Fig. 1. (A) Areas of activation (red) and deactivation (blue) for all frequencies of GVS used in this study (main effect of GVS). Results obtained by group analysis are superimposed onto transverse sections of a standard brain template (P < 0.001, uncorrected). Activated areas for all stimulation frequencies compared to rest condition included the supramarginal gyrus, lateral sulcus, superior temporal gyrus, anterior and posterior insula, inferior and middle frontal gyri, anterior cingulum, and precentral sulcus. Deactivated areas were found bilaterally in large parts of the precuneus, precentral gyrus adjacent to central sulcus, middle occipital gyrus, middle temporal gyrus, parahippocampal gyri extending into the hippocampus, medial frontal gyrus, as well as in the middle and superior frontal gyri. (B) Further activations were located in the pontine brainstem bilaterally, representing the medial longitudinal fasciculus (orthogonal sections at MNI coordinates x,y,z = 6, −34, −34).

Fig. 2. Borders of the activation clusters obtained by stimulation with six different frequencies of GVS. Results obtained by group analysis are superimposed onto transverse sections of a standard brain template (P < 0.001, uncorrected). All stimulation frequencies led to activations of comparable cortical areas. Stimulation with the highest frequency used (5 Hz) caused additional activations in the left postcentral gyrus.

Fig. 3. Borders of the deactivation clusters obtained by stimulation with different frequencies of GVS. Results obtained by group analysis are superimposed onto transverse sections of a standard brain template (P < 0.001). During stimulation with 0.1 Hz, deviations from the common deactivation pattern (Fig. 1) were found, i.e., deactivations were only in the postcentral gyri bilaterally.

Fig. 4. Brain areas showing a significant linear relationship of BOLD signal with the frequency of GVS (correlation analysis) obtained by group analysis and projected onto a standard brain template. Orthogonal sections through (A) left supramarginal gyrus, (B) right posterolateral thalamus, (C) vermal lobule III, and (D) glass-brain view showing the additional clusters in vermal lobule VI (declive) and insular gyrus V (P < 0.001).

Fig. 5. Brain areas showing a significant linear relationship between BOLD signal and the negative inverse of the GVS frequency (brain areas obtained by correlation with a factor that is proportional to the perceived change in position), obtained by group analysis and projected onto a standard brain template. Orthogonal sections showing (A) a cluster covering parts of the left posterolateral thalamus, putamen, and retroinsular region, (B) cerebellar vermis lobule IV/V, (C) right posterolateral thalamus and insular gyrus V/transverse temporal gyrus, and (D) glass-brain view showing the additional clusters in the left supramarginal gyrus and right calcarine sulcus.

Summary of subject reports on perceptions during AC-GVS

The number of subjects who gave the corresponding answers is given for all six stimulation frequencies.

Results of the comparison GVS versus rest for all stimulation frequencies used in the experiments (P < 0.001, uncorrected)

Cluster size, anatomical description, Brodmann areas, MNI coordinates, and t values are given for the maximum (bold) and local submaxima of each cluster. Positive x, y, and z coordinates indicate locations right, anterior, and superior to the middle of the anterior commissure, respectively.