GazeBase, a large-scale, multi-stimulus, longitudinal eye movement dataset

This manuscript presents GazeBase, a large-scale longitudinal dataset containing 12,334 monocular eye-movement recordings captured from 322 college-aged participants. Participants completed a battery of seven tasks in two contiguous sessions during each round of recording, including a – (1) fixation task, (2) horizontal saccade task, (3) random oblique saccade task, (4) reading task, (5/6) free viewing of cinematic video task, and (7) gaze-driven gaming task. Nine rounds of recording were conducted over a 37 month period, with participants in each subsequent round recruited exclusively from prior rounds. All data was collected using an EyeLink 1000 eye tracker at a 1,000 Hz sampling rate, with a calibration and validation protocol performed before each task to ensure data quality. Due to its large number of participants and longitudinal nature, GazeBase is well suited for exploring research hypotheses in eye movement biometrics, along with other applications applying machine learning to eye movement signal analysis. Classification labels produced by the instrument’s real-time parser are provided for a subset of GazeBase, along with pupil area.

Saccades and pre-saccadic stimulus repetition alter cortical network topology and dynamics: evidence from EEG and graph theoretical analysis

Parietal and frontal cortex are involved in saccade generation, but their output signals also modify visual signals throughout cortex. These signals produce well-documented behavioral phenomena (saccades, saccadic suppression, various perisaccadic perceptual distortions) but their underlying influence on cortical network dynamics is not known. Here, we combined high resolution electroencephalography (EEG) with frequency dependent source localization and graph theory analysis (GTA) to understand how saccades and presaccadic visual stimuli interactively alter cortical network dynamics in humans. 21 participants viewed series of 1-3 vertical or horizontal grids, followed by grid with the opposite orientation just before a horizontal saccade or continued fixation. EEG signals from the presaccadic interval (cue + 200ms, or equivalent fixation period) were used for source localization. Source localization (saccade minus fixation) identified bilateral dorsomedial frontoparietal activity across frequency bands, whereas stimulus repetition produced band-specific modulations in left prefrontal, posterior parietal, and central-superior frontal and/or parietal cortex, with significant saccade-repetition interactions in frontal and parietal regions. GTA analysis revealed a saccade-specific functional network with major hubs in inferior parietal cortex (alpha) and the frontal eye fields (beta), and major saccade-repetition interactions in left prefrontal (theta) and supramarginal gyrus (gamma). Overall, quantitative measures of whole-brain network topology and dynamics (segregation, integration, synchronization, complexity) were enhanced during the presaccadic interval, but repetition interactions reduced synchronization and complexity. These results show that presaccadic signals have widespread, coherent influence on cortical network dynamics, likely responsible for both saccade production and the perceptual phenomena associated with saccades.

Horizontal Saccadic Velocity in Patients with Exotropia before and after Unilateral Resection and Recession Surgery

Purpose. The effects of strabismus surgery on eye movement are not known in detail, as few studies have compared saccade velocities before and after strabismus surgery. In this study, horizontal saccades were recorded using an eye-tracker in patients with only exotropia to compare the peak velocities (PVs), before and after undergoing strabismus surgery of the same type (unilateral resection and recession). Methods. Horizontal saccades of monocular vision were recorded using an eye-tracking device in 18 patients with exotropia and 20 normal subjects. All patients were examined using the same method after strabismus surgery. Results. The PVs of adduction and abduction in the patients were higher than those in the normal subjects (in dominant eye, P=0.032 for adduction and P=0.049 for abduction; in nondominant eye, P=0.016 for adduction and P=0.037 for abduction). Following the surgery, the PVs of abduction of the surgical eye (nondominant eye) decreased to the level of the normal subjects (P=0.016). However, there were no correlations between changes in the PVs and the extent of surgery (resection and recession). Conclusion. Strabismus surgery normalized the patient’s increased PV in the operated eye for abduction of horizontal saccade. Not only peripheral (extraocular muscle) but also central sensory-motor mechanisms may be involved in the changes in PV of horizontal saccades, both of which could result from the improvement of the primary eye position.

