Generalization in Adaptation to Stable and Unstable Dynamics

Abstract

Humans skillfully manipulate objects and tools despite the inherent instability. In order to succeed at these tasks, the sensorimotor control system must build an internal representation of both the force and mechanical impedance. As it is not practical to either learn or store motor commands for every possible future action, the sensorimotor control system generalizes a control strategy for a range of movements based on learning performed over a set of movements. Here, we introduce a computational model for this learning and generalization, which specifies how to learn feedforward muscle activity in a function of the state space. Specifically, by incorporating co-activation as a function of error into the feedback command, we are able to derive an algorithm from a gradient descent minimization of motion error and effort, subject to maintaining a stability margin. This algorithm can be used to learn to coordinate any of a variety of motor primitives such as force fields, muscle synergies, physical models or artificial neural networks. This model for human learning and generalization is able to adapt to both stable and unstable dynamics, and provides a controller for generating efficient adaptive motor behavior in robots. Simulation results exhibit predictions consistent with all experiments on learning of novel dynamics requiring adaptation of force and impedance, and enable us to re-examine some of the previous interpretations of experiments on generalization.

Motor learning of novel dynamics is not represented in a single global coordinate system: evaluation of mixed coordinate representations and local learning

“…Our findings add to this body of work, and further argue that adaptation is local. Consistent with the behavioral evidence, previous computational studies that examine how movement errors generalize during learning have found narrow bases of representation ( Donchin et al 2003 ; Ingram et al 2010 ; Kadiallah et al 2012 ; Thoroughman and Shadmehr 2000 ; Thoroughman and Taylor 2005 ). Finally, there are also neural data that suggest localized tuning curves ( Cohen and Andersen 2002 ; Coltz et al 1999 ; Georgopoulos et al 2007 ; Paz et al 2003 ).…”

The effects of training breadth on motor generalization

“…The implication is that neural populations encode local features of learning and should have difficulty extrapolating to completely novel circumstances. Similarly, computational models that examine how movement errors generalize during learning have found narrow bases of representation Ingram et al 2010;Kadiallah et al 2012;Thoroughman and Shadmehr 2000;Thoroughman and Taylor 2005). Our findings add to this body of work and further argue that adaptation is a form of local learning.…”

Proprioception in musculoskeletal rehabilitation. Part 1: Basic science and principles of assessment and clinical interventions

“…As training progresses, muscle activity declines due to adaptation of feedback and feedforward control (Franklin, Liaw et al 2007). It has been suggested that this learning effect occurs partly by the CNS building an internal forward model used in feedforward control to minimize motion error and effort while maintaining stability (Kadiallah, Franklin et al 2012), and that muscle spindle afferents contribute in predicting future kinematic states by acting as forward internal sensory models in learned skills (Dimitriou and Edin 2010). Feedback control, including long latency feedback responses, also adapt due to context and task demands as learning occurs (Pruszynski andScott 2012, Cluff andScott 2013).…”