In order to have an accurate assessment of task performance in the fMRI environment, the timing of the stimulus and response mode of the RGS were adapted in accordance with the fMRI scanning requirements and timings (Fig. 2). Subjects were presented with image sequences generated by the VR machine, showing the arms of an avatar in a green landscape following the standard RGS protocol. Colored balls moving at various speeds and angles relative to the subject approached the avatar in the right or left visual field from the horizon in a first or third person perspective (Fig. 1). When a ball approached a virtual hand, the subjects had to press a button

with the index finger of their corresponding right or left hand. The time window for successfully catching the ball was 1000 ms (500 ms before and 500 s after crossing BMS-777607 the flight direction of the ball and the path of the catching hand). This was chosen to account for the fact that, in the RGS, the avatar’s position is fixed, whereas in real life

one would be able to move one’s body forwards or backwards in order to catch a flying ball. When the ball was missed, it passed by and left the field of view. When the ball was caught, the subjects could view the caught ball for the subsequent 8 s to let the hemodynamic Protein Tyrosine Kinase inhibitor response return to baseline. After a short blank display of the landscape, the next trial

began with a reappearance of the avatar. There were 24 repetitions of each trial, and each trial lasted 24 s. In a mixed event-related experimental design, subjects were presented with three different experimental conditions in separate Benzatropine scanning sessions in a pseudo-random order (Fig. 2): (i) action condition – the subjects were required to actively catch the balls by pressing the corresponding button (left/right) with their index finger; (ii) observation condition – the subjects were required to observe the avatar catching the balls; and (iii) imagination condition – the balls disappeared during their flight towards the avatar, and the subjects were required to imagine catching the ball at the right moment; for balls on the right, they had to indicate this by a right button press, and vice versa. Passive viewing of the landscape served as the baseline. Behavioral data were analysed with spss software (Version 20; IBM, Armonk, NY, USA). Prior to statistical analysis, data were tested for normal distribution with the Kolmogorov–Smirnov test. In case of a deviation from normal distribution, median scores were calculated, and the non-parametric Wilcoxon test was used to compare data (corrected α = 0.008). Imaging data were analysed with the brainvoyager qx software package (Brain Innovation, Maastricht, the Netherlands).