CVSA and LHO

[Charles Collins]

My reading of this study is that it tested the efficacy of using CVSA in obtaining confessions (much like “lie-detector” tests are used for), not in actually detecting deception.

As a key example, in the Grand Larceny case with 20 suspects, 19 of the examinations resulted in No Stress Indicated, and only one resulted in Stress Indicated – which in turn resulted in a confession. The Bernoulli Probability of having had 20 successful evaluations out of 20 examinations was less than 1 in 1,000,000. If VSA were simply a prop to obtain confessions, it would have been virtually impossible to achieve such specific and unerring results.


The deception is determined by the answers to carefully crafted questions which are designed for that purpose. (Hence, Barney Fife might have a little problem using the system.)

[John Iacoletti]

As a key example, in the Grand Larceny case with 20 suspects, 19 of the examinations resulted in No Stress Indicated, and only one resulted in Stress Indicated – which in turn resulted in a confession. The Bernoulli Probability of having had 20 successful evaluations out of 20 examinations was less than 1 in 1,000,000. If VSA were simply a prop to obtain confessions, it would have been virtually impossible to achieve such specific and unerring results.

That doesn’t follow at all. What “successful evaluation”? They indicated stress in one case and they got a confession, that’s all. That doesn’t prove that there was any deception involved in that case, or no deception in the other 19.

[Charles Collins]

That doesn’t follow at all. What “successful evaluation”? They indicated stress in one case and they got a confession, that’s all. That doesn’t prove that there was any deception involved in that case, or no deception in the other 19.

Exerpt from “Malcontent” by Sean R. DeGrilla:

Computer Voice Stress Analyzer (CVSA) is a computer-based instrument which records the stress in an individual’s voice that is interpreted as deception. Stress associated with both jeopardy and consequences can be detected in the human voice. For this instrument, jeopardy is known as the danger or hazard of being found guilty. The consequence is seen as the unpleasant result.

[John Iacoletti]

Exerpt from “Malcontent” by Sean R. DeGrilla:

Computer Voice Stress Analyzer (CVSA) is a computer-based instrument which records the stress in an individual’s voice that is interpreted as deception.

Exactly. Which is why it’s not scientific.

[Charles Collins]

Exactly. Which is why it’s not scientific.

I have an MRI image that one of my doctors gave me. But it takes someone trained in reading and interpreting it to understand what it depicts and relate that to most people. Do you really believe that makes the process “not scientific”?

[Steve M. Galbraith]

I have an MRI image that one of my doctors gave me. But it takes someone trained in reading and interpreting it to understand what it depicts and relate that to most people. Do you really believe that makes the process “not scientific”?

If it indicates Oswald killed JFK then he'll reject it. He rejects fingerprint evidence, handwriting evidence, photographic evidence. He rejects eyewitness accounts, circumstantial evidence and anything else.

You're trying to use reason with a person who is unreasonable.

[John Iacoletti]

I have an MRI image that one of my doctors gave me. But it takes someone trained in reading and interpreting it to understand what it depicts and relate that to most people. Do you really believe that makes the process “not scientific”?

Nobody is trained to interpret whether a stress response is due to a lie being told. It’s why polygraphs and CVSA are generally not admissible in court.