Abstract
There is considerable interest in movement direction encoding in the central nervous system, in part because of potential applications to the development of neural prosthetic devices. Most studies in this field focus on single movements, with little attention given to the interactions that may occur between successive movements. The purpose of our study was to determine, independent of movement effector system, if a prior movement results in directional reaction time biases for subsequent movements. In our experiments, participants made two consecutive eye movements or two consecutive arm movements in directions indicated by arrows presented at fixation. In experiment one, two target locations were shown during each trial. Here, movements were faster when offset by 90 or 180 degrees from the first movement (relative to movements back to the original target location). This pattern is consistent with 'inhibition of return' (IOR) typically found for repeated movements made in the same direction. In experiment two we presented four possible target locations. Here we found that arm and eye movements were faster only when offset by 90 degrees from the initial movement; the well-established advantage for movements offset by 180 degrees was eliminated. Our results reveal an effect of set size (i.e. 4 vs. 2 possible movement locations) on the spatial gradient of RTs for consecutive movements. That similar results were found for eye and arm movements suggests the existence of a common motor programming principle that may be useful for computational models attempting to 'decode' neural signals for the implementation of neuroprosthetic devices.
