Behavioral and Functional Brain Activity Alterations Induced by TMS Coils with Different Spatial Distributions

Electric field distributions induced by the TMS coils when placed over the right DLPFC, obtained in a phantom and overlaid on MRI images. The colored maps describe the absolute magnitude of the electric field for the various TMS coils. These were measured in a phantom model of the human head with an equivalent stimulation amplitude to that used in this study (i.e., stimulator output equivalent to 120% of an average motor threshold). In the color scale, red indicates a field magnitude above neural activation threshold (100 V/m), while white and yellow indicate field magnitude below the threshold. In the top left inset, the H-Coil is illustrated by the blue wire, while the sham coil is illustrated by the white wire wound on a cylindrical former. Anatomical images are shown in radiologic coordinates.

Group activation maps obtained while performing the IGT in the PRE (i.e., prestimulation) condition. Statistical maps (task > rest) are overlaid on a rendered cortical surface (A) or orthogonal slices (B) for a threshold of p = 0.05 with FWER correction. (L/R indicates left/right hemisphere in A and left/right sides of the image in B, respectively.)

A, Effect of stimulation group on functional activation from the significant interaction (TIME × COIL) from the complete factorial model (TIME × COIL × ROI). Activation, shown on the y-axis, is represented by the SPM contrast image for task > rest. The plots thereby indicate the stimulation-induced activation for the three stimulation groups over the two time points (PRE and POST) and the corresponding slopes characterize the sign of this activation-induced change. The ROIs were in areas of the putative reward network, in the right OFC, right insula, and pgACC. Extended Data Figure 4-1 in the extended material shows the corresponding plot for changes in the behavioral performance metric measured during the IGT task. B–D, Exploratory analysis of TIME × COIL interaction in each of the individual ROIs, the right orbitofrontal cortex (B), the pregenual ACC (C), and the right insular cortex (D). The same pattern of opposing slopes shown in A is consistently observed in these individual ROIs that represented a nonsignificant trend following correction for multiple comparisons (right OFC ROI: F_(2,17) = 3.52, p_uncorrected = 0.048, p_corrected = 0.14, η²_p = 0.30^a13; pgACC ROI: F_(2,17) = 3.68, p_uncorrected = 0.047, p_corrected = 0.14, η²_p = 0.31^a14; right insular cortex: F_(2,17) = 1.9, p_uncorrected = 0.18, p_corrected = 0.54, η²_p = 0.18^a15). Errors bars are ±SEM, * indicates a significant interaction (p < 0.05).

Explorative analysis showing intercoil pattern of slopes, defined as the change in the activation contrast of task > rest between the POST and PRE scans, i.e., as POST-PRE. Two representative orthogonal slices are shown with Talairach slice positions: axial (left): z = 20 cm; sagittal (right): x = −4 cm. Yellow voxels represent the stereotypical pattern of activation seen across the three selected ROIs in the reward system, i.e., with a positive slope in the H-Coil stimulation group and a negative slope in the Figure-8 coil and Sham stimulation groups. Blue voxels represent the nondifferential pattern of activation (namely, negative slopes for all stimulation groups).