Monday, September 03, 2007

Would I Lie To You Yet Again?

The Deception Blog alerts our attention to yet another breathless and swooning report on reading the fMRI tea leaves of lie detection in their post, More fMRI stuff and nonsense. The ABC report says virtually nothing specific about what purportedly happens in the brain of someone lying compared to telling the truth, only:
The FMRI results visibly showed more blood rushing to the specific parts of the brain when I was lying.

"We're looking beyond just the anxiety, what you're doing cognitively when telling a lie. When you're telling a lie, you're overriding a normal response of telling the truth, " said George. "So your brain is putting on a brake and overlearning a response. It's harder to tell a lie then tell the truth, [to] remember what you're lying about. You've created a false way the world is. So we're looking at overworking, inhibiting, multitasking. Those areas are pretty predictive in catching someone when you're lying."
That, and the lovely figure below. Notice the only activity in the truth-telling brain on the left seems to located mostly outside the cerebral cortex. Great.

Who needs Pinocchio's nose to find a lie? The FMRI scan on the right detects a brain processing a false statement; the less colorful brain on the left corresponds to someone in the middle of a truthful statement. (ABC)

Subscribe to Post Comments [Atom]


At September 04, 2007 9:27 AM, Anonymous Anonymous said...

The neurocritic writes:
> That, and the lovely figure
> below. Notice the only activity
> in the truth-telling brain on
> the left seems to located mostly
> outside the cerebral cortex.
> Great.

Whee! Well, I have no idea what the study actually did and every reason to believe that it was done incorrectly, but I think the Neurocritic is making a slightly enthusiastic inference here.

You will note that the activations are being plotted on a single subject's anatomy (a no-no, in my book), but they are almost certainly average (and Gaussian-blurred) activations from many subjects. So, although there are no guarantees, I am willing to give the benefit of the doubt that, on average, there was actually cortex beneath the plotted activation.

Not exactly a ringing endorsement, but there you go.

At September 04, 2007 12:44 PM, Blogger James Marcus Bach said...

Well if I were the defense expert on a case involving that technology, I would have some good counterarguments against this.

I would challenge the testing process by which the machine was certified, based on the arbitrariness of calling something "true" and something else "false". It's one thing to do this in a laboratory with standardized questions and test subjects who are not in actual fear for their lives or liberty, but quite another to try this on an unwilling subject.

The technology relies on an inference that the test subjects are identical to all future field subjects in their brain functions, attitudes about being questioned, and understanding of the questions.

It ignores the problem of interpreting the question and polite phrasing of answers. The technology assumes that that the process of lying is always a process of cognizing something, then by an act of will deciding not to say that something, and substituting something else instead. There's another term for this: being polite. And another: being careful to tell the truth.

I have testified under oath several times, and the way I answer questions in that situation is far different than how I answer them naturally. That's not because I'm lying, but because I must not be loose in my replies such that the opposing lawyer can infer spurious meanings into my testimony that I did not intend.

I think perhaps what this technology is really doing is showing the difference between blurting out an answer and considering an answer before making it. I would point out that no fancy technology is needed to look at a suspect and see if he's hesitating or halting in his replies-- see President Clinton's famous "definition of is" answer.

At September 04, 2007 12:56 PM, Blogger The Neurocritic said...

Yes, Anonymous, I'll agree that I was being overly critical (gasp!) based on the blurred activation presumably from many subjects -- but maybe not, since the technique is only useful on the single subject level -- and superimposed on one person's brain. And we don't know the comparison conditions for either the lying or the truthful brain. Plus, an ABC news report isn't exactly a manuscript in Science or even Neuroimage.

The next question is why only a small activation in Brodmann area 10 of the frontal pole is associated with telling the truth...

At September 04, 2007 2:21 PM, Blogger The Neurocritic said...

James - thanks for your comments, you made many good points.


Post a Comment

<< Home

eXTReMe Tracker