Polygraph vs. Voice Stress: What’s the difference?
According to David Krapohl, President of the American Polygraph Association:
Various investigative techniques for detecting deception have appeared in the past 80 years. Some were developed by scientists and researchers, like reaction time tests, the polygraph and brain wave methods. Others were created by manufacturers without the help of researchers, such as the B&W lie detector and the various voice stress devices. The most recent method being heralded as the new lie detector is the Computerised Voice Stress Analyser (CVSA). What separates the CVSA from previous voice stress methods is that it is displayed on a computer screen, rather than on paper. There are no validated algorithms, scoring systems, or sophisticated analytical methods. These shortcomings have not prevented the manufacturer from making remarkable claims regarding the efficiency of its product.
Are the claims correct?
Those of us in the lie detection profession would like to believe it, because switching to this new device would allow us to better serve our clients in a shorter time. Before you accept the self-endorsements that the operators of these voice stress devices make, it is best that we first look at what scientists have to say.
Below is a list of the university grade research studies that have investigated voice stress as a deception detection approach. Some studies looked at the CVSA device in particular, while others investigated whether voice stress analysis in general, could be used to detect stress or deception. Copies of these studies can be obtained from many university libraries or can be accessed through the American Polygraph Association website at www.polygraph.org
Conclusion: Accuracy was not significantly greater than chance for the CVSA.
Conclusion: Validity of voice stress measures was poor.
Conclusion: Chance-level detection of stress. Chance-level detection of lies.
Conclusion: Voice stress did not detect deception greater than chance.
Conclusion: Chance-level detection of deception for voice analysis.
Conclusion: Voice measures were not reliable or useful.
|
 |
