Generalization of force-field adaptation in proprioceptively-deafferented subjects

Abstract

Humans have the remarkable ability to adapt their motor behaviour to changes in body properties and/or environmental conditions, based on sensory feedback such as vision and proprioception. The role of proprioception has been highlighted for the adaptation to new upper-limb dynamics, which is known to generalize to the opposite, non-adapted limb in healthy individuals. Such interlimb transfer seems to depend on sensory feedback, and the present study assessed whether the chronic loss of proprioception precludes interlimb transfer of dynamic adaptation by testing two well-characterized proprioceptively-deafferented subjects. These had to reach toward visual targets with vision of the limb. For both deafferented subjects, we observed adaptation of the dominant arm to Coriolis forces and after-effects on non-dominant arm movements in different movement directions, thus indicating interlimb transfer. Overall, our findings show that motor learning can generalize across limbs and movement directions despite the loss of proprioceptive afferents.

Proprioceptive loss and the perception, control and learning of arm movements in humans: evidence from sensory neuronopathy

“…They also showed that training reshaped the profile of this variability across different tasks, implying its active regulation (see also Pekny et al 2015). Wong and Shelhamer (2014) reported similar relationships between baseline errors and adaptation of saccadic eye movements, while Lefumat et al (2015) showed that variability of arm movement during the adaptation phase was correlated with the amount of inter-manual transfer of force-field adaptation, a finding that also held up for two participants with PL (Lefumat et al 2016). …”

“…IW (male, 61 years old at time of participation, 100% left-handed according to the 10-item version of the Edinburgh inventory, Oldfield 1971) has no tactile, muscle or joint proprioceptive sensation from below the collar (C3) following a selective peripheral neuronopathy brought about by a viral infection at age 19 years. GL (female, 66 years old at time of participation, right-handed with a lateral quotient of 77%; Lefumat et al 2016) suffered a similar neuronopathy except the loss of sensation is from the mouth down (trigeminal V3, Forget and Lamarre 1987; Cole and Paillard 1995). IW has apparently normal afferent sensations from neck muscles whilst GL has no afferent information from neck muscles or lower face.…”

“…However, adaptation to new forces acting on the limb, while possible with visual feedback (Sarlegna et al 2010; Yousif et al 2015; Lefumat et al 2016), would seem a more challenging task for an individual with PL (Sainburg et al 1995; Pipereit et al 2006; Bock and Thomas 2011). In the absence of limb proprioception, a perturbation such as an unexpected force is most likely perceived via visual feedback, as the distorted trajectory and/or the mis-reached goal would provide a visual error signal.…”

“…However, velocity-dependent forces are only present during motion. It is, therefore, remarkable that PL participants can not only adapt to these forces (Sarlegna et al 2010; Lefumat et al 2016), but also seem to do so selectively (Yousif et al 2015), such that their responses to either position-dependent or velocity-dependent forces are distinct and appropriate to the perturbation profile.…”

“…What features of the visual feedback used in dynamic adaptation is not yet known, but the adaptation processes could be driven by comparison between the planned trajectory and the observed movement: indeed, as the forces are often introduced abruptly, they lead in the first few trials to substantial trajectory errors which can be easily observed by the participant (Brashers-Krug et al 1996; Sarlegna et al 2010; Lefumat et al 2016). Lago-Rodriguez and Miall (2016) recently showed that normal subjects actively de-adapt learned responses based on “errorless” visual feedback, supporting the idea that de-adaptation, and most likely adaptation, is driven by the difference between an internally predicted trajectory and visual feedback.…”