Individual differences in motor noise and adaptation rate are optimally related

Abstract

Individual variations in motor adaptation rate were recently shown to correlate with movement variability or “motor noise” in a forcefield adaptation task. However, this finding could not be replicated in a meta-analysis of visuomotor adaptation experiments. Possibly, this inconsistency stems from noise being composed of distinct components which relate to adaptation rate in different ways. Indeed, previous modeling and electrophysiological studies have suggested that motor noise can be factored into planning noise, originating from the brain, and execution noise, stemming from the periphery. Were the motor system optimally tuned to these noise sources, planning noise would correlate positively with adaptation rate and execution noise would correlate negatively with adaptation rate, a phenomenon familiar in Kalman filters. To test this prediction, we performed a visuomotor adaptation experiment in 69 subjects. Using a novel Bayesian fitting procedure, we succeeded in applying the well-established state-space model of adaptation to individual data. We found that adaptation rate correlates positively with planning noise (r=0.27; 95%HDI=[0.05 0.50]) and negatively with execution noise (r=−0.41; 95%HDI=[−0.63 −0.16]). In addition, the steady-state Kalman gain calculated from state and execution noise correlated positively with adaptation rate (r = 0.31; 95%HDI = [0.09 0.54]). These results suggest that motor adaptation is tuned to approximate optimal learning, consistent with the “optimal control” framework that has been used to explain motor control. Since motor adaptation is thought to be a largely cerebellar process, the results further suggest the sensitivity of the cerebellum to both planning noise and execution noise.SIGNIFICANCE STATEMENTOur study shows that the adaptation rate is optimally tuned to planning noise and execution noise across individuals. This suggests that motor adaptation is tuned to approximate optimal learning, consistent with “optimal control” approaches to understanding the motor system. In addition, our results imply sensitivity of the cerebellum to both planning noise and execution noise, an idea not previously considered. Finally, our Bayesian statistical approach represents a powerful, novel method for fitting the well-established state-space models that could have an influence on the methodology of the field.

Can we improve slow learning in cerebellar patients?

“…This classical model was originally presented by Smith and colleagues (Smith et al, 2006) and posits a fast state , that learns and forgets quickly, and a slow state , that learns and forgets slowly, both adapting to the motor error. The model was fit using a Bayesian approach similar to that in van der Vliet et al (2018). Details of the model and the Bayesian fitting procedure used to estimate individual subject parameters are available in the supplementary materials.…”

An implicit memory of errors limits human sensorimotor adaptation

“…1 and 2 illustrates a frequently observed characteristic of motor 190 adaptation: even at the late stages of training, performance continues to suffer from small, persistent 191 errors 3,4,8,15,31,[35][36][37] . Such persistent, steady-state errors are one of the fundamental predictions of 192 mathematical models of adaptation [11][12][13][14][15] in which performance is driven by an interaction between two 193 opposing forces, error-based learning, and trial-to-trial forgetting: 194…”

Statistical determinants of visuomotor adaption along different dimensions during naturalistic 3D reaches

“…Previous studies (Burge, Ernst et al 2008, van Beers 2009, Wei and Kording 2010, He, Liang et al 2016, van der Vliet, Frens et al 2018), provided evidence for the OI theory during visuomotor adaptation. They showed that the planning component of motor noise correlates positively with the adaptation rate while the measurement component correlates negatively, at least when the respective noise component experimentally is kept constant.…”

“…Existing tools for fitting adaptation models (Albert and Shadmehr 2018, van der Vliet, Frens et al 2018) can only estimate parameters from single experimental runs (changes over repeated trials in one task condition and one subject) despite the often hierarchical nature of the experiment (repeated trials interleaved with multiple conditions, multiple subjects). This is particularly problematic for two-state models due to their low identifiability (see results).…”

“…Higher variability in the forward model (high planning variability) would increase the Kalman gain and allow faster corrections of the experienced error, whereas a higher variability in the sensory feedback (high measurement variability) would determine a lower gain and therefore a slower adaptation. Previous studies found that human behavior approaches that of an optimal learner (Burge, Ernst et al 2008, Wei and Kording 2010, He, Liang et al 2016, van der Vliet, Frens et al 2018) showing that adaptation rate positively correlates with the Kalman gain calculated from the planning and measurement variabilities. Recently, it was found that individual differences in the rate of motor adaptation are largely explained by individual differences in the Kalman gain (van der Vliet, Frens et al 2018).…”