Lie Detectors - here we go again?

[Blondin]

Brain Scan Shows Differences in Truth, Lying (http://story.news.yahoo.com/news?tmpl=story&cid=585&e=1&u=/nm/20041129/sc_nm/science_lying_dc)

Do you think this sounds any more plausible than measuring heart rate, breathing rate, skin resistance and all that other stuff as done in 'traditional' lie detectors? Sounds like there is still an awful lot of interpreting to reach a conclusion.

[BobK]

It's a pathetically small sample size and how do they determine whether the difference in brain activity is caused by lying, or the fact that only the liars were active participants in shooting the gun?
Seems to me that would make a difference in how the brain would remember the event.

[Blondin]

That was my feeling, too. This is an awfully premature announcement.

[Michael Redman]

How can you design an experiment where your "controls" are undertaking different activities? How can anyone possibly think the results are meaningful? Am I missing something?

[Bikewer]

Another technology that seems to be resurgent in police circles is Voice Stress Analysis.
I recall when this first became popular some years ago, and various folks were "analyzing" the speeches of prominent politicians and whatever, duly pronouncing that they were lying...(like that was a revalation....)

At any rate, I hadn't heard much about the technique for some time, until one of the local departments began using an individual to interview suspects as part of investigations.

I do not know the current status of this (if any!) in the scientific community, and to my knowledge the results of any such test would not as yet be admissable in court.
Of course, neither are lie-detector (polygraph) tests, but the device is still in general use both in criminal investigation and background checking.

[EdipisReks]

Robo Cop had a built in voice stress analyzer. Robo Cop is about as scientifically valid as the experiment linked.

[Elind]

Originally posted by Blondin
Brain Scan Shows Differences in Truth, Lying (http://story.news.yahoo.com/news?tmpl=story&cid=585&e=1&u=/nm/20041129/sc_nm/science_lying_dc)

Do you think this sounds any more plausible than measuring heart rate, breathing rate, skin resistance and all that other stuff as done in 'traditional' lie detectors? Sounds like there is still an awful lot of interpreting to reach a conclusion.

There's another thread on this.

Traditional lie detectors have holes large enough to drive a bus through but, as I said elsewhere; brain paterns for manufacturing a fantasy (lie) would seem quite logically to be totally different from simply repeating a memory, in theory.

[Vitnir]

What about human lie detectors? I saw an episode of Derren Brown's (http://www.derrenbrown.co.uk/) Mind Control where he was walking blindfolded and his guide tried to trick him into walking the wrong way through various obstacles. Rigged or is it possible to read cues that obscure? Why would it be impossible for a machine to do the same?

[Bikewer]

Martin Garnder had an extensive article in one of his books on the numerous ways one can "cheat" blindfolds.

[Earthborn]

Originally posted by Elind
brain paterns for manufacturing a fantasy (lie) would seem quite logically to be totally different from simply repeating a memory, in theory.Shouldn't be so hard to fool such a thing. If it can measure whether you are remembering something or making something up, just make something up before you are tested. Then when you are tested all you have to do is remember what you made up.

If that is not enough to fool the machine, and it can distinguish between a memory you made up and a memory of something you really remember, all you have to do to beat the machine is believe your own lies. Doesn't seem hard to do for most people.

[Elind]

Originally posted by Earthborn
Shouldn't be so hard to fool such a thing. If it can measure whether you are remembering something or making something up, just make something up before you are tested. Then when you are tested all you have to do is remember what you made up.

If that is not enough to fool the machine, and it can distinguish between a memory you made up and a memory of something you really remember, all you have to do to beat the machine is believe your own lies. Doesn't seem hard to do for most people.

This is not mind reading, it's pattern recognition. It reminds me of experiments that determine that small children cannot lie until a certain age around 3 or 4. It requires a different way of thinking.

Any lie is "made up", regardless of how far in advance it was prepared.

I'm not saying I know this is or can be foolproof, but it seems logically consistent to me and my bet would be that with the advances in real time and high resolution monitoring of brain activity that we read about, it will be practical in the future.

[Giz]

A friend of mine saw this story, his take was that you could fool it if when they ask you "did you kill Colonel Mustard in the Library with the lead piping?" you ask yourself (internally) "Is mauve my favourite colour?"

Any "No!" then comes out nice and true...