Visual Feedback is not necessary for the Learning of Novel Dynamics
File(s)
Published version
Author(s)
Franklin, DW
So, U
Burdet, E
Kawato, M
Type
Journal Article
Abstract
Background. When learning to perform a novel sensorimotor task, humans integrate multi-modal sensory feedback such as
vision and proprioception in order to make the appropriate adjustments to successfully complete the task. Sensory feedback is
used both during movement to control and correct the current movement, and to update the feed-forward motor command
for subsequent movements. Previous work has shown that adaptation to stable dynamics is possible without visual feedback.
However, it is not clear to what degree visual information during movement contributes to this learning or whether it is
essential to the development of an internal model or impedance controller. Methodology/Principle Findings. We examined
the effects of the removal of visual feedback during movement on the learning of both stable and unstable dynamics in
comparison with the case when both vision and proprioception are available. Subjects were able to learn to make smooth
movements in both types of novel dynamics after learning with or without visual feedback. By examining the endpoint
stiffness and force after learning it could be shown that subjects adapted to both types of dynamics in the same way whether
they were provided with visual feedback of their trajectory or not. The main effects of visual feedback were to increase the
success rate of movements, slightly straighten the path, and significantly reduce variability near the end of the movement.
Conclusions/Significance. These findings suggest that visual feedback of the hand during movement is not necessary for the
adaptation to either stable or unstable novel dynamics. Instead vision appears to be used to fine-tune corrections of hand
trajectory at the end of reaching movements.
vision and proprioception in order to make the appropriate adjustments to successfully complete the task. Sensory feedback is
used both during movement to control and correct the current movement, and to update the feed-forward motor command
for subsequent movements. Previous work has shown that adaptation to stable dynamics is possible without visual feedback.
However, it is not clear to what degree visual information during movement contributes to this learning or whether it is
essential to the development of an internal model or impedance controller. Methodology/Principle Findings. We examined
the effects of the removal of visual feedback during movement on the learning of both stable and unstable dynamics in
comparison with the case when both vision and proprioception are available. Subjects were able to learn to make smooth
movements in both types of novel dynamics after learning with or without visual feedback. By examining the endpoint
stiffness and force after learning it could be shown that subjects adapted to both types of dynamics in the same way whether
they were provided with visual feedback of their trajectory or not. The main effects of visual feedback were to increase the
success rate of movements, slightly straighten the path, and significantly reduce variability near the end of the movement.
Conclusions/Significance. These findings suggest that visual feedback of the hand during movement is not necessary for the
adaptation to either stable or unstable novel dynamics. Instead vision appears to be used to fine-tune corrections of hand
trajectory at the end of reaching movements.
Date Issued
2007-12-19
Date Acceptance
2007-11-24
Citation
PLOS One, 2007, 2 (12)
ISSN
1932-6203
Publisher
Public Library of Science
Journal / Book Title
PLOS One
Volume
2
Issue
12
Copyright Statement
© 2007 Franklin et al. This is an open-access article distributed under
the terms of the Creative Commons Attribution License, which permits
unrestricted use, distribution, and reproduction in any medium, provided the
original author and source are credited.
the terms of the Creative Commons Attribution License, which permits
unrestricted use, distribution, and reproduction in any medium, provided the
original author and source are credited.
License URL
Subjects
General Science & Technology
MD Multidisciplinary
Publication Status
Published
Article Number
e1336