Home Research Project Details D3 - Dynamics and variability of planar human tracing movements
Personal tools

D3 - Dynamics and variability of planar human tracing movements

Armin Biess and Hans Scherberger

This research project investigates the dynamical control principles that underlie human tracing movements of planar contours. Tracing movements are particularly suitable for the quantitative analysis of the motor control system because they define a class of human arm movements that can be measured with high accuracy and have a well-defined task-goal description, that is, to move as close to a contour while keeping a certain, a-priori defined timing. Our research objective is to quantify and predict the movement dynamics and variability for tracing of closed and open contours by a common model description. How do fluctuations in hand position around the given contour depend on movement parameters, such as speed, acceleration and contour shape? How does the motor system adapt the movement dynamics to changes in contour shape, for example, when changing a circular path smoothly into a peanuts-shaped contour? Is there a parsimonious representation of motor information and if so how can it be quantified? We intend to approach these questions using optimization principles and geometrical methods that have been successfully applied in our previous studies [Biess et al 2011, 2007, 2006]. Existing models of human tracing movements [Lacquaniti et al. 1983, Todorov et al. 1998] do not address these questions. In this project we aim towards a dynamical model for periodic and discrete planar tracing movements by pursuing the following goals:

Aim 1. Quantitative phenomenology of movement dynamics and its fluctuations: To gain insight into the movement dynamics we will measure kinematic movement parameters (e.g., speed, acceleration and direction) during tracing of different contours using a digitizing tablet (WACOM) while controlling the geometric properties of the shape and timing of the movement. We intend to study the functional dependence of the contour-error on the movement parameters. This analysis will identify the kinematic variables that are most likely to be controlled by the motor system to achieve the task-goal and will provide insight into the characteristics of movement dynamics and signal-dependent noise [Harris et al. 1998]. From these results we will try to infer the movement dynamics in form of stochastic differential equations and test their predictive power by numerical integration and analytical approximate solutions of the corresponding Fokker-Planck equation.

Aim 2. Global optimization principles in tracing movements: Based on these data we will further investigate whether the movement dynamics can be derived from global optimization principles [Richardson et al. 2002]. To predict the mean motor behaviour we test whether one can accurately describe tracing movements that can predict the movement dynamics for open and closed contours by an optimization principle.

Aim 3. Movement adaptation to changes in contour shape: Using the accumulated data set we will explore how the motor system modifies the movement dynamics due to changes in contour shape. Starting from a known motor pattern, such as the periodic tracing of a circle, we will generate new motor patterns with closed contours by applying different classes of maps. First results that we have obtained demonstrate that this approach is promising in generating novel motor patterns from existing ones.

Belongs to Group(s):
Primate Neurobiology

Is part of  Section D 

Members working within this Project:
Biess, Armin 
Scherberger, Hansjörg 

Selected Publication(s):