Body-Goal Variability Analysis of Performance

Paola Cesari
Dipartimento di Scienze Motorie; Universita' di Verona

Joseph Cusumano
Engineering Science & Mechanics, Penn State University

     Full text: Not available
     Last modified: March 15, 2007

Abstract
The large numbers of joints and muscles of the human human body give rise to
equifinality [1] in task execution. That is, there are typically an infinite
number of body configurations that can achieve the same goal. This set of
configurations is now known to in many cases of practical importance to have
the structure of a manifold [2]. In this work, we use this fact to develop a
new type of experimental task performance analysis based on the idea of a
body-goal variability mapping [3]. We show how this mapping arises naturally
from the idea of a goal function that theoretically defines a task and directly
determines the goal equivalent manifold (GEM), the set of all possible task
solution strategies. In addition, the body-goal map determines the sensitivity
of different strategies to perturbations in the body state during execution,
and thus can be used to show that even though two different strategies may be
geometrically goal-equivalent (i.e., lie within the GEM), they may not be
equivalent in their robustness to perturbation.

The experimental application of these ideas does not require a detailed
kinematic model, but instead is uses data to estimate the body-goal matrix that
relates body and goal-level variability. This results in a characterization of
the sensitivity of goal-level errors to body-level perturbations. We show how
the goal-level performance can in general be decomposed into factors involving
sensitivity, GEM alignment, and body variability. The method is applied to the
analysis of redundant kinematic data from subjects performing an aiming task
carried out with and without a laser pointer. The results show that the
decomposition obtained from the body-goal map allows one to see how a specific
performance level is achieved by a given subject, and to distinguish the
behavior of subjects even when their goal-level performance is
indistinguishable. The analysis also allows one to see precisely how subjects
respond to changes in perceptual information (laser on/ laser off) and
precision requirements (target size). We find evidence that control is exerted
on all factors in the body-goal performance decomposition, not just that
related to GEM alignment.

[1] Bernstein N (1967) The coordination and regulation of movements. Oxford,
Pergamon.

[2] Scholz J, Sch?ner G (1999) The uncontrolled manifold concept. Experimental
Brain Research, 126, 289?306

[3] JP Cusumano and Paola Cesari (2006) Body-Goal Variability Mapping in an
Aiming Task. Biological Cybernetics, 94(5), 367-379.