VSS, May 13-18
Talk 1, 10:45 am, 22.21
Fast Smooth Pursuit Inhibition Reveals Mechanisms of Multisensory Integration
Philipp Kreyenmeier1,2 (), Ishmam Bhuiyan1, Hiu-Mei Chow1, Miriam Spering1,2,3,4; 1Department of Ophthalmology & Visual Sciences, University of British Columbia, Vancouver, Canada, 2Graduate Program in Neuroscience, University of British Columbia, Vancouver, Canada, 3Djavad Mowafaghian Center for Brain Health, University of British Columbia, Vancouver, Canada, 44Institute for Computing, Information, and Cognitive Systems; University of British Columbia, Vancouver, Canada
Humans have the remarkable ability to react quickly to sudden changes in the visual environment, presumably via a fast, subcortical pathway. However, how these rapid orienting responses are affected by stimuli from mixed modalities, such as light and sound, is less well known. Here, we study the fast inhibition of smooth pursuit eye movements, occurring reflexively in response to the sudden onset of auditory, visual, or audiovisual distractors. Unlike saccades, smooth pursuit provides a continuous and graded measure of sensorimotor control, allowing us to reveal mechanisms of multisensory integration for fast sensorimotor control. Observers (n=16) tracked a horizontally moving disc while eye movements were recorded. After a random delay (0.5-1s) an auditory, visual, or audiovisual distractor was presented for 50 ms in 60% of trials; no distractor was presented in the remaining 40% of trials (control trials). Distractor presentation elicited a transient inhibition in smooth pursuit velocity ~100 ms after distractor onset, followed by an increase in pursuit velocity above baseline (rebound). Pursuit inhibition revealed graded velocity changes to the auditory (-2.7%), visual (-5.5%), and audiovisual distractors (-9.1 %; all p<.01 compared to control trials). Similarly, the increase in pursuit velocity during the rebound period scaled with distractor type (auditory: +3.7%, visual: +6.2%, and audiovisual: +9.6%; all p<.01). The audiovisual responses were significantly stronger than the visual-only (p=.004) or auditory-only conditions (p<.001). Strikingly, the linear sum of the unisensory conditions predicted the outcome of the multisensory condition, indicating an additive effect of multisensory stimulation on oculomotor inhibition. Our findings demonstrate strong effects of multisensory stimulation on fast sensorimotor control (smooth pursuit inhibition). The short latency of these responses suggests the involvement of a fast sensorimotor system, presumably mediated via the superior colliculus—a midbrain structure involved in both oculomotor orienting and multisensory integration.
Talk 2, 11:00 am, 22.22
Modeling the impairment of smooth pursuit eye movements in macular degeneration
Age-related macular degeneration (AMD) is the most prevalent cause of central visual field loss. Since the fovea (oculomotor locus) is often impaired, individuals with AMD typically have difficulties with saccadic and smooth pursuit eye movements (Verghese, Vullings, & Shanidze, 2021). We propose that smooth pursuit eye movements are impaired in macular degeneration due to two factors: 1) the transient disappearance of the target into the scotoma and 2) noise that depends on the eccentricity of the oculomotor locus from the target. To assess this claim, we measured performance in a perceptual baseball task where observers had to determine whether a target would cross or miss a rectangular region (plate) after being extinguished (Kim, Badler, & Heinen, 2005), when instructed to either fixate a marker or smoothly track the target. We recorded eye movements of 3 AMD eyes and 6 control eyes with simulated scotomata (matching those of individual AMD participants) during the task. We found that controls with simulated scotomata could better discriminate strikes from balls compared to AMD participants, particularly in the smooth pursuit condition (d' = 0.77 controls; d' = 0.64 AMD). We also developed a model that predicted performance on the task using visible portions of the target trajectory given the scotoma and position uncertainty given the eccentricity of the eye from the target. The model showed a similar trend to participant results, with better discrimination for simulations using control eye position data (foveal oculomotor locus) than for MD data (peripheral oculomotor loci). However, the model's discrimination performance was roughly 1.37x better than actual participant performance. These findings suggest that while the disappearance of the target due to the scotoma and noise due to the eccentricity of the peripheral oculomotor locus from the target in AMD affect perceptual discrimination, these factors account only partially for the impairments.
