When AntiPolygraph.org
requested the Department of Defense Polygraph Institute (DoDPI)
Federal Psychophysiological Detection of Deception Examiner Handbook under the Freedom of Information Act, the Defense Security Service (DSS) withheld, among other portions, Chapter 7, which concerns the "acquaintance test" (also called a "numbers test" or "stim test"), a bit of gimmickry used to trick examinees into believing that the polygraph can detect deception.
DSS claimed that "disclosure could enable circumvention of the polygraph test by others."
However, it turns out that a DoDPI report previously approved for public release includes detailed instructions on how to administer the "numbers test." Richard Widup and Gordon H. Barland's March 1994 report, "Effect of the Location of the Numbers Test on Examiner Decision Rates in Criminal Psychophysiological Detection of Deception Tests" may be downloaded as an 870 kb PDF file here:
http://antipolygraph.org/documents/dodpi94-r-0015.pdf The procedure for the "numbers test" is described at pp. 4-5 (14-15 of the PDF file):
Quote:Procedure
The manner in which the numbers test was explained to the examinee was left to the discretion of the individual examiner. Although all examiners learned a standard procedure at DoDPI, with experience they typically develop an individualized style of administration. No attempt was made to standardize the administration of the numbers test just for this study. A typical numbers test might be conducted as follows.
"John, I am now going to conduct a test in which I want you to answer 'No' to all of the questions. The main purpose of this test is to determine if you are physiologically capable of being administered a PDD examination today. The way in which this test will be conducted is that I want you now to pick a number between 3 and 8, that is not either 3 or 8. Now, have you picked your numebr? Good. Now, please tell me your number." (The examinee answers that his number is 5.)
"Okay, John. As I recall, you are right handed, correct? Now, I want you to take this pen, and with your left hand write the number you picked on this piece of paper, somewhere in the middle. Please write the number a couple of lines high. Now, I am going to surround the number you wrote with numbers that I will write. I am now going to place this paper on the wall in front of you."
"We will now go over the way in which this test will be conducted. Keep in mind that I want you to answer 'No' to all the questions. The test will go this way. Regarding the number you wrote on that piece of paper, did you write the number 2? (The examinee answers 'No.') Did you write the number 3? (Answer: No.) Did you write the number 4? (Answer: No.) Did you write the number 5?" (Answer: No.)
"John, what is that answer? (The subject answers that the response was a lie.) That's right, John; that is a lie. Now, both you and I know that you wrote the number 5, correct? In fact, I saw you take your left hand and write that number on that piece of paper. I want you to answer 'No' nonetheless, and I will tell you why after this test is over. Did you write the number 6? (Answer: No.) Did you write the number 7? (Answer: No.) Did you write the number 8? (Answer: No.) Very good, John. Now I am going to start at the top of the list and go to the bottom, and I want you to follow along with me with your eyes, answering each question with a 'No.' John, do you have any questions? Now I will start the test."
At this point the test is administered. Immediately after the numbers test, the examiner shows the chart to the subject, pointing out the responses which occurred at the selected number. The examiner asks the subject what caused those responses. Most subjects reply that they do not know, whereupon the examiner replies "John, these responses occurred when you lied about the number you wrote. Now, lying about a number is normally not a big thing, is it? But it was to you, John, because you saw how big those responses were. Now the other questions on the main test are more important, aren't they? John, you can imagine how large those responses would be if you are lying to me about any or [sic, correct "of"] those, can't you? That's the reason you need to be 100% truthful with me here today."
Many polygraph subjects do show a significant physiological response when they reach the number they chose.
But such responses have nothing to do with deception. When the subject replies 'No' to the number he wrote, he isn't attempting to deceive the polygrapher: he is merely following instructions. The process of having the subject pick the number and write it with his weak hand tends to make that number significant to him compared with the other numbers in the series. For more on "stim tests," see Chapter 3 of
The Lie Behind the Lie Detector.