Another type of deception detection device is the Psychological Stress Evaluator (PSE).  People who advocate this method claim that the voice itself reveals deception.  The PSE measures variations in emotional stress, although there is little evidence that it is highly accurate.  The idea is that when subjects lie, their voice reaches a higher pitch.  The advantage of this machine is that it avoids physical contact with the subject, but instead uses a microphone or tape recorder into which the subject speaks.  The machine then analyzes the sound of the voice and prints the results onto a graph, detecting differences in the voice that may not be heard by the human ear.

Has it been used successfully in an investigation?  Yes.

In 1973, in Modesto, California, the body of Gloria Carpenter, 59, was found in her bathtub.  A thin mark around her neck indicated that she had been strangled by a nylon stocking.  Further examination by the coroner showed that she had also been raped and had been placed in the tub to make it look as if she had died there by natural causes.  Nothing was taken from her apartment and there was no evidence of a break-in, so it was conjectured that her killer was an acquaintance.

It wasn't long before detectives had traced her movements on the night of her death to a local bar, and then to a drinking partner named John Wayne Glenn.  He admitted knowing her and claimed that while he had taken her home that night, he had not entered her apartment.

However, a search of his place turned up a detective magazine that contained a story that mirrored how the victim was killed.  He continued to deny any involvement and since there was no physical evidence against him, officials decided to try the newly invented PSE.  Glenn agreed to submit to the procedure.  He was asked questions in a taped session and the tape was fed into the PSE.  The results indicated obvious deception on Glenn's part, which made the investigative team redouble their efforts in a more focused way on this one suspect. 

They searched the victim's bathroom again and eventually found a faint palm print on the tub that was linked to Glenn.  He quickly revised his story to admit that he'd been in the victim's home many times and had slipped once in the bathroom.  He was arrested and went to trial.  The jury didn't buy his story, and he was convicted.  Although the results of the PSE could not be used as evidence, the device had helped to focus the investigation until hard evidence could be turned up.

Other devices like this that rely on using the voice to detect stress include the Mark II Voice Stress Analyzer (VSA), and the Computerized Voice Stress Analyzer (CVSA).

Advantages over the polygraph, besides more comfort to the subject, include the fact that voice analysis can be done over the phone and in a variety of conditions as long as a tape of good quality exists.  That means that voice analysis can be done secretly, especially in cases of suspected security breaches.

However, the American Polygraph Association did a study on the CVSA and concluded that voice stress analysis is no better than chance.  They also pointed out that, while the Department of Defense uses polygraphs, it does not employ voice stress analysis in any investigative context.  In their 1996 study, the DOD found no evidence in support of using the CVSA as an instrument for detecting deception.

Even so, lie detection via polygraph has still not achieved solid respect.  In the case of the 1996 bludgeoning death of six-year-old JonBenet Ramsey, the parents, John and Patsy, came under heavy suspicion.  They refused to take a polygraph test administered by the FBI because they did not trust the agency.  Instead, over three years after the crime, they agreed to submit to tests by independent examiners.  One was inconclusive, but another appeared to clear them of knowledge of, or complicity in the crime.  Still, the Boulder, CO, investigative team were not impressed.  Apparently, the age-old suspicion of the lie detector's potential for ambiguous interpretations was still in play on both sides.  This standoff might have been resolved with paired results—test administration and interpretations done by examiners chosen by both sides.  Until then, the Ramseys probably will not have a publicly viable claim to honesty, based on their test administrator.

Yet it cannot be denied that since virtually all studies have shown that most people show poor skills at detecting deception, and since the polygraph is currently the most accurate device available for this purpose, its use is a plus for assessing credibility of a subject being questioned.  The Committee of Concerned Social Scientists submitted an amicus curiae on the polygraph in United States V. Scheffer, concluding that the polygraph is "a valid application of psychological science" which is "generally accepted by the majority of the informed community of psychological scientists."  The error rate is acceptable in scientific research, and most of the problems have to do with human error, which arises in any situation where there is a human evaluator of data.  The polygraph has a long, solid history of testing and ought to be considered in court as a viable part of testimony.

Whether or not this device will ever become routine as admissible evidence, it clearly has the respect of many investigators and will likely continue to be utilized in both criminal and civil inquiries.  Increasingly more states are requiring licensure, which will pressure training schools to tighten their standards.  If this trend continues and if more high profile cases make use of it, the polygraph has the potential to make a significant contribution to future criminal cases.