Lie detector tests have become a popular image thanks to many popular TV shows and movies such as Meet the Parents, Lie to Me and Maury, making us believe that they are reliable. In reality, the idea that a machine is able to detect the validity of what a person is saying based on their psychophysiological changes is more of a lie than anything. Three indicators are commonly looked for in "lie detection" tests: heart rate/blood pressure, respiration and skin conductivity. There are three common types of questions that are asked during a lie detection test: relevant questions ("Did you do it?"), irrelevant questions ("Is your name John Doe?") and control questions ("Have you ever betrayed anyone who trusted you?"). If there is more physiological responses to the relevant questions than the control questions, a diagnosis of "deception" will be made, if there is a greater response to control questions, a "nondeception" diagnosis is reached, and finally if there is no difference between the relevant questions and the control questions, the test is considered "inconclusive."
The problem with these tests is that they typically aren't separated from placebo-like effects. It is very easy to alter the results of the test simply by clenching ones' teeth, biting ones' tongue, curling ones' toes or trying to solve complicated math problems in ones' head. Lie detector tests have "lied" to the people administering them many times, falsely convicting people of crimes that they didn't commit.
This information brings us down to the final question: when will we stop believing in the machines that are supposed to detect lies, when all they do is lie to us about the results?
For more information on "lie detection" or polygraph tests, click here.