Related Experiment Video
Updated: Jun 18, 2026
07:05
Visualization Method for Proprioceptive Drift on a 2D Plane Using Support Vector Machine
Published on: October 27, 2016
Quantizing and characterizing the variance of hand postures in a novel transformation task
Ramana Vinjamuri1, Mingui Sun, Douglas Weber
1Department of Physical Medicine and Rehabilitation, University of Pittsburgh, USA. rkv3@pitt.edu
Summary
This study shows that learning reduces hand posture variance in novel tasks. Principal component analysis (PCA) revealed that subjects developed task-specific postural synergies as they improved cursor control.
Area of Science:
- Human-Computer Interaction
- Biomechanics
- Motor Learning
Background:
- Understanding how humans adapt to novel control schemes is crucial for designing intuitive interfaces.
- Hand posture variability is a key factor in motor control and learning.
Purpose of the Study:
- To investigate the changes in hand posture variance during a novel cursor control task.
- To identify if learning leads to the adaptation of postural synergies.
Main Methods:
- Utilized a data glove with 14 sensors to measure joint angles.
- Employed principal component analysis (PCA) to analyze hand posture data from five subjects across one- and two-dimensional cursor control tasks.
- Subjects performed multiple trials to achieve smooth cursor trajectories.
Main Results:
- A decrease in the number of principal components was observed across trials in both tasks, indicating reduced posture variance with learning.
- Visualization of postural synergies revealed the emergence of task-specific patterns.
Conclusions:
- Learning in novel motor tasks leads to a reduction in hand posture variability.
- Subjects adapt task-specific postural synergies, suggesting efficient motor control strategies develop with practice.
