To control our movements, the brain relies on sensory feedback. This feedback is thought to take two forms: one from sensory receptors that measure the state of our body and our environment, and a second from a process that predicts that state using efference copy, called a forward model (FM). While the idea of FM is attractive from a theoretical perspective, there is as yet no experiment that convincingly demonstrates that control of movements depends on a FM. This is because most previous experiments have focused on reaching, a movement in which it is difficult to dissociate effect of FM vs. actual sensory feedback. Yet, testing the hypothesis of FMs is crucial because while many have argued that the cerebellum plays a role in computing FMs, there is scant evidence that control of movements depends on FMs. To solve this problem, I suggest that a better approach may be to consider control of saccades. These movements typically complete in 60ms and are too fast for external sensory feedback to be useful. For this reason, saccades are often thought to be stereotypical, ballistic movements. However, I will show that the motor commands that initiate saccades are in fact highly variable. If left uncompensated, this variability would causes dysmetria in our saccades. Yet, saccade endpoints appear immune to this variability, consistent with the idea that control of saccades depends on an internal feedback that monitors the outgoing motor commands and """"""""steers"""""""" the eyes. If this internal monitoring is a FM, then the FM should adapt in response to persistent errors. Therefore, endpoint errors should produce a change in the way that motor commands are internally monitored, producing a change in the control system, and the resulting saccade trajectory. Are saccades controlled via a FM? First, I will characterize sources of natural variability in motor commands that initiate saccades to a given visual target. Second, I will perform adaptation experiments to alter the hypothetical FM and test whether there is a change in the trajectory of the saccades, as predicted by adaptive control theory. Third, I will test whether this variability is corrected via internal feedback through the cerebellum.
One objective of this project is to understand the mechanisms by which eye movements are calibrated for optimal visual function. Without exquisitely accurate eye movements, vision is degraded and patients suffer disability during tasks such as reading, driving, searching ones visual environment, and using vision to make accurate movements of the limbs during reaching or grasping. How this is accomplished is a major focus of contemporary visual neuroscience research with strong implications for patient care.
| Xu-Wilson, Minnan; Tian, Jing; Shadmehr, Reza et al. (2011) TMS perturbs saccade trajectories and unmasks an internal feedback controller for saccades. J Neurosci 31:11537-46 |
| Xu-Wilson, Minnan; Zee, David S; Shadmehr, Reza (2009) The intrinsic value of visual information affects saccade velocities. Exp Brain Res 196:475-81 |
| Xu-Wilson, Minnan; Chen-Harris, Haiyin; Zee, David S et al. (2009) Cerebellar contributions to adaptive control of saccades in humans. J Neurosci 29:12930-9 |
