Interactions between Saccades and Smooth Pursuit Eye Movements in Marmosets

Abstract

Animals use a combination of eye movements to track moving objects. These different eye movements need to be coordinated for successful tracking, requiring interactions between the systems involved. Here, we study the interaction between the saccadic and smooth pursuit eye movement systems in marmosets. Using a single-target pursuit task, we show that saccades cause an enhancement in pursuit following a saccade. Using a two-target pursuit task, we show that this enhancement in pursuit is selective toward the motion of the target selected by the saccade, irrespective of any biases in pursuit prior to the saccade. These experiments highlight the similarities in the functioning of saccadic and smooth pursuit eye movement systems across primates.

Keywords: eye movements; marmoset; saccades; smooth pursuit.

Figure 1.

Single-target pursuit task examples. A, The scheme for single-target step-ramp pursuit task. Animals begin by fixating at a target at center of the screen. The target then makes either a positive or negative step and starts to move. B, C, Example pursuit traces from animals. Top row is showing position trace in the direction of target movement. Bottom row is showing velocity trace in the direction of target movement. Black lines indicate the eye traces. Colored blue and red lines indicate the target motion, along vertical (blue) or horizontal (red) direction. Inset schematics show which direction the target movement is in for each example. Shaded regions indicate the 30 ms windows used to compute pre-saccadic (gray) and post-saccadic (green) pursuit velocity. B is a trace from Animal 1, for positive step target moving upward at 5°/s. C is a trace from Animal 1, for negative step target moving rightward at 5°/s. D is a trace from Animal 2, for negative step target moving leftward at 8°/s.

Figure 2.

Post-saccadic enhancement of single-target pursuit. A, Analysis from Animal 1 for positive step-ramp trials (n = 169). First panel shows average eye velocity before (gray) and after (green) the saccade for rightward target motion. Dotted lines indicate bootstrapped confidence intervals. Inset shows the target motion. Second panel shows the distribution of pre and post-saccadic eye velocity for all rightward single-target motion trials. Green indicates post-saccadic pursuit and gray indicates pre-saccadic pursuit. Third panel compares pre (gray) and post (green) eye velocities in rightward single-target motion trials binned by time of measurement since motion onset. The center of the analysis window is used as the time of measurement. B, Same as A for Animal 1, negative step trials, motion condition is leftward (n = 98). C, Same as A for Animal 2, negative step trials, motion condition is downward (n = 234). D, Mean velocities in the direction of target motion for different target directions for pre- (gray) and post-saccadic (green) pursuit for Animal 1, positive step. Error bars are 1 SEM. E, Same as D for Animal 1, negative step. F, Same as D for Animal 2, negative step.

Figure 3.

Two-target pursuit task examples. A, The scheme for two-target step-ramp pursuit task. Animals begin by fixating at a target at center of the screen. The targets then make either a positive or negative step (only negative step is shown) and start to move. After a variable interval, unselected target disappears. B–E, Example pursuit traces from animals. Motion of the target moving horizontally is shown in red and motion of target moving vertically is shown in blue. Eye movement is shown in black. Shaded regions in velocity plots indicate windows used to compute pre- (gray) and post-saccadic (green) eye velocities. Target steps and choice is indicated in schematic for each example. Weights for each example are provided. B and C are from Animal 1, C and D are from Animal 2.

Figure 4.

Post-saccadic enhancement of selected target pursuit. A, Pre- (gray) and post-saccadic (green) eye positions for trials within each condition. B, Compares pre- (gray) and post-saccadic (green) pursuit weights to saccade weights. Thick lines indicate the fitted linear regression lines. C, ROC curves for pre- (gray) and post-saccadic pursuit weights. First column in all rows is data for Animal 1, positive steps, targets are moving along 0 and 90° (n = 401). Second column in all rows is data for Animal 1, negative step, targets are moving along 0 and 90° (n = 97). Third column in all rows is data for Animal 2, negative step, targets are moving along 180 and 90° (n = 301). Fourth column in all rows is data for Animal 2, negative step, targets are moving along 0 and 270° (n = 198).

Figure 5.

Target selection evolution in time and relation to saccade. A, Weight distributions for pre-saccadic pursuit (gray), post-saccadic pursuit (green) and saccades (yellow). Animal, step, and direction conditions are indicated above the histograms. B, Choice probability in different intervals surrounding the saccade. Error bars are 95% confidence intervals on choice probability (n: first column = 401, second column = 97, third column = 198, fourth column = 301).

Figure 6.

Pre-saccadic pursuit bias does not affect post-saccadic pursuit enhancement. A, B, Distribution of post-saccadic eye velocity in the target direction. Left column represents trials where the saccade choice was target 1, right column is for trials with target choice 2. Blue and red bars indicate trials where the pre-saccadic pursuit choice was target 1 (blue) and target 2 (red). A is data from Animal 1, positive step condition (n: left distribution = 155, right distribution = 246). B is data from Animal 2, negative step, 0–270 condition (n: left distribution = 65, right distribution = 133).

Figure 7.

Scheme of target selection in the marmoset saccade and pursuit system.