Changes in control of saccades during gain adaptation

Abstract

In a typical short-term saccadic adaptation protocol, the target moves intrasaccadically either toward (gain-down) or away (gain-up) from initial fixation, causing the saccade to complete with an endpoint error. A central question is how the motor system adapts in response to this error: are the motor commands changed to bring the eyes to a different goal, akin to a remapping of the target, or is adaptation focused on the processes that monitor the ongoing motor commands and correct them midflight, akin to changes that act via internal feedback? Here, we found that, in the gain-down paradigm, the brain learned to produce a smaller amplitude saccade by altering the trajectory of the saccade. The adapted saccades had reduced peak velocities, reduced accelerations, shallower decelerations, and increased durations compared with a control saccade of equal amplitude. These changes were consistent with a change in an internal feedback that acted as a forward model. However, in the gain-up paradigm, the brain learned to produce a larger amplitude saccade with trajectories that were identical with those of control saccades of equal amplitude. Therefore, whereas the gain-down paradigm appeared to induce adaptation via an internal feedback that controlled saccades midflight, the gain-up paradigm induced adaptation primarily via target remapping. Our simulations explained that, for each condition, the specific adaptation produced a saccade that brought the eyes to the target with the smallest motor costs.

Figure 1.

Group average, trial-by-trial saccade kinematics for gain-down adaptation versus corresponding control. The dotted lines represent set breaks, and the solid line divides the baseline (error-clamp) trials from the adaptation/control trials. In red are the mean adaptation kinematics and in black the mean control (n = 5). Error bars are SEM. A, Saccade amplitude, peak velocity, duration, time with respect to saccade onset at which the velocity reaches its peak, peak deceleration, and peak acceleration. To highlight the within-set structure of the data, bin size was two trials for the first bin, and then four trials, and then six trials for all subsequent bins for each set. B, Group average speed profile. This represents the average kinematics for the last 40 trials of each of the last four sets. C, Group average acceleration profile.

Figure 2.

Group average, trial-by-trial saccade kinematics for gain-up adaptation versus corresponding control. The format is same as in Figure 1. A, Saccade amplitude, peak velocity, duration, timing at peak velocity, peak deceleration, and peak acceleration. B, Group average speed profile. This represents the average kinematics for the last 40 trials of each of the last four sets. C, Group average acceleration profile.

Figure 3.

Comparison of saccade parameters in the adaptation versus control condition. Data represent the last 40 trials of the last four sets. Error bars are SEM.

Figure 4.

Saccade latencies. The format is the same as in Figure 1. Error bars are SEM. A, B, Saccade latencies in the gain-down and gain-up paradigms. The dotted lines represent set breaks, and the solid line marks the end of error-clamp trials. C, Data represent the last 40 trials of each of the last four sets. Error bars are SEM. Both adaptation conditions produced significantly longer latencies than the control condition, although the difference was smaller in gain-up.

Figure 5.

Simulation of adaptive control of saccades. A, Gain-down paradigm. The black trace is the speed profile of a 15° control saccade in a baseline condition. The blue trace is the saccade to a remapped target at 10° [i.e., at saccade onset the target is remapped by −5° (Δr = −5°)]. This is the saccade profile expected in the control condition of group 1. The green trace is the speed profile of a saccade in which the forward model had adapted, producing a 10° saccade in response to a 15° target (Δr = 0°; β = −0.5). Gain-up paradigm. The black trace is the speed profile of a 15° control saccade in preadaptation condition. The blue trace is the saccade to a remapped target at 20° (Δr = +5°; β = 0). This is the profile expected in the control condition of group 2. The green trace is the speed profile of a saccade in which the forward model had adapted, producing a 20° saccade in response to a 15° target (Δr = 0°; β = 0.25). B, Corresponding profiles of the motor commands u_t (torque) for the saccades shown in A. Units are newton · meters. C, Sum of the squared motor commands as a function of time: Σ_tu_t². Forward model adaptation is the least costly policy to reduce saccade error in a gain-down paradigm. Target remapping is the least costly policy to reduce saccadic error in a gain-up paradigm.