Acknowledgements: This work was supported by NIH Institutional training grant to Smith-Kettlewell T32EY025201 (JR) and grant NIH R01EY027390 (PV)
Talk 3, 11:15 am, 22.23
Relating curvature to speed: How smooth pursuit of predictable and unpredictable 2D target motions complies with the two-thirds power law
Natural movements, including smooth pursuit, comply with the two-thirds power law, which holds that movement speed (S) depends on the radius of curvature (R) raised to the power 1-2/3 (de’Sperati & Viviani, 1997). We examined the effect of the predictability of the motion path on the compliance of smooth pursuit with the law. A target that moved at a constant speed along a V-shaped path whose turn angle (0-180 deg) was unpredictable (unmarked) or predictable (marked by a line) (VSS 2019; Experiment 1). The time of the turn was chosen randomly from three values (Experiment 2). Eye speed decreased over time, reaching a minimum ~50ms after the direction change for marked, and ~150 ms for unmarked paths. The radius of curvature was found by fitting a circumscribed circle to eye positions over a 67-192 ms interval ending 60 ms after the minimum speed. The start was chosen to equate distance traveled before and after minimum speed was reached. Minimum eye speeds (S) were well fit (p<10-41) by the power law: log S = log K + (1-β)log R. Estimates of β were lower and closer to 2/3 for more predictable paths: β=.63 for marked, .74 for unmarked paths in Experiment 1; .65 for late turns and marked paths, .82-.85 for early turns and unmarked paths in Experiment 2. Larger estimates of β are associated with shallower slopes, suggesting that unpredictability weakened the relationship between curvature and speed, while predictability increased the compliance with the two-thirds power law. Predictability has been assumed to benefit pursuit mainly by reducing latencies. Here we find that another benefit is to regulate the relationship between the geometry and kinematics in a way that makes pursuit more compatible with likely natural targets: the movements of living things.
Talk 4, 11:30 am, 22.24
Modelling the neural control of ocular accommodation
Ocular accommodation is the process of adjusting the eye’s crystalline lens so as to bring the retinal image into sharp focus. The major stimulus to accommodation is therefore retinal defocus, and in essence, the job of accommodative control is to send a signal to the ciliary muscle which will minimise the magnitude of defocus. This is complicated by the sensorimotor latencies within the system, which delay both information about defocus and the accommodation changes made in response, and by the sluggish response of the motor plant. Both of these can lead to instability during steady fixation, phase delays during tracking of moving objects, and overshoots and oscillations in the response to a step change in distance, and/or excessively large oscillations in response to changing distance. A common technique to reduce these problems in motor control is the use of a “forward model” or Smith predictor, whereby the control system uses an internal model to predict the future effect of its planned actions, allowing for sensorimotor delays. Following previous work, we conclude that most aspects of accommodation are well explained by dual integral control, with a “fast” or “phasic” integrator enabling response to rapid changes in demand, but which hands over control to a “slow” or “tonic” integrator which maintains the response to steady demand. For the first time, we combine dual integral control with a Smith predictor. In addition, we propose a novel proportional-control signal, not available to the Smith predictor, to account for the power spectrum of accommodative microfluctuations during steady fixation, which may be important in hunting for optimal focus, and for the nonlinear resonance observed for low-amplitude, high-frequency input.
Acknowledgements: Magic Leap Inc
Talk 5, 11:45 am, 22.25
Active recalibration of visual localization
We have recently shown that motor errors following the execution of a saccade recalibrate not only saccade but also visual space. In addition, we found that in the absence of saccade errors, visual localization became biased toward the fovea. It is an open question whether the simultaneous recalibration of motor and visual space arises because both rely on a shared resource or because both are separate but interacting. To dissociate between both models, we used saccade adaptation. We adapted saccades in outward direction by displacing the saccade target during saccade execution. We also tested visual localization in interspersed trials. After saccade adaptation trials, we applied "no error" trials in which we annulled each post-saccadic error. To this end, we predicted the saccade landing position online and presented the saccade target at the predicted position. If saccade and motor space rely on a shared resource, visual localization and saccades should show the same distortion after application of the "no error" trials. However, if both are separate, we expected that saccades remain adapted because the absence of error information would confirm the adapted amplitude. However, because visual processing does not receive information about errors, visual localization should become biased toward the fovea. We found evidence for the latter hypothesis. First, we replicated the finding of visual mislocalization following saccade outward adaptation. Second, in sessions in which "no error" trials followed adaptation trials saccade adaptation remained strong but localization became biased toward the fovea. Our data support a model of active recalibration in which saccade errors shape visual localization while processing for visual and motor space is separate.
