Lie-detection using functional MRI

American last week reported that
psychiatrist Sean Spence and collaborators at the University of
Sheffield are
developing a lie-detection test based on functional magnetic resonance
(fMRI) technology.
Using fMRI, Spence and colleagues are able to monitor blood flow to
certain areas of the brain’s
prefrontal cortex that are implicated in the regulation of truth
telling, spotting patterns of blood flow associated with lying or
witholding the truth. The
technique promises to outperform existing polygraph technology, which
relies on
signs of anxiety that are notoriously unreliable indicators of lying.
It has already been used on the British reality TV show Lie Lab,
though Spence emphasizes that it is not yet ready for widespread forensic use.

the United States, the commercialization of fMRI lie-detection tests by
two companies – ‘No Lie MRI’ and ‘Cephos’ – has already provoked some
ethical worries.

One concern is that fMRI lie detection will be
adopted before there is good information on its reliability. The thought is
that undue significance might be attached to results which are in fact highly
uncertain. But suppose that the fMRI approach could be developed into a
lie-detection test that far outstrips polygraphs in reliability. Would its use
remain problematic?

Some worry that the technology might be misused
by, for example, anti-terrorism agencies. Such agencies might be tempted to
apply the test in situations where its use would amount to a gross infraction
on individual privacy and autonomy. This is a real worry, but it applies
equally to existing lie-detection technology which, though controversial, has
been widely used. It’s not clear why misuse of a more reliable test should be
more problematic than misuse of a less reliable one.

concern is that it is difficult to
define lying or truth-telling in a precise and uncontroversial way. But
also difficult to define many personality traits or motives that serve
grounds for suspicion in criminal investigations. Yet judgments about
personality or motive are routinely made in such investigations. If
investigators were prevented from making judgments that rely on poorly
defined concepts,
their forensic capabilities would be drastically limited.

concerns about the risk of exaggerating
the reliability of test results might remain; even a fully developed
fMRI lie-detection test would surely give some false positives
and false negatives. But again, it’s not clear why the reliability of
an fMRI
test would be exaggerated any more than, say, the reliability of
polygraph technology, or the various other technologies routinely used
forensic scientists. The tendency of some to exaggerate the reliability
of technology-based tests has been a longstanding problem in forensics,
and it’s one that should certainly be confronted. But it hasn’t
generally be seen as a good ground for rejecting new technologies,
presumably because the benefits of those technologies have been thought
to outweigh the risks of misinterpretation.

The possibility of an fMRI lie-detection test certainly raises
some ethical issues. But its not clear that any of these issues are new, nor
that the ethical concerns surrounding the use of such a test would be any
greater than those surrounding the use of existing forensic technologies.

Main sources:

L. Greenemeier, Are you a
liar? Ask your brain. Scientific American, 15
November 2007. Available at

Lure of lie detectors spooks
Nature 441, 918-919; 22 June 2006. Available at


D. D. Langleben et al. NeuroImage 15, 727–732;

J. Wild. Nature 437, 457; 2005

F. A. Kozel et al. Biol. Psychiatry 58, 605–613;

E. Check. Nature 435, 254-255; 2005

J. Knight. Nature 428, 692-694; 2004

  • Facebook
  • Twitter
  • Reddit

Comments are closed.


Subscribe Via Email