Activity of near-response cells during disconjugate saccades in strabismic monkeys

Infantile strabismus is a common disorder characterized by a chronic misalignment of the eyes, impairment of binocular vision, and oculomotor abnormalities. Nonhuman primates with strabismus, induced in infancy, show a pattern of abnormalities similar to those of strabismic children. This allows strabismic nonhuman primates to serve as an ideal animal model to examine neural mechanisms associated with aberrant oculomotor behavior. Here, we test the hypothesis that impairment of disparity vergence and horizontal saccade disconjugacy in exotropia and esotropia are associated with disrupted tuning of near- and far-response neurons in the supraoculomotor area (SOA). In normal animals, these neurons carry signals related to vergence position and/or velocity. We hypothesized that, in strabismus, these neurons modulate inappropriately in association with saccades between equidistant targets. We recorded from 62 SOA neurons from 4 strabismic animals (2 esotropes and 2 exotropes) during visually guided saccades to a target that stepped to different locations on a tangent screen. Under these same conditions, SOA neurons in normal animals show no detectable modulation. In our strabismic subjects, we found that a subset of SOA neurons carry weak vergence velocity signals during saccades. In addition, a subset of SOA neurons showed clear modulation associated with slow fluctuations of horizontal strabismus angle in the absence of a saccade. We suggest that abnormal SOA activity contributes to fixation instability but plays only a minor role in the horizontal disconjugacy of saccades that do not switch fixation from one eye to the other. NEW & NOTEWORTHY The present study is the first to investigate the activity of neurons in the supraoculomotor area (SOA) during horizontally disconjugate saccades in a nonhuman primate model of infantile strabismus. We report that fluctuations of horizontal strabismus angle, during fixation of static targets on a tangent screen, are associated with contextually inappropriate modulation of SOA activity. However, firing rate modulation during saccades is too weak to make a major contribution to horizontal disconjugacy.

Eye Movements

Chapter 23 discusses eye movements, including the final common pathway, horizontal saccade circuitry and details, vertical saccades, smooth pursuit, and the pupillary light reflex.

Muscimol inactivation of caudal fastigial nucleus and posterior interposed nucleus in monkeys with strabismus

Previously, we showed that neurons in the supraoculomotor area (SOA), known to encode vergence angle in normal monkeys, encode the horizontal eye misalignment in strabismic monkeys. The SOA receives afferent projections from the caudal fastigial nucleus (cFN) and the posterior interposed nucleus (PIN) in the cerebellum. The objectives of the present study were to investigate the potential roles of the cFN and PIN in 1) conjugate eye movements and 2) binocular eye alignment in strabismic monkeys. We used unilateral injections of the GABAA agonist muscimol to reversibly inactivate the cFN (4 injections in exotropic monkey S1 with ∼4° of exotropia; 5 injections in esotropic monkey S2 with ∼34° of esotropia) and the PIN (3 injections in monkey S1). cFN inactivation induced horizontal saccade dysmetria in all experiments (mean 39% increase in ipsilesional saccade gain and 26% decrease in contralesional gain). Also, mean contralesional smooth-pursuit gain was decreased by 31%. cFN inactivation induced a divergent change in eye alignment in both monkeys, with exotropia increasing by an average of 9.8° in monkey S1 and esotropia decreasing by an average of 11.2° in monkey S2 ( P < 0.001). Unilateral PIN inactivation in monkey S1 resulted in a mean increase in the gain of upward saccades by 13% and also induced a convergent change in eye alignment, reducing exotropia by an average of 2.7° ( P < 0.001). We conclude that cFN/PIN influences on conjugate eye movements in strabismic monkeys are similar to those postulated in normal monkeys and cFN/PIN play important and complementary roles in maintaining the steady-state misalignment in strabismus.