Effective MRI lie detector

[youfailedx]

Even the person who formed this operation, U Penn scientist Daniel Langleben, says that using the fMRI as a real "lie detector" is over reaching.

[youfailedx]

Too bad Grant from Mythbusters fooled this thing.

This is supposed to work by revealing the brain's activity when formulating a response (i.e. lying, instead of just remembering something). All Grant did was make his brain work hard on every question he was asked. This caused the operator to misidentify the question that Grant was lying on.

[youfailedx]

It seems that you need to do is have your story all planned out and remain calm. Or just do what Grant did. Maybe feign an elaborate story in your head for every question you are asked. Then the operator won't have any base for comparison.

No "lie detector" is accurate. The polygraph, Computer Voice Stress Analysis, and Layered Voice Analysis have all been revealed as junk by scientific study. I suspect it is only a matter of time before this thing is revealed as fallible too.

Good riddance.

[ClamCrunchy]

@youfailedx Not 100% accurate, but better than chance accuracy. The fact that a method is not 100% accurate doesn't make it junk, rather the fact that it can do no better than chance does. Both fMRIs and polygraphs have an accuracy of about 60% - this may seem low, but it's still significantly higher than chance accuracy.