE x CORRECTNESS interaction. Error bars represent standard errors.
Table 1.
Summary output of the LMM of first dorsal interosseus’ MEPs.
Table 2.
Summary output of the LMM of opponens pollicis’ MEPs.
Fig 4.
Interact plot for the EEG analysis of the first time-window of interest (action observation).
The figure shows the estimated marginal means of the ERP amplitude averaged across the identified channels (PO7, PO3, PO4, PO8, O1, Oz, O2) and time-points (from 478 ms to 682 ms) of interest for the 2 groups x 2 sessions interaction. Error bars represent standard errors.
Table 3.
Results of the 2 groups x 2 sessions mixed ANOVA of the first time window of interest (action observation).
Table 4.
Results of the 2 groups x 2 sessions x 2 correctness mixed ANOVA of the second time window of interest (TMS pulse deliver) in the first cluster (Fz, F2, F4, FC1, FC2, FC4, Cz, C2).
Table 5.
Results of the 2 groups x 2 sessions x 2 correctness mixed ANOVA of the second time window (TMS pulse deliver) of interest in the second cluster (FT7, T7, C5, TP7, CP5, CP3, P7, P5).
Fig 5.
Comparison between coils made before the training and at the end of the training.
a-c: Coils produced by three different participants (black bar is 1 cm).
Fig 6.
Significant differences in the shape of the coils.
Significant differences were identified through MANOVA, performed both on the elliptic Fourier coefficients (p = <0.001) and the PCA scores (p = <0.001). More specifically, significant differences between the coils produced before and after the training were shown performing a MANOVA Pairwise, which returned a p value of <0.001 on the PCA scores.
Fig 7.
Comparison between experimental vessels produced at the beginning and at the end of the training.
a-c: Vessels shaped at the beginning of the training; d-f: Vessels shaped at the end of the training. The vessels made by three diverse participants were selected to show the improvements achieved at the end of the training (black bar is 1 cm).