Due to the foveated nature of our visual system, eye movements are essential for perception. They are not only the most frequent movements of our bodies, but they have also been studied quite frequently. Because they can serve as objective, implicit and continuous measurements of sensory and decision processes, they are often considered to be a ‘window into the mind’. Past research has investigated the two most important types of eye movements, fast saccades and slow smooth pursuit, separately. However, in the real world these two movements always occur together.

Building on previous work, we have shown across multiple studies how tightly saccadic and pursuit eye movements interact and share information for an optimal eye movement control.

Coming Soon

By leveraging individual differences across a battery of eye movement tasks, we could show that oculomotor behavior across different tasks is linked by the relevant sensory information. Within a task, individual observers rely on their own strengths to decide how to combine saccadic and pursuit eye movements.

Goettker, A. & Gegenfurtner, K.R. (2023). Individual differences link sensory processing and motor control. PsychArchives. https://doi.org/10.23668/psycharchives.12508

Review on saccade-pursuit interactions

Building on previous work, we summarized recent evidence from our work and the work of others to show how saccadic and pursuit eye movements interact and how we are able to perceive a stable world during our constant eye movements

Saccadic and pursuit eye movements are part of a single sensorimotor system, which is based on separate, continuous streams of position and velocity-related information. Within this framework, compensation for eye movements is also based on efferent information about upcoming eye positions and eye velocities, allowing to discount for retinal effects of any combination of saccadic and pursuit eye movements. Based on this proposed model of shared representation of eye position and velocity, dynamic interactions and synergies between the two movements allow for optimal tracking performance, while at the same time maintaining perceptual stability. 

Goettker, A., & Gegenfurtner, K. R. (2021). A change in perspective: The interaction of saccadic and pursuit eye movements in oculomotor control and perception. Vision Research, 188, 283-296.

Saccadic and pursuit eye movements are based on a combination of position- and velocity-related signals

We used isoluminant stimuli in the periphery to attenuate the velocity information of a moving stimulus, which led to a very interesting error pattern: Saccades towards these stimuli landed at the position the target had 100 ms before saccade onset! In addition, the scaling to target velocity of the following pursuit response, that happened within 50 ms for “normal” stimuli, was absent and the pursuit response impaired. This suggests that when missing information about the velocity of a target, the eyes are targeted at an accurate, but delayed representation of the position of the target.

Goettker, A., Braun, D. I., & Gegenfurtner, K. R. (2019). Dynamic combination of position and motion information when tracking moving targets. Journal of Vision, 19(7), 2-2.

How you move your eyes affects how you perceive the world…

While a lot of research has investigated how we perceive a stable world either during a saccadic or during a pursuit eye movement, how these two processes interact was unclear. To test this, we leveraged the probabilistic nature of the oculomotor system: When you present certain target movements multiple times, the eyes sometimes use pure pursuit and sometime pursuit with additional corrective saccades to track the moving target (Figure A). We then compared the perceived speed of these targets depending on how you moved your eyes and were surprised: The perceived speed of trials with additional corrective saccades differed and it differed even depending on the direction of the saccade (Figure B). If you used a saccade to jump ahead to catch-up with the target, you perceive the stimulus to move faster. If you used a saccade to jump back to wait for the target, you perceive the stimulus to move slower.

We were able to predict these results with one critical assumption: The efferent signal you use to compensate for your eye movement is just an eye velocity signal, independent of whether you executed a saccadic or pursuit eye movement! During pursuit trials variations in eye velocity did not affect the perceived speed (Figure C), but variations in eye speed based on the integrated velocity of saccades did affect it (Figure D).

Goettker, A., Braun, D. I., Schütz, A. C., & Gegenfurtner, K. R. (2018). Execution of saccadic eye movements affects speed perception. Proceedings of the National Academy of Sciences, 115(9), 2240-2245.

… and how you intercept moving targets

In a follow-up study we investigated whether the changes in perceived speed were also reflected in interception of these targets. We replicated the influence of corrective saccades on perceived speed and found that it also influenced the interception endpoints. Targets that were perceived faster due to a forward catch-up saccade were intercepted further ahead than comparable trials tracked with pure pursuit. The opposite pattern was present for backward saccades. However, although qualitatively comparable the effects between perception and interception were not correlated, which seemed to be based on a different temporal integration of information: While saccades at any point in the trial influenced the perceived speed, only saccades shortly before the interception also biased interception endpoints.

Goettker, A., Brenner, E., Gegenfurtner, K. R., & de la Malla, C. (2019). Corrective saccades influence velocity judgments and interception. Scientific Reports, 9(1), 5395.

Tracking curved movements

Most studies for pursuit eye movements use linear target movements with a constant velocity. Here we asked, how good people could track curved target movements and how their tracking performance compared to perceptual judgements about the amount of curvature. We observed that while observers could discriminate trajectories with different curvatures already at short presentation durations, pursuit eye movements took around 300 ms to show an accurate curvature. Again, saccadic eye movements came to the rescue, and corrected for the initially inaccurate pursuit.

Ross, N. M., Goettker, A., Schütz, A. C., Braun, D. I., & Gegenfurtner, K. R. (2017). Discrimination of curvature from motion during smooth pursuit eye movements and fixation. Journal of Neurophysiology, 118(3), 1762-1774.