Subjects were asked to perform each exercise at a self-selected pace and the velocity of movement during each exercise was not controlled.
We did not control for velocity of movement because we wanted to mimic the conditions under which these exercises are most likely to be performed in a clinical setting.
Subjects practiced each movement prior to recording until they could execute it correctly.
This typically took only 1 or 2 practice trials. Three trials of each movement were recorded and stored offline for further analyses.