Fig. 2

Behavioral effects induced by tSMS of the SMA. a Schematic representation of the choice-reaction time tasks. All tasks start with a fixation screen with four empty circles, followed by a cue screen (S1) and a GO screen (S2). In the fully-cued (FC) task, the cue is an empty circle in one position (bottom right in the example) representing the target button, and the GO is a filled circle in the same position, triggering the response. In the uncued-compatible (UC) task, the cue screen is equal to the fixation screen, so no target information is provided before the go signal (top right in the example). The uncued-incompatible (UI) task is identical to the UC task, but the subject is instructed to respond to the target that is opposite to the one indicated by the go signal (in the example the go signal indicates top left and the subject needs to respond to bottom right). b Average reaction times (RTs), initiation times (ITs), and movement times (MTs) for the three tasks after real or sham tSMS. c Corresponding error rates. Error bars represent standard errors (n_real = 20 and n_sham = 22). *p < 0.05, **p < 0.01. d Schematic representation of the drift-diffusion model (modified from Fig. 16 by Murata et al.¹⁵², used under CC BY 4.0). Reaction times are modeled as generated by a noisy process that integrates evidence with a positive drift rate v from a starting point z (z = a/2 in the EZ-diffusion model) until reaching either a higher threshold a, leading to a correct decision, or a lower threshold 0 (i.e. zero), leading to an error decision. The non-decision time T_er includes both perceptual processes occurring before decision onset and motor processes occurring after the decision. The EZ-diffusion model estimates a, v, and T_er from the overall probability of correct responses, the mean, and the variance of the reaction times. e Distributions of initiation times pooled across all subjects separately for the three tasks and for the groups that received tSMS (red) or sham stimulation (blue). Distributions were smoothed for illustration purposes. f Corresponding parameters estimated with the EZ-diffusion model, using bootstrapping techniques. On each box, the central mark represents the median, the edges of the box are the 25th and 75th percentiles, and the whiskers extend to the largest and smallest values within 1.5 times the inter-quartile range from the edges. *P < 0.05, **p < 0.01. Data for Fig. 2 are provided at https://osf.io/n3au4/