Individual differences in TMS sensitivity influence the efficacy of tDCS in facilitating sensorimotor adaptation

Abstract

Background: Transcranial direct current stimulation (tDCS) can enhance cognitive function in healthy individuals, with promising applications as a therapeutic intervention. Despite this potential, variability in the efficacy of tDCS has been a considerable concern.

Objective: /Hypothesis: Given that tDCS is always applied at a set intensity, we examined whether individual differences in sensitivity to brain stimulation might be one variable that modulates the efficacy of tDCS in a motor learning task.

Methods: In the first part of the experiment, single-pulse transcranial magnetic stimulation (TMS) over primary motor cortex (M1) was used to determine each participant's resting motor threshold (rMT). This measure was used as a proxy of individual sensitivity to brain stimulation. In an experimental group of 28 participants, 2 mA tDCS was then applied during a motor learning task with the anodal electrode positioned over left M1. Another 14 participants received sham stimulation.

Results: M1-Anodal tDCS facilitated learning relative to participants who received sham stimulation. Of primary interest was a within-group analysis of the experimental group, showing that the rate of learning was positively correlated with rMT: Participants who were more sensitive to brain stimulation as operationalized by our TMS proxy (low rMT), showed faster adaptation.

Conclusions: Methodologically, the results indicate that TMS sensitivity can predict tDCS efficacy in a behavioral task, providing insight into one source of variability that may contribute to replication problems with tDCS. Theoretically, the results provide further evidence of a role of sensorimotor cortex in adaptation, with the boost from tDCS observed during acquisition.

Keywords: Individual differences; Sensorimotor learning; TMS; rMT; tDCS.

Figure 1

Experiment Procedure. The experiment was run in a single session, divided into two stages. At the start of the experimental session (Stage 1), TMS was used to determine the resting motor threshold (rMT) for each participant, the minimum stimulation intensity required to evoke MEPs of at least 50 μV peak-to-peak amplitude in FDI. This measure served as a proxy of sensitivity to brain stimulation, under the assumption that there exists a negative correlation between rMT threshold and sensitivity to tDCS (e.g., lower rMT = higher sensitivity). Participants were then tested on a visuomotor adaptation task (Stage 2), composed of baseline, perturbation, and washout phases. In the perturbation phase, the cursor feedback was rotated 30° in the CCW direction. To examine the effects of early and late learning and forgetting the perturbation and washout phases were each divided into two blocks (Adapt1 and Adapt2; Post1 and Post2). Trials were grouped into epochs by averaging the data over eight consecutive trials (1 reach/target). tDCS stimulation (gray line) over M1 was applied during the end of the baseline phase and during most (or all) of the perturbation phase.

Figure 2

Flowchart depicting the primary data analysis pipeline.

Figure 3

Anodal tDCS over M1 induces faster learning in visuomotor adaptation. A: Angular error for each group across the reaching task. Data points show mean for epochs of eight trials (1 reach/target). Positive errors indicate movements deviated in the clockwise direction of the target, thus counteracting the perturbation during the adaptation epochs (Adapt1, Adapt2). Negative errors in the washout phases (Post1, Post2) indicate aftereffects from adaptation. B: Mean movement time, presented in the same format. C-F: Mean angular error during the early and late phases of the perturbation and washout phases. In the washout phase, the mean angular errors are shown with respect to 0, and not with respect to the end of the perturbation phase. * = p < 0.05.

Figure 4

Forgetting in the absence of feedback is similar for the Anodal and Sham groups. A: Change in heading angle during the washout phase, relative to a baseline (0°), set on an individual basis as the mean heading direction over the last five epochs of the perturbation phase. Data points show mean for epochs of eight trials (1 reach/target). Larger values indicate more forgetting of the adapted internal model acquired during the perturbation phase. B, C: Mean forgetting values during the early and late phases of the washout phase.

Figure 5

Efficacy of tDCS during adaptation varies as a function of sensitivity to brain stimulation. A: For illustrative purposes, the data from Figure 2A are replotted with the anodal group divided by a median split into those with a low rMT threshold (mean ± SD, 36 ± 3, high sensitivity) and those with a high rMT threshold (mean ± SD, 49 ± 7, low sensitivity). All statistics used continuous measures of rMT. B, C: Correlations between rMT and angular error during the second half of the perturbation phase (Adapt2) and washout phase (Post2) for the anodal and sham groups. The gray lines show the mean (± SE) values of the sham group. β = standardized regression slope. * = p < 0.05.

Figure 6

Permutation test identifying epochs in which efficacy of tDCS varies as a function of sensitivity to brain stimulation. Functions for each individual in the M1-Anodal (A) and Sham (B) groups on the visuomotor adaptation task, color coded by rMT, ranging from blue (low rMT, high sensitivity) to purple (high rMT, low sensitivity). In the M1-Anodal group (A) the black horizontal line indicates consecutive epochs in which a cluster-based permutation test revealed significant correlation between rMT and performance. The insert shows the correlation between rMT and average angular error in the identified cluster. Angular error data were smoothed using a 3-epoch window to allow better visualization; all statistical analyses were conducted on unsmoothed data.