Fig. 2.

Dependence of pursuit velocity and acceleration with target velocity. A and E: the mean horizontal (H) velocity for both subjects is shown time-locked to the onset of target motion for a set of different velocities sampled in a single behavioral session (velocity indicated by color with thick lines indicating a Gaussian smoothing, σ = 20 ms). The upward arrow indicates the beginning of target motion. Velocity and time calibration bars apply for both A and E. B and F: the closed-loop pursuit velocity measured in the interval from 400 to 600 ms is shown as a function of target velocity averaged across sessions (±1 SD). The pursuit velocity shows a linear dependence for slower (<10°) target velocities. Each gray dot indicates the average value computed from trials completed at that velocity in a single behavioral session (marmoset P, 9 sessions; marmoset B, 3 sessions). Points for the example sessions shown in A and E are labeled in green. C and G: the closed-loop gain as a function of stimulus velocity reveals left and right asymmetries. D and H: the open-loop acceleration measured in the interval from 200 to 300 ms is shown as a function of target velocity averaged across sessions (±1 SD). Same conventions as in B and F.