Experimental protocol.

<p>The main experiment was divided in two parts. The first part consisted in phosphene mapping. The subjects received 7 magnetic pulses at 70% output intensity. They were then asked to draw on the screen the perceived phosphene using the mouse. This area, which we called the “phosphene region”, was during the second part –the visual search experiment– the region where the stimuli were presented. This ensured that the TMS effect was specific to the stimulated area on the cortex. The control used was the symmetric zone relative to the vertical midline, which we called the “control region”. The visual search experiment comprised blocks of either a serial task (L versus T) or a parallel task (L versus +). The SOA (Stimulus Onset Asynchrony) between the search array and the masks was adjusted individually in each task to fix performance at around 75%. A double-pulse (25 ms interval) was applied at random delays between 100 and 450 ms after the onset (8 different delays by steps of 50 ms). Every 72 trials (2 blocks of 36 trials), the experiment restarted from the first step to make sure that the phosphene had not changed.