Dynamics of oculomotor adaptation and its interaction with perception
Final Report Abstract
Saccadic eye movements are the visual system’s dexterous camera work that allows exploration of the fine details of a visual scene. Maintaining their accuracy during arousal and fatigue, throughout developmental or pathological modifications, is thus of primary importance for our ability to see and act. In this collaborative project, we studied oculomotor plasticity as it arises in response to incongruities between our movements and their visual goals, and explored its relationship with visual perception in healthy adult participants. In particular, we used a combination of eye tracking, visual psychophysics, and computational modeling to: 1. Extend the repertoire of paradigms to reveal the dynamics of adaptation. 2. Assess the impact of single saccades on the gain of future saccades. 3. Establish the effect of intermediate and long time scales on adaptation. 4. Investigate the temporal evolution of the spatial profile of adaptation. 5. Determine the dynamics of perceptual vs. motor changes and their interactions. First, we characterized, in phenomenological terms, the dynamic changes of movement amplitude in a novel saccade adaptation paradigm. We showed that saccadic eye movements adapted stereotypically and systematically to visual errors that varied continuously as a sinusoidal function of trial number. The oculomotor response had the same frequency and followed the variation with a short phase lag in the order of tens of trials. This contribution extends the repertoire of paradigms to reveal the dynamics of adaptation beyond the traditional protocols. In addition, our protocol reveals this fast learning process in single short experimental sessions, qualifying it for the study of sensorimotor learning in health and disease. Second, we introduced a generative model that appropriately described the sinusoidal adaptation data. We characterized the dependence of the generative parameters (the learning rates) on experimental conditions and stimulus features and identified the weights of an impulse response function that, when convolved with the stimulus, provided a reasonable fit of the observed response. Third, using a fast-paced adaptation protocol, we were able to make detailed measurements of the temporal evolution of the spatial profile of adaptation. When adapting a single saccade vector, the strength of the transfer of adaptation to other saccade vectors depended on the similarity between adapted and tested vectors (i.e., giving rise to a so-called adaptation field), replicating earlier findings. More interestingly, transfer developed over the time-course of the experiment in a way that was similar to adaptation itself. This result suggests that the adaptation field rises steadily over time (with little change in tuning), emerging gradually as adaptation unfolds. Fourth, in two projects, we studied the interactions between saccade adaptation and perception. In one study, we showed that the speed, but not the magnitude, of adaptation was a function of participants’ visual sensitivity to intrasaccadic changes. Adaptation took longer to emerge when the target step was close to the individual’s displacement threshold. In a second project, our French collaborators examined sensitivity to target displacements using a paradigm in which the saccade target steps during the eye movement, but to a different position on every trial. Participants had to indicate whether the target stepped left or right. By using reverse correlation, they showed that participants have an internal estimate of their own oculomotor variability, and that this knowledge contributes to trans-saccadic perception. Finally, in collaborations with other labs, we have transferred the sinusoidal adaptation protocol to a different domain of learning in the active visual system (transsaccadic size recalibration) and derived ways to use our findings in translational research on psychosis (in particular, schizophrenia). Together, the results of this project contribute to a core question in the interdisciplinary field of active vision – how sensorimotor plasticity allows humans to exhibit adaptive intelligent behavior.
Publications
- (2015). Attention in active vision: A perspective on perceptual continuity across saccades. Perception, 44, 900-919
Rolfs, M.
(See online at https://doi.org/10.1177/0301006615594965) - (2015). Failure to use corollary discharge to remap visual target locations is associated with psychotic symptom severity in schizophrenia. Journal of Neurophysiology, 114, 1129-1136
Rösler, L., Rolfs, M., van der Stigchel, S., Neggers, S.F.W., Cahn, W., Kahn, R.S., & Thakkar, K.N.
(See online at https://doi.org/10.1152/jn.00155.2015) - (2016). Saccadic adaptation to a systematically varying disturbance. Journal of Neurophysiology, 116, 336-350
Cassanello, C., Ohl, S., & Rolfs, M.
(See online at https://doi.org/10.1152/jn.00206.2016) - (2017). Oculomotor prediction: a window into the psychotic mind. Trends in Cognitive Sciences, 21, 344-356
Thakkar, K.N., Diwadkar, V.A., & Rolfs, M.
(See online at https://doi.org/10.1016/j.tics.2017.02.001) - (2018). Probing transsaccadic correspondence with reverse correlation. Journal of Vision, 18(3):10, 1-14
Joosten, E.R., & Collins, T.
(See online at https://doi.org/10.1167/18.3.10) - (2019). A generative learning model for saccade adaptation. PLoS Computational Biology, 15(8), e1006695
Cassanello, C.R., Ostendorf, F., & Rolfs, M.
(See online at https://doi.org/10.1371/journal.pcbi.1006695) - (2019). Disrupted corollary discharge in schizophrenia: evidence from the oculomotor system. Biological Psychiatry: Cognitive Neuroscience and Neuroimaging, 4, 473-481
Thakkar, K.N. & Rolfs, M.
(See online at https://doi.org/10.1016/j.bpsc.2019.03.009) - (2019). Structural thalamofrontal hypoconnectivity is related to oculomotor corollary discharge dysfunction in schizophrenia. Journal of Neuroscience, 39(11), 2102-2113
Yao, B., Neggers, S.F.W., Rolfs, M., Rösler, L., Thompson, I.A., Hopman, H.J., Ghermezi, L., Kahn, R.S., & Thakkar, K.N.
(See online at https://doi.org/10.1523/JNEUROSCI.1473-18.2019)
