Scientific Control and PL-CQT Comparison Questions

[Drew Richardson]

Polygraphers,

I was scheduled to testify later this week in a criminal matter in which I would have opposed any significance being placed on a particular law-enforcement administered polygraph exam and results.  I was to be countered with a well-known research scientist in the polygraph community.  I have been informed that the prosecutor has decided not to call the expert and not to introduce anything regarding the stipulated exam and results.  The area I would like to stimulate some discussion regarding here has nothing to do with the aforementioned but one narrow aspect of the criticism I would have made of the PL-CQT exam that was administered to the defendant.  The "test" included three relevant and three comparison/control questions.  All of the relevant questions were along the line of "Did you touch xxx's vagina in a sexual manner?" and the controls were lie controls (e.g., Prior to your 21st birthday, did you ever tell a lie to someone who really trusted you?)  I would have suggested that in addition to the fact that the identity (as to question type) of the relevant questions is more than blatantly obvious to anyone accused of touching xxx's vagina and quite likely to cause autonomic reactivity with anyone concerned with the consequences of being found deceptive to such a question (i.e., that is everyone—innocent and guilty examinee alike), the relevant questions differ in many respects, chief among them emotional content (highly inflammatory, highly offensive in nature) from the relatively neutral comparison questions.  In other words the latter group of questions offer absolutely no scientific control whatsoever with regard to autonomic reactivity in this exercise.  I would maintain any suggestion otherwise involves completely contorted reasoning which would fly completely in the face of common sense and that any effort expended at "setting" comparison questions under such circumstances is completely wasted effort.  Any thoughts, darkcobra2005, nonombre, others?

[Drew Richardson]

Darkcobra,

Part I response (Part II will follow when I return from travel.)

You are correct.  This exam was of the "Did you do it?", "Did you do it uh huh?" variety of (tri-) ZCT (two relevants related to touching the vagina of the young female in question (the phrase "bare vagina" used in the latter of the two) and a third question regarding touching any child in a sexual manner).

You write:

...It is my opinion that the pretest interview is extremely important in presenting the questions and the examiner must be totally unbiased and present him/herself in that manner....

I assume then that you would agree with me that it would be inappropriate for the examiner to be a participant in pre-polygraph initial (non-polygraph) investigative interviews of the victim in which such bias might develop, yes?

You further write:

...I do not suggest that any polygraph examination should be admissible in court, nor would I suggest that any person should be assumed guilty as a result of a polygrpah examination, it is simply a tool for investigative purposes....

Then I would assume that you would also agree with me that it would ALWAYS be improper for law enforcement/prosecution (or the defense for that matter) to seek a stipulated exam whose results would be admitted into trial testimony/evidence, yes?

Your further write:

...I would be of the opinion that an innocent person would be very focused on the relevant questions, the consequence is prison and loss of any respect from the community at large.  The consequence for the control must be of the same intensity.  That is the problem with the pretest and making a control have the same weight to the innocent as the relevant.  It would be difficult, not impossible....

In this passage you make 4 statements.  In order they are:

(1)   "I would be of the opinion that an innocent person would be very focused on the relevant questions, the consequence is prison and loss of any respect from the community at large."  TRUE and a statement whose significance can not be overstated and which can not be dwelled on for too long a period of time.

(2)  "The consequence for the control must be of the same intensity."  TRUE as far as it goes.  Not only must the intensity (emotional content) be equivalent, but also the nature/type of question must be indistinguishable from the relevant.  Not going to happen.

(3)  "That is the problem with the pretest and making a control have the same weight to the innocent as the relevant." Same as before-not going to happen.

(4)  "It would be difficult, not impossible." IT IS SO NEARLY IMPOSSIBLE AND THEREFORE SO UNTRUSWORTHY AS TO NEVER BE A VIABLE TECHNIQUE.  But for the sake of conversation, let's say it's not completely impossible.  I will suggest that any degree of success with your efforts (at setting comparison questions) to make it possible is completely unverifiable.  You can watch videotapes all day long of your performance, but you have no idea what is going through the mind of the examinee, and to continue (with the procedure and in-test phase) you have to ASSUME the most unreasonable of outcome(s) to proceed, i.e., that the examinee doesn't recognize relevant questions or is impacted by their clearly greater significance.  And so we have arrived at the crux of this discussion.  You have no idea what impact your comparison questions have.  Perhaps they have some (not likely in this case), perhaps they don't.  You just don't know.  As a drug chemist when I add an internal standard to my assay, I know exactly what it is doing and how it behaves in my experiment.  You do not have any similar control with you experiment and therefore have no scientific control whatsoever.  Some years ago when members of your community decided to no longer refer to "control questions" but to "comparison questions", that was in response to a cry for honesty.  That being said, your community is no less responsible for scientific control now that it has removed the word from your nomenclature.  Best till later and thanks for your reply...