Acknowledgements: Supported by European Research Council (project moreSense grant agreement n. 757184).
Talk 6, 12:00 pm, 22.26
Seeing the unconscious? Limited awareness for involuntary microsaccades
Microsaccades are miniscule motor acts widely assumed to be generated in complete absence of the observers’ awareness. Recent evidence, however, demonstrated that microsaccades can be generated voluntarily and with high precision (Poletti et al. 2020), suggesting conscious accessibility. Here, we examined if observers can recognize the microsaccades they spontaneously generate and, moreover, if microsaccade-contingent visual stimulus presentations alter the awareness of these involuntary motor acts. We displayed a vertically oriented grating with a rapid temporal phase shift (>60 Hz) that rendered the stimulus invisible during slow fixational eye movements. The grating became briefly visible when it slowed down on the retina, either because the observer spontaneously generated a microsaccade in the direction of the phase shift (active condition) or when the retinal consequence of a previous microsaccade was replayed back to the observer (replay condition). In additional control trials, no stimulus was displayed at all. At the end of each trial, observers reported first if they perceived the stimulus and second whether they believed to have generated a microsaccade. Depending on the answers to these questions, observers additionally rated their confidence that the perceived stimulus was or was not caused by a microsaccade. In both the active and replay condition, observers were highly sensitive to the stimulus in trials in which their microsaccade—or the replayed microsaccade—matched the stimulus’ phase shift’s direction and velocity. This suggests that the retinal consequences of microsaccades alone (i.e., retinal motion) account for the observed sensitivity. In contrast, observers’ sensitivity to whether they generated a microsaccade was very low, in particular, when a stimulus was displayed. Finally, confidence about the cause of the stimulus percept was comparable for active and replay condition trials. Together, these results suggest that microsaccades, if generated spontaneously, indeed escape awareness.
Acknowledgements: J.N.K. was supported by a graduate school scholarship by the Berlin School of Mind and Brain. S.O. was supported by a DFG research grant (OH 274/2-2). M.R. was supported by the Heisenberg Programme of the DFG (grants RO 3579/8-1 and RO 3579/12-1).
Talk 7, 12:15 pm, 22.27
A small foveated target is not the optimal fixation stimulus
Fixating a small dot is a universal technique for stabilizing gaze in vision and eye movement research, and for clinical imaging of normal and diseased retinae. However, during fixation microsaccades and smooth “drifts” occur that presumably benefit vision by refreshing the retina and sharpening image perception. Yet microsaccades compromise image stability and usurp task attention. Motivated by evidence of similar control of microsaccades and catch-up saccades we ask if microsaccades are mitigated while fixating a peripheral target, since peripheral targets mitigate catch-up saccades during smooth pursuit. Human observers fixated either a small dot, the center of an 8-dot circular array (6° diameter), or a 9-dot composite stimulus. Microsaccade rate was significantly lower with the peripheral array (0.48 sac/sec) than the dot (1.0 sac/sec). However, reinserting the dot into the array increased microsaccade rate (0.72 sac/sec). Drift speed also decreased with the peripheral array, with and without the central dot (average 1.19 deg/sec), relative to the small dot (1.41 deg/sec). In contrast, eye position was more variable with the array than the dot alone or the dot plus array. The results suggest that analogous to pursuit of a foveal target, fixating a stationary one engages the saccadic system and compromises retinal image stability. In contrast, fixating a peripheral stimulus improves stability, thereby affording better retinal imaging and more attention for other tasks.