Sensory-motor adaptation is (mostly) linear.

UNLABELLED Sensory-motor adaptation is usually conceived as an automatic process that maintains the calibration between motor plans and movement outcomes. Viewed as a mechanism that monitors disturbances and produces compensatory motor outputs, sensory-motor adaptation can be thought of as a filter. The first question one normally asks regarding filter performance is whether it is linear. We test homogeneity and additivity using a sinusoidal sensory-motor perturbation of reach endpoints (Landy & Hudson, VSS 2012). METHODS Subjects made center-out reaches on a tabletop with fixed starting point, and with target direction and distance chosen to fall randomly within an annulus centered on the start position. Feedback was shown on a frontoparallel display. During each reach, only the target was shown. Fingertip endpoint was shown (shifted) on reach completion. The amount of shift was either a single or the sum of two sinewaves (over trials), with a peak shift of never more than 6 mm. Homogeneity was tested by measuring the response to sinewave-perturbed endpoints following a single sinusoidal disturbance with amplitude A and, in a separate session, 2A. As a test of additivity, the adaptive response to two sinewaves (A and B, of different frequencies) measured separately was compared to the response to perturbation using their sum. RESULTS The sensory-motor adaptive response to perturbations in (Cartesian) x- and y-dimensions are consistent with linearity, in that the response to A and B sinusoidal perturbations measured in isolation predict the adaptive response to the 2A and the A+B perturbations. By the same criteria, the polar gain dimension also displays linearity. However, the polar angle dimensions displays a small but statistically significant deviation from linearity in its phase response in the additivity condition: its response to the A+B perturbation lags differently than predicted by its response to the A and B perturbations applied separately. Meeting abstract presented at VSS 2015.