In a small room at the end of a corridor on the third floor of a concrete building, a woman is perched on a chair facing a TV screen and holding an Xbox controller.
A plastic and metal contraption has been fitted to her head. Three clusters of metal pins attached to the headset poke into the woman’s scalp at the front, crown and rear of her skull. Behind her, a man sits at a long table studying a laptop. Zigzags like the ones used to show a pulse on a heart monitoring machine shimmy across his screen.
The word ‘Matakana’ flashes onto the TV screen. Before it disappears a few seconds later, the woman clicks the right-hand button on the Xbox controller. The wavy lines on the computer screen pick up their tempo as memories of holidays in the village of Matakana, north of Auckland, flood into her mind.
It may sound like a scene out of a science fiction or crime thriller movie, but this is very much reality. The room is at University of Canterbury’s law school, and the woman – undergoing a demonstration of landmark research into a forensic investigation technique known as brain fingerprinting – is me.
*Can we trust forensic science?
*‘Rapid DNA’ promises breakthroughs in solving crimes. So why does it face a backlash?
*National forensics plan raises miscarriage of justice fears
*How science may just end up killing crime
While brain fingerprinting was invented in the 1980s by US neuroscientist Dr Larry Farwell, it’s never been used in New Zealand – until now. Kiwi academics are investigating it as some believe it could revolutionise the justice system.
The fingerprinting is based on an electroencephalogram test, which detects electrical activity in the brain through metal discs attached to the scalp. Doctors often use a similar method to diagnose epilepsy and other health conditions that affect the brain.
The basic premise is that the technology measures a person’s brainwave responses to certain stimuli. For example, during a murder investigation, a police officer could present a suspect with a picture of a weapon and if the suspect’s brainwaves responded in a particular way, that could indicate they recognise the weapon and, therefore, have knowledge of the crime.
Proponents say it could help solve cold cases, catch terrorists and exonerate the wrongfully convicted. In the US, it’s been used in three criminal cases and has led to a man confessing to murder, and an innocent man being freed from prison after an accuser admitted lying. But critics say there isn’t enough independent evidence to prove its validity.
University of Canterbury law professor Robin Palmer and his criminologist colleague professor Greg Newbold say that if the technology had been available in New Zealand when Teina Pora was first accused of the murder of Susan Burdett in 1992, he may have never been charged, as a test could have shown he had no knowledge of the crime.
In what would become one of New Zealand’s most notorious miscarriages of justice, Pora was convicted at two trials and spent more than 20 years in prison, despite being innocent.
As a former prisoner, Newbold has always taken an interest in high profile cases like Pora’s, the Bain family murders and that of Arthur Allan Thomas, whom Newbold met while serving time for selling heroin.
He says brain fingerprinting could help the wrongfully convicted, but it will never be a silver bullet to solving crime.
“What happens in the case like Michael Wayne October, who’s very likely innocent of the murder he was convicted of [but] was so befuddled by alcohol, he has no memory of the whole of the night at all?” Newbold says.
He also doubts brain fingerprinting could help solve rapes as it would be difficult to prove from the test whether or not sexual contact was consensual. For brain fingerprinting evidence to be accepted in New Zealand courts laws would need to be created, which could take a long time.
“If a defendant or a witness refuses to take the test, what does that indicate? And what does the judge tell the jury? Is a person going to be legally compelled to take the test? Or are they able not to take the test?”
Palmer, who is leading the New Zealand studies, has a vision of one day having labs set up at police stations across the country. Documents obtained by Stuff under the Official Information Act show police, including the Canterbury district commander, support the research, which has been funded by the New Zealand Law Foundation.
“Imagine how much time and effort could be saved for the police in investigation,” Palmer says.
Instead of officers with limited resources chasing different leads, they could bring suspects into the lab and test them on information only the perpetrator of the crime would know. The tests would indicate who, if any, of them had knowledge of the crime and they would have to explain why and how they had knowledge of it.
