New Haven Register (New Haven, CT)
SCSU PROFESSOR DEVELOPS TEST AIMED AT DETECTING LIES
Idea is to weed out dishonest witnesses, jurors
NEW HAVEN — We all reportedly lie, though some more regularly.
Kevin Colwell will tell you that’s the truth, and his recent work could help law enforcement and trial lawyers tell truth from lies.
Colwell has developed a simple test that may help determine whether a witness to a crime is lying or whether potential jurors are being honest about how much they know about a case.
Colwell, a professor of psychology at Southern Connecticut State University, believes his method, in which an investigator or lawyer asks a series of either/or questions, will lead to crimes being solved more quickly and to fewer mistrials because of jurors who are dishonest about what they know.
By using statistics, his “quick paper and pencil test” enables police to weed out the dishonest witnesses — who then may be further questioned — in a short amount of time.
The test is deceptively simple: “Did you see a man? Yes or no. Did he have a gun or a bat? Did he have a hat or a hoodie? And with those things — given you already know information — you can compare what they say to what they claim to know and see whether or not they’re being honest with you,” Colwell said.
Michael Lawlor, associate professor of criminal justice at the University of New Haven and former adviser on criminal justice policy to former Gov. Dannel Malloy, said Colwell’s technique could be helpful in jury selection.
“I can see if the goal was to weed out people that know about the details of the case, this might be effective,” he said. Connecticut is unique among states in that the state Constitution requires individual questioning of each potential juror, known as voir dire, he said. Colwell’s questions could be asked by a lawyer during that process.
“The lawyers can ask whatever questions they want, so there’s no reason a lawyer couldn’t ask these questions of every single witness,” Lawlor said. If a potential juror failed the test, a lawyer could reject him by using a peremptory challenge, which requires no specific cause.
“It might make it easier to convince the judge that this person can’t be fair,” he said.
But Lawlor said advertising the technique as a way to speed up jury selection may not be a winning argument with all lawyers. Dragging out a case may be seen as an advantage, with a defendant hoping for additional evidence to turn up. “It’s a leap to assume that everybody would want to speed it up,” he said.
Credibility assessment
Colwell said it’s important to be able to tell liars from truth-tellers.
“Most of us don’t go two days without lying. It’s not quite every day, but on average you tell two lies every three days,” he said. “Men lie slightly less frequently than women. But women lie more often to make other people feel good, whereas men lie to make themselves look better or to get out of trouble.”
When it’s done for a pro-social reason, Colwell said, lying is good. “People that lack social awareness lie less frequently than people that have social awareness, and most lies are done as a way to maintain some group cohesion.”
But that’s not what his work is about. “I’m focusing on that specific group of lies that is more anti-social, and I’m also doing a lot of what’s called credibility assessment, which means hoping to corroborate the truth,” he said. “I’m really looking for honest people so I can help investigators know that that person’s being honest. That’s my real goal. I’m not really trying to convict people.”
But even though people lie, aren’t police and attorneys trained interrogators, who can tell whether people are telling the truth?
No, Colwell said. “What’s lingering in the background is this: Law enforcement is no better at detecting deception than you are. And so if we don’t give them tools, they’re not any better than the average person.”
This has been studied dozens of times, he said.
“I start my training seminars by having police tell the truth or lie to each other and having other police try to … determine whether they were just told a true or a deceptive statement by their colleagues,” he said. “Last time I did it, the cops were wrong 13 out of 16 times.”
Overall, he said police will identify a lie or a truth no more than 60 percent of the time. The same goes for attorneys or forensic psychologists, he said.
When there is a crime scene, police need to separate the honest from the dishonest. “A bombing, a shooting, any big crime where there’s a whole lot of witnesses, we can take statements from the first couple of witnesses, take the information they provided, and create a paper-and-pencil test really quickly from it,” Colwell said.
The test is given to all the witnesses “as a way to isolate those people that are trying to hide information,” he said.
The trick is in choosing the witnesses to interview first. “That’s the hardest part of the whole thing, is that you’ve got to start with cooperative witnesses,” Colwell said.
“You need people that are likely to have been in the middle of the event and they’re willing to cooperate with you,” he said. “That’s the trick. That’s where law enforcement experience, etc., is going to be important, where they try to find a couple of people that they believe they can trust and that had a good perspective on what happened.”
The other use is in selecting potential jurors, “if somebody knows about the case from the press, but wants to hide the fact that they already know about it, and that happens in high-profile cases,” he said. An example would be he highly publicized O.J. Simpson trial.
“In the case of jury selection, it’s easier because there’s already been a significant amount of information gathered by law enforcement. … You use the information that’s in the popular press to create the instrument in the jury selection case,” Colwell said.
A defense attorney and a prosecutor may have different agendas when it comes to jurors knowing what has been publicized in the news media.
“What we don’t want is somebody to have an opinion about the case already and lie and act as if they don’t know anything about it during jury selection,” Colwell said.
“Each side is going to want to know whether or not the potential jurors already have information about this particular case,” he said. “And different attorneys can use that different ways. They just want to know.”
Someone who is guessing on such a test will get half right, on average, Colwell said. Someone answering truthfully will get most correct. And someone who is lying will get a high number wrong, Colwell said.
The more someone misses, the higher the chance they are lying. Six or seven wrong wouldn’t be statistically helpful, but eight tells the interrogator what he wants to know. That computes to a 95 percent chance the test taker is lying.
“If I am wondering whether or not you are deliberately withholding information from me, then you would have to get eight out of 10 wrong for me to say, ‘Aha, I know this guy is deliberately scamming the system because he missed so many. He had to know what was going on,’” Colwell said.
The system isn’t perfect. “There will be, every now and then, four or five people out of a hundred that we have a false positive where we think they’re deliberately making mistakes when they’re not,” he said. “This is just a screen. This is just: who do you elevate to the next level?” That may be a more indepth interview with an attorney present.
Colwell, whose work has been published in the journal of the American Polygraph Association, among others, has been consulting for the military and intelligence agencies for 20 years. He has trained agents of the National Counterterrorism Center, the FBI and the High-Value Detainee Interrogation Group, an “elite task force that Obama created.”