The All About Forensic Psychology Blog (28 May) picked up on an article by Robert Ellman (Intrepid Liberal Journal) in Political Cortex about the implications of brain scanning for lie detection in forensic settings. AAFP author David Webb writes:
Fascinating article adressing the potential use and implications of Functional Magnetic Resonance Imaging (FMRI) i.e. brain fingerprinting. [...] Given the concerns regarding the documented unreliability of the polygraph (lie detector), Silberman contends that subject to further empircial testing, Functional Magnetic Resonance Imaging (FMRI) may prove effective in solving crimes or preventing terrorism.
I’m enjoying the AAFP Blog a lot: David often finds interesting FP articles that I wouldn’t otherwise come across. But, at the risk of being picky, I need to correct a few things David’s post about brain scanning. To be fair, his post reflects confusion in the original article about the differences between brain fingerprinting, brain scanning and fMRI. They are not the same thing. Let’s try and sort this out, without getting too technical:
Brain scanning is a generic term used in many different contexts by many different authors usually to refer to any technique that measures brain activity.
Brain Fingerprinting is the name given by Lawrence Farwell to a technique for detecting guilty knowledge using EEG. EEGs measure electrical activity across a subject’s brain (electrical signals known as Event Related Potentials or ERPs) when the subject is exposed to a stimulus. Farwell claims that Brain Fingerprinting can detect whether a stimulus is novel or familiar to the subject.
fMRI is a technique for measuring brain activity by measuring changes in the concentration of oxygenated haemoglobin in the brain. In deception detection, using fMRI can, researchers suggest, detect changes in the pattern of subjects’ brain activity when lying compared to when they are telling the truth.
None of these techniques consitutes a ‘lie detector’. Nor is the polygraph (sorry, David, I’m being picky again). The techniques simply measure changes in physiology or brain activity in response to a stimulus. The decision as to whether that change is a result of deception depends on the stimulus that caused the change (e.g. the question being asked, or, perhaps, the photo being shown) and how that change is interpreted.
For anyone new to this area – or even with a bit of knowledge of the field – I’d recommend a great article that appeared in the American Journal of Bioethics last year on the promises and perils of lie detection. The paper, by Paul Root Wolpe, Kenneth R Foster and Daniel L Langleben (all at the University of Pennsylvania) does a terrific job of setting out clearly some of the problems in deploying Emerging Neurotechnologies for Lie Detection and is also a good source for summaries of the theoretical, ethical and practical issues involved in deploying EEG measurements and fMRI for deception detection.
As far as the authors know, Farwell’s Brain Fingerprinting is the only one of the new neurotechnologies that has been used in real-life forensic settings and only once (according to Wolpe et al.) has it been used in court in the US. As Wolpe et al explain, in Harrington vs Iowa (a 2003 post-conviction relief hearing), Farwell submitted Brain Fingerprinting evidence that Harrington had no knowledge of the crime scene. Harrington was freed, but Wolpe et al. contend that the Brain Fingerprinting evidence was essentially irrelevant to the court’s decision. They add,
as the State of Iowa complained in its brief against Brain Fingerprinting in Harrington, the most critical problem with admission of Brain Fingerprinting evidence is the lack of any track record establishing its reliability (p47).
Despite this, it appears that the Indian Police is using Brain Fingerprinting in real-life cases, where suspects’ liberty is at stake. This blog – and OmniBrain and Neurofuture – mentioned these concerns before and Robert Ellman picks up on more evidence that this untested technology is being used by the Indian Police. He provides links to articles in New Kerala newspaper highlighting how Brain Fingerprinting (not fMRI, as Ellman suggests) is being used, for instance:
The court had earlier awarded life imprisonment to Javed and six others [...] for rioting and murdering a man on November 11, 2003 [...] The judge awarded the sentence after considering the results of the brain fingerprinting tests performed on the accused, among other facts in this case.
Ellman also highlights concerns that ‘cognitive profiling’ might one day be applied to
[...] search suspects for brain waves that suggest a propensity toward violence — a sort of cognitive profiling. ‘You can do an FMRI scan showing that the structures in the brain responsible for impulse control and empathy are underactive and the parts of the brain responsible for aggression and more animalistic, violent activities are overactive,’ Snead explained. ‘Maybe with these nascent technologies, we’ll be able to develop some kind of profile for a terrorist.’ Suspects who show a propensity for violence might be detained indefinitely as enemy combatants even though they committed no crimes.
Minority Report here we come.
UPDATE (16 June):
Sandra helpfully points out in the comments below that the ‘Brain Fingerprinting’ being used in India is not the same as the BF developed by Lawrence Farwell, though it seems to be based on the same principle.