However, even if a test shows the person has knowledge, the result would have to be tested in court, like any other evidence. A positive result wouldn’t definitively prove guilt, as a witness or an accomplice could also have knowledge of a crime.
But before it’s used at all, Palmer’s team must prove brain fingerprinting actually works.
While experts agree the science behind the technique stacks up, Farwell’s claims that the tests are 99.9 per cent accurate have attracted criticism from some in the field. In 2005, one of his most vocal critics, professor Peter Rosenfeld of Northwestern University in Chicago, tried unsuccessfully to replicate Farwell’s studies. In some tests, Rosenfeld’s accuracy was as low as 54 per cent. He argued that was evidence brain fingerprinting was not as effective as Farwell claimed.
The Canterbury project is the firsttest of the technology where Farwell has not been directly involved. Last month, the researchers completed the first stage, using university students as guinea pigs, but are unable to release the results until they’ve been peer reviewed.
In a couple of weeks, the second phase of testing is due to begin. That will see ex-prisoners who plead guilty to offences tested on the specifics of their crimes. The hypothesis is that the brain fingerprinting system will correctly identify which prisoners have knowledge of each crime. If all goes to plan, they’ll then investigate whether alcohol, drugs or mental health difficulties impact the results.
My own test, which used names and places of significance to me as stimuli instead of details of a crime, revealed a 96 per cent likelihood that I knew the information. Beforehand I was asked to name people, places and objects that were meaningful to me. The researcher then entered them into a computer programme and generated a list of random similar words, including a town in Canada I’d never heard of.
The test was broken up into 10 or so four minute blocks. Every few seconds a word would flash on to the screen and I was instructed to click the left-hand button on the Xbox if I did not recognise the word and the right-hand button if it was significant to me. I have to keep still and quiet during each test and refrain from blinking as even minor movements can trigger a brain wave response and render the test inconclusive.
One academic who’s confident in the validity of brain fingerprinting is professor Richard Jones, a neuroscientist and neuroengineer at the New Zealand Brain Research Institute. He designed the experiments being conducted in Christchurch.
“Some people would say [they’re] unbelievable results. In science you think, ‘that’s not possible, you can’t get 100 per cent’. But I do actually extensively believe them. Although I’m a naturally a skeptical-type person, I don’t feel skeptical about it,” he says.
Jones concluded it’s scientifically sound as the statistical averaging it uses to determine whether the person recognises the stimuli is legitimate.
Seattle-based Farwell, who has spent two weeks in Christchurch to teach the researchers how to use the machines, says it’s impossible to cheat because the brain’s response to stimuli is measured at one-300th of a second.
He’d stay at the laboratory until late at night to work on different projects. First, he built a computer interface that allowed a young man who was paralysed from the eyeballs down to speak by using his brain waves to activate a voice synthesiser.
Then came brain fingerprinting, which ended up forming his PhD research.
He says the reason some academics – such as Rosenfeld – have failed to replicate his results in their own brain fingerprinting research is because of method errors.
“In every case, whether it’s in my lab or six other laboratories, where the brain fingerprinting scientific standards are met, people have gotten very close to 100 per cent accuracy. When the scientific standards are not met, sometimes it’s no better than chance.”
University of Auckland associate professor of psychology Paul Corballis has been following developments in brain fingerprinting with interest since 1991 and summarises his position as cautiously optimistic.
Most of Corballis’ concerns stem from the perception that because it’s neuroscience, juries might view brain fingerprinting evidence as infallible – which he doesn’t believe it is.
Because alcohol and drugs can reduce the voltage of brainwaves, testing people who are under the influence could generate false positives and, ultimately, lead to the guilty walking free, Corballis warns.
“I think there is value there. [But] I think you have to be very careful how you interpret it,” he says. “I just kind of feel like we’re not there yet. There’s enough kind of little fish hooks that make me think it’s not quite ready for prime time.”