In this experiment, it seems that applying TMS in the build-up of the difference between ERPs on Raf activity trials containing a figure stimulus and trials containing a homogenous stimulus affects performance, whereas stimulating at the peak of this difference in ERPs does not alter performance. This suggests that during build-up, the neural processes leading to figure border detection are more vulnerable to interference than when they have fully evolved. Stack versus frame: neural correlates of surface segregation To isolate signals related to surface segregation and to cancel out signaling related to figure border detection, we subtracted

activity evoked by frame stimuli from activity evoked by Inhibitors,research,lifescience,medical stack stimuli (as both stimuli have exactly the same amount of figure borders on exact the same locations, see “Task design”). Figure 6A shows a significant deflection between responses evoked Inhibitors,research,lifescience,medical by stack and frame stimuli appearing around 230 msec (significant interval: 227–313 msec, FDR corrected, P < 0.05) in the no TMS condition. This stack–frame difference was abolished in the early TMS condition (Fig. 6B), where behaviorally stimulation resulted in decreased stack and frame detection. In the intermediate TMS condition, responses evoked by stack and frame stimuli remained to significantly Inhibitors,research,lifescience,medical deflect from one another between 230

and 348 msec (FDR corrected, P < 0.05; see Fig. 6C). Due to interpolation of the EEG data, we were not able to test the difference between stack and frame stimuli when TMS was applied in the late time window (see “EEG measurements and analyses”). TMS stimulation in an early time window decreased figure detection and disrupted relatively early neural signaling associated Inhibitors,research,lifescience,medical with figure border detection. In addition, TMS in an early time window disrupted later occurring figure–ground signals related to surface segregation, while neural correlates of surface segregation remained intact when TMS was applied in the intermediate time window. To test

whether there is a difference between the different Inhibitors,research,lifescience,medical second TMS conditions, we compared the difference signals (responses evoked by stack stimuli minus responses evoked by frame stimuli) of three TMS conditions: the no TMS condition, the early TMS condition, and the intermediate TMS condition (the late TMS condition is missing due to data interpolation, see “EEG measurements and analyses”). For each TMS condition, we cumulated values of this difference signal in the time interval between 227 and 313 msec (based on the significant deflection of stack from frame stimuli in the no TMS condition). Figure 7A shows a clear reduction in the difference between ERPs on trials containing a stack and trials containing a frame stimulus when TMS was applied in an early time window in comparison with the no TMS condition (t = 2.97, P = 0.