saccades
To efficiently explore our visual environment, we humans incessantly make brief and rapid eye movements. These so-called saccades inevitably shift the entire visual image across the retina, thereby inducing - like a moving camera with long exposure duration - a significant amount of motion blur, transforming single objects into elongated smeared motion streaks. While simultaneously recording electroencephalography and eye tracking, we asked human observers to make saccades to a target stimulus which then rapidly changed location while their eyes were in mid-flight. Critically, we compared smooth target motion to a simple jump, thus isolating neural responses and behavioral benefits specific to motion streaks. For continuous motion (i.e., when streaks were available), the post-saccadic target location could be decoded earlier from electrophysiological data and secondary saccades went more quickly to the new target location. Indeed, decoding of target location succeeded immediately after the end of the saccade and was most efficient on occipital sensors, suggesting that saccade-induced motion streaks are represented in visual cortex. Computational modeling of saccades as a consequence of early visual processes suggests that fast motion could be efficiently coded in orientation-selective channels, providing a parsimonious mechanism by which the brain exploits motion streaks for goal-directed behavior.
When looking at data recorded by video-based eye tracking systems, one might have noticed brief periods of instability around saccade offset. These so-called post-saccadic oscillations are caused by inertial forces that act on the elastic components of the eye, such as the iris or the lens, and can greatly distort estimates of saccade duration and peak velocity. In this paper, we describe and evaluate biophysically plausible models (for a demonstration, see the shiny app) that can not only approximate saccade trajectories observed in video-based eye tracking, but also extract the underlying – and otherwise unobservable – rotation of the eyeball. We further present detection algorithms for post-saccadic oscillations, which are made publicly available, and finally demonstrate how accurate models of saccade trajectory can be used to generate data and mathematically tractable ground-truth labels for training ML-based algorithms that are capable of accurately detecting post-saccadic oscillations.
