Researchers in the U.K. have identified two simple techniques to beat a brain imaging lie detector that’s been touted as a possible replacement for earlier methods, such as the polygraph test.

Functional magnetic resonance imaging (fMRI) machines are purported to provide evidence of a person’s honesty by scanning certain areas of the brain that are activated when they’re lying.

Unlike polygraph or other lie detection tests that track a person’s physiological signs -- such as sweating, heart rate, and blood pressure -- to determine guilt, fMRI tests measure the amount of blood flow to activated areas of the brain during a lie.

“When you lie, your brain is engaged much more than when we’re telling the truth because you’re trying to suppress the truth when you lie,” study author Giorgio Ganis, an associate professor in cognitive neuroscience at the University of Plymouth, told CTVNews.ca during a telephone interview from England on Thursday.

The assumption is the brain will show signs of recognition when presented with a “concealed” piece of information or item, which the person is trying to hide, during the test.

In order to conceal those signs of recognition, the areas of the brain responsible for directing attention and decision making will work harder, and the increased blood flow to these regions will light up in the scan, according to the researchers.

Ganis used the example of a thief who stole a ring and is later shown an image of the ring during a concealed information test, which involves showing a person objects taken from the crime scene along with other random “control” items.

“If you’re the perpetrator and I show you this ring together with a bunch of other valuable items, your brain is going to show a much higher response,” he explained. “If you’re an innocent person who was not present at the crime scene, there won’t be any difference because you won’t be able to tell that that ring was actually the item that was stolen.”

It’s been suggested that fMRI tests are more difficult to trick than physiological ones where a person can conceal their physical “tells” by intentionally recreating them when they’re provided with an innocent “control” question or item, thus confusing the machine.

However, Ganis and his team of researchers from the University of Plymouth, in collaboration with the University of Padova in Italy, have discovered that it may not be as hard to beat fMRI technology as some would believe.

Mental countermeasures

In the first study of its kind, academics set out to explore the effects of “mental countermeasures” on brain activity displayed in fMRI scans.

To do this, the researchers taught 20 participants two mental countermeasures to use during concealed information tests to see if the results of the fMRI scans would be affected.

During the test, the participants were asked to conceal information about a secret item they saw inside of an envelope. They were then presented with a series of random control items along with the item they were trying to conceal and asked to use the countermeasures they were just taught.

The first countermeasure was to attach meaningful memories to the control items to make them more significant so they activate the scanned areas of the brain. Ganis used the example of a bracelet as one of the control items used in the stolen ring scenario.

“You have to very quickly try to remember something interesting about a bracelet. Perhaps you remember that your mother gave you a bracelet or something from movie,” he said. “Now this control item, the bracelet in this case, is no longer meaningless. It becomes meaningful to you.”

The second countermeasure was to focus on the superficial or aesthetic aspects of the item they were trying to conceal in order to make it less meaningful to them so there would be less activation in the studied region of the brain.

“If you see for example a ring, focus on the shape of the ring, the shape of the gem, something like that. Try not to think that this item is actually familiar to you,” Ganis said.

The results showed that the countermeasures lowered the accuracy of the fMRI tests by 20 per cent because it was more difficult for the machine to detect differences in brain activity.

“The point of this study, the big picture, is that we wanted to show that it’s relatively easy to train somebody to beat this test,” Ganis said.

In light of their findings, the researchers said more research into fMRI tests is needed before they are used by law enforcement or in criminal investigations.

“This study shows that the process can be manipulated if someone associates meaningful memories to the control items, or focuses on the aesthetics, rather than the memory, of the item they’re trying to hide,” leader author Chun-Wei Hsu from the University of Plymouth said in a press release. “None of our participants were seasoned liars or criminals, they were just everyday people.”

Ganis said they hope to study whether it’s possible to identify when people are using countermeasures and if there is a way to prevent them from doing so in these types of lie detection tests.

“Deception is a really challenging area of psychology, and the more we can find out about the techniques used to detect it, the better,” Hsu said.