Supporting information
Procedure to calculate reaction time
We hereby describe the procedure to determine the instant when the examinee initiates the reaction to the stimulus. In order to obtain a better understanding of the procedure, Fig.2S shows the screen coordinates of the examinee's dominant hand from the moment that the stimulus appears until the next stimulus is displayed.In this figure, n represents the number of samples taken by the Kinect sensor. For two stimuli separated by 2 s, Kinect provides between 60 and 80 samplesdepending on the current computer load, each containing 20 body points. In addition, f(n) represents the position of the dominant hand for each of the samples,M is the maximum value of f(n), nMdenotes the value of n where M is attained, and m represents the minimum value of f(n).
The hit reaction time for the stimulus being considered is calculated at the moment in which the next stimulus appears. To facilitate the procedure, the original signal was simplified as follows. Instead of considering the full signal, we reduced it by considering the segment f(n'), where n' takes the values from 1 to T; and T is given by the maximum value of n satisfying both, f(n) is less than (M+m)/2; and n is less than nM. The reduced signal is fitted by a piecewise function, which is composed of two linear components and relies on three parameters. This function is defined by:
where t represents the intersection point of both linear functions; a characterizes the first, constant, function and m is the slope of the second function. These parameters are fixed to minimize the mean square error between the signal and the function, namely,
The value of the parameters can be obtained using any optimization library that provides a gradient descent algorithm. The fitted function is displayed in Fig.2S.
Fig.1S Experimental set-up
Fig. 2S Computation of the reaction time. n represents the number of samples taken by the Kinect sensor from the stimulus onset. f(n) represents the position of the dominant hand for each of the samples.The solid line represents the trajectory of the dominant hand; the red dot shows the estimated moment in which the reaction starts; and the dashed line is the fitted piecewise function. The horizontal distance between the onset of the stimulus and the red dot corresponds to the reaction time
Fig.3STwo examples of the trajectory of the dominant hand. The solid line represents the trajectory of the hand. The dashed line indicates the moment when the next stimulus appears.(A) The hand of the participant returns toa resting position before the nextmovement; (B) the hand has not yet reached a resting position before the appearance of the next stimulus. n represents the number of samples taken by the Kinect sensor.f(n) represents the position of the dominant hand for each of the samples
Fig.4S Representation of the parameters used to calculate the Motion-based Impulsivity Index (MBI).The reaction of the participant to an X stimulus (dashed line) is preceded and followed by two non-X stimuli (solid lines). M is the maximum height attained in the interval that follows the previous or the posterior stimulus. In this example,the participant initiates the action but inhibits it almost immediately. The value of the MBIin this particular case was MBIx=0.35.f(n) represents the position of the dominant hand for each of the samples
Fig.5S Trajectories of the hands of two different participantsresponding to X stimuli. (A) Non-impulsive participant (low BIS-11 score); (B) impulsive participant (high BIS-11 score). Three different reactions to X stimuli can be appreciated for participant B: E3 (commission error), E10 (partial “hit”) and E12 (correct inhibition)
Fig.6SMotion-based Impulsivity Indexes(MBIs)of two participants responding to the X stimuli (corresponding to registers shown in Fig.5s). The participant (B) with a high BIS-11 score also shows high MBIs
1