Contact: Dr. Cusumano, Dr. Cesari, A'ndrea Elyse Messer
A new approach to analyzing the steadiness of a marksman's aim, developed by researchers from Penn State and the University of Verona, Italy, eventually may also be useful in diagnosing and monitoring tremors in patients with neuromuscular disorders.
Dr. Joseph P. Cusumano, Penn State associate professor of engineering science and mechanics, says, "The analyses that we are developing, with cooperation from the members of an Italian air pistol team, may enable us to develop a sort of steadiness profile or body tremor 'fingerprint' for any individual. Such a fingerprint can be useful in a clinical setting to diagnose and track the progression of a neuro-muscular disorder or injury, or the recovery from such a disorder or injury."
He adds, "The key thing - and what is new about our approach - is that the profile we generate is based on an analysis of an individual's movement while they are accomplishing a specific task, namely, aiming. Most previous work, for example with Parkinson's patients, has looked at tremor independent of any task - without, as it were, any context."
The new procedure allows the researchers to evaluate an individual athlete and understand the specific configurations in which they hold their limbs to maximize the way in which the natural tremor in their body is controlled when they aim at a target.
Cusumano and his co-investigator, Dr. Paola Cesari, director of the Movement Science Laboratory, University of Verona, Italy, described their technique today June 27 at the Eleventh International Conference on Perception and Action at the University of Connecticut in a paper, The Analysis of Kinematic Redundancy in Pointing.
Cusumano explains that everyone, even the steadiest marksman, experiences unavoidable inherent body tremor. People can minimize or amplify the way this natural movement affects their pointing accuracy by positioning their limbs in a variety of ways - some positions being more or less effective than others.
The new analysis technique, developed by Cusumano and Cesari, allows the researchers to use video cameras to collect movement data from a person shooting at a target, reduce the person's movements to their basic elements and then explain how and why tremor varies as the person adjusts his or her joint angles to try to maintain a steady aim.
"It is clear that we can track and, to some extent, explain the sources of tremor on the target in terms of different body configurations, but the best way to take the results of our analysis and actually predict performance, given a set of measurements, is not yet clear to us," says Cusumano. "However, the results of this research will nevertheless provide a powerful tool to coaches, since the results of changes in the athlete's form can be directly measured, interpreted and related to performance in a physically meaningful way."
In their approach, Cusumano and Cesari have combined a method of statistical data analysis, called Principal Component Analysis (PCA), with stability analysis, which relates tremor in the body joint positions to tremor of the target point.
Cusumano says, "This combination of analyses is familiar in robotics but is new in the area of movement science."
The researchers asked 16 people with varying degrees of expertise to aim an air pistol and shoot at a target. Each subject had a reflective marker attached to the tip of his gun, and on his wrist, elbow, shoulder, neck, head and hip.
Two opto-electrical cameras equipped with infrared illuminators tracked the markers and collected movement data from the subjects. Using a computer, the researchers analyzed each image from the cameras and extracted the position of each marker as a function of time to produce a data stream.
The researchers found that all shooters, expert and non-expert alike, had one dominant pattern of movement which corresponded, at the target, to relatively large movements left to right and relatively small movement vertically.
"Since even nonexperts, such as me, did pretty much the same thing, this suggests that these basic movements are related to our bio-mechanical nature, not to skill per se," says Cusumano.
However, higher scoring marksmen tended to hold their limbs in one of only a few consistent, coordinated patterns. These favored positions were such that the natural tremor in the body was strongly decreased in its effect on motion at the target.
"While the number of coordinated movement patterns needed to accomplish a given task indicates an individual's level of coordination, increased stability or control at the target is related to a subject's ability to couple his or her motor coordination with the perception of the target," he adds.
The research is continuing and Cusumano and Cesari hope eventually to be able to predict an individual's performance based on their analysis. Looking to the future, they hope to use the same basic approach to define criteria that would, for example, help a physician to measure the decline over time in a Parkinson's patient's ability to perform a specific task.