Supplementary material - Surface EMG
Materials and methods
Participants
Six neurologically intact participants from the University of British Colombia volunteered for this investigation (4 male, 2 female, mean age = 23.1, SD = 3.2 years). All participants were self-declared right-handers with normal or corrected to normal vision. None of the sample from this experiment has participated in the main experiment.
Materials and apparatus
Participants were seated in a chair in front of a target board at a distance of 45 cm from a 5 cm yellow square target. Participants began each trial with their right arm in either an ‘elbow’ posture (~10 deg shoulder flexion, ~40 deg shoulder abduction, ~90 deg elbow flexion) or a ‘shoulder’ posture (~0 deg should flexion, 0 deg shoulder abduction, 90 deg elbow flexion). These different postures were included to account for the various strategies that participants may have utilized when reaching toward the targets in the head-free, unconstrained manner of Experiment 1 (described in more detail in the ‘Procedure’ section below).
Accelerometry.Due to differences between the equipment available to the testing locations, infrared recording of kinematics was replaced by accelerometry in this portion of the study. An accelerometer (V94-41, Coulbourn Instruments, Whitehall, Pennsylvania, USA) was taped to the right index finger of each participant to determine movement onset time. Participants were asked to remain in an upright position with their back resting against the chair throughout the protocol and for every pointing movement.
Electromyography. Four pairs of 4 mm Ag/AgCl surface electrodes were placed on the bellies of the right biceps brachii (BB), the anterior deltoid (AD), the triceps brachii (TB), and superior trapezius (ST). The first pair was placed on the belly of the BB at midshaft of the humerus and the second pair was placed on the belly of the AD. The third pair was placed on the belly of the TB at midshaft of the humerus and the fourth pair was placed on the ST. The electrodes in each pair were placed 2 cm apart (center to center). Also, each pair of electrodes utilized a ground reference electrode which was placed on the olecranon process of the ulna. The surface EMG signal was amplified (1000x) with an isolated bioamplifier (Coulbourn Instruments, Whitehall, Pennsylvania, USA), band-pass filtered (13Hz-1kHz), sampled at 1024 Hz and converted from analog to digital format (1401 Plus A/D converter, CED, Cambridge, England). EMG from all muscles was smoothed and averaged (root mean square) over a window that corresponds to the acceleration record for each pointing movement. Any deflection detected at the beginning of the trial, relative to baseline, was defined as EMG initiation.
Procedure
Participants started the experiment seated, with their right hand on the start point, 45 cm away from the target. The experimenter started a trial with a keyboard press, playing a short auditory tone from the computer which acted as the cue to begin the reach. Upon hearing this tone, participants were asked to reach as quickly and accurately as possible with their right arm to the target infront of their hand. When seated in the ‘elbow’ posture, participants executed this task using an elbow extension strategy, thus requiring the trapezius (arm elevation) and triceps (elbow extension) to be the prime movers. In the ‘shoulder’ posture, participants reached the target via shoulder flexion (anterior deltoid) and elbow extension (eccentric bicep). Participants each completed 10 ‘elbow’ reaches and 10 ‘shoulder’ reaches. During all pointing movements, EMG was recorded from the ST, TB, AD and BB muscles, despite the task-induced variation in prime-movers: ST and TB were analyzed for ‘elbow’ trials while BB and the AD were analyzed for ‘shoulder’ trials. Reaching posture was ordered in a blocked fashion; block order was randomized across participants.
Data reduction and analysis
The expected trial sequence was thus: (1) keyboard press by the experimenter followed by (2) initial EMG activity and, finally, (3) acceleration of the limb. A trial was excluded from subsequent analysis if it deviated from this sequence. Motor time (i.e., neuro-mechanical delay) was measured as the difference between the time of EMG initiation and the initial acceleration (i.e., movement onset).
Results and discussion
The mean motor time (the EMG initiation time subtracted from the time of movement onset) across all the muscles in all the strategies was 41 ms (SD = 9 ms). This value was fairly consistent across the recorded muscle groups (see supplementary Table 1). Therefore, even if participants in the two experiments were forced to use slightly differing postures or strategies to complete their reaches in the different testing locations, it is unlikely that systematic variation exists between muscle activation and movement onset that could invalidate the conclusion that gating precedes movement onset. In fact, the motor delay values for all recorded muscles under all conditions are lower than the gating onset value (55 ms) from the main experiment (see also Cavanaugh & Komi, 1979). Thus, the tactile gating described in the main text precedes even the activation of the upper arm muscles required to undertake the reach – the first likely source of reafferance that could affect detection performance on distal appendages. Therefore, we can conclude that peripheral reafferance is not a viable contributor to the sensory suppression in our task. Rather, it is likely that the tactile gating demonstrated in the current work has a central origin related to efference copy (Voss et al., 2008).
Supplementary Table 1. Individual’s mean motor times (time from first muscle activity to actual movement onset) for all recorded muscles.
P / BB (ms) / TB (ms) / AD (ms) / T (ms)1 / 37.0 / 31.9 / 22.6 / 48.8
2 / 57.2 / 32.7 / 51.2 / 46.9
3 / 53.5 / 40.7 / 33.4 / 45.9
4 / 43.1 / 34.7 / 29.0 / 48.9
5 / 30.0 / 42.9 / 38.8 / 50.8
6 / 52.7 / 37.3 / 35.5 / 48.9
Mean / 45.6 / 36.7 / 35.1 / 48.4