Motor short-term memory measures as a function of methodology

Nine combinations (n=8) of methods for having subjects move their limb during presentation of criterion movements and return to the starting location were studied on a linear positioning apparatus. The results indicated, for absolute, variable, and constant errors, an interaction between the methods used in returning the limb from the endpoint of the criterion movement and the reproduction performance at two delay conditions of 0 and 20 sec. In addition, for absolute and constant errors, both the methods of moving the limb for the criterion movement and the methods of returning to the starting position interacted with reproduction of three movement amplitudes (15, 30, and 45 cm). For variable error only the methods of returning to the starting position differentially influenced performance at the three amplitudes. These results indicated that methodological variables can interact with independent variables in the short-term memory paradigm and design of future investigations should take this into account.