Saccades and pursuit: two outcomes of a single sensorimotor process

Abstract

Saccades and smooth pursuit eye movements are two different modes of oculomotor control. Saccades are primarily directed toward stationary targets whereas smooth pursuit is elicited to track moving targets. In recent years, behavioural and neurophysiological data demonstrated that both types of eye movements work in synergy for visual tracking. This suggests that saccades and pursuit are two outcomes of a single sensorimotor process that aims at orienting the visual axis.

Figure 1. Oculomotor strategies in visual tracking

For each panel, eye and target position are represented versus time. A, the eye velocity saturates and cannot match target velocity. Therefore, catch-up saccades are triggered (thick traces). Target velocity is 50 deg s⁻¹. B, the oculomotor system predicts the time of target motion onset and the eyes begin to move before the target does. The inset provides a zoom around target motion onset. The arrows highlight the advance in position of the eye with respect to the target at its onset. C, a catch-up saccade (thick blue trace) is executed to suppress the position error around 200 ms after target motion onset. For B and C, target velocity is 18 deg s⁻¹.

Figure 2. Position input to the smooth pursuit system and velocity input to the saccadic system

A, schematic representation of the protocol: the target follows a linear path and a second target (red star) is flashed during ongoing smooth pursuit. The position of the moving target at the time of the flash is represented by the grey disk. For the illustrated trial, the target velocity is 27 deg s⁻¹ along the //axis. The target is flashed for 10 ms at 2.6 deg perpendicularly to the target path (along the ⊥ axis). B, representation of the eye velocity perpendicular to the target trajectory (⊥ axis) versus time. Eye velocity perpendicular to the target path increased after the apparition of the flash (red vertical line). C, a curved saccade (blue circles) occurring after both position and velocity steps of the target (black dashed lines) is presented in two dimensions. Target positions just before and after the target step are represented by the open and closed squares, respectively. Before and after the saccade, smooth pursuit eye movements (continuous blue lines) are present to track the target. The grey dotted lines represent isochronic lines that connect eye and target at the same moment in time. Red arrows give the position error (PE) and retinal slip (RS) orientation before the saccade. The saccade is elicited with an initial orientation close to the orientation of PE and final orientation close to the orientation of RS.

Figure 3. Trigger mechanism and extraretinal interactions

A, illustration of the eye-crossing-time parameter which corresponds to the time that the eye trajectory would need to cross the target at constant velocity (position versus time representation with T: target; E: eye; PE: position error; RS: retinal slip; T_xe: eye-crossing-time). To evaluate the eye-crossing-time at a certain instant in time (red dot), we compute the tangent to the eye trace (red line). The distance from the defined instant in time and the intersection between target and tangent yields the eye crossing time. B, C and D, representation of eye and target position versus time for three different trials that illustrate the trigger mechanism of the catch-up saccade (thicker traces). About 100 ms (red dots) before the saccade, we have represented the eye-crossing-time parameter following the definition in A. In trials B and C, the eye-crossing-time is outside the smooth zone (B: < 40 ms and C: > 180 ms). For trial D, there is no saccade because the eye-crossing-time remains within the smooth zone. E, illustration of the compensation for a decrease in the smooth eye velocity by the saccadic system during the occlusion of a pursuit target. The upper panel represents position versus time while the lower panel gives the velocity versus time. During the occlusion (grey areas), the smooth eye velocity tends to decrease (thick blue trace on velocity panel). To compensate for this decrease, saccades are generated during the occlusion (thick blue traces on the position panel). The collaboration between saccades and pursuit yields a small position error at target reappearance (at the right of the grey area on the upper panel). F, representation of how the saccadic system can integrate the smooth eye velocity to account for the smooth eye displacement while localizing a stationary target (same protocol as in Fig. 2: position input to the smooth pursuit system). During ongoing pursuit (continuous blue line for the eye and dashed black line for the target), a second target is flashed (red star). The position of the eye at the time of the flash is represented by the grey disk. A first saccade (blue circles) is triggered with a short latency toward the flashed target and the saccade vector matches the position error at the instant of the flash (grey dashed line). Subsequently, a second saccade (red circles) is triggered and accounts for the smooth eye displacement (SED, red arrow) following the flash.

Figure 4. Model of the synergies between saccades and pursuit and brain circuitry subserving saccades and pursuit

A, the retinal inputs are processed into retinal slip (RS, blue lines) or position error (PE, red lines). In the absence of sensory inputs, RS is estimated from efference copies (Eff. copy pursuit) of the pursuit commands whereas PE is estimated from efference copies of both saccade (Eff. copy saccade) and pursuit commands. Estimates of position and velocity errors are sent to both saccadic and pursuit systems, which process them into motor commands. The minor role of the position input to the smooth pursuit system is represented by the red dash lines. The trigger uses these inputs to decide whether a catch-up saccade should be triggered. Pursuit and saccadic commands are then summed up and conveyed to the premotor structures. B, colours were assigned to each brain area following its involvement in each motor pathway (saccade- or pursuit-related activity; red or blue disks). BG: basal ganglia; CB: cerebellum; FEF: frontal eye field; LIP: lateral intraparietal area; MST: medial superior temporal area; MT: middle temporal area; PON: pontine nuclei; SC; superior colliculus; SEF: supplementary eye field; TH: thalamus.