posted 07-08-2008 08:39 AM            

Initially I liked it but then I thought damn that is a seriously hot question...

posted 07-08-2008 01:39 PM            

------------------
Ex scientia veritas

posted 07-08-2008 03:17 PM            

With all due respect to Dr. Abrams, somebody better have a darn good reason for putting the victim's name in the comparison question. Maybe that will be necessary if the subject has been in treatment for 10 years, and is non-naive after completing 30 to 40, or more, previous polygraphs. But is it necessary? Is it wise?

In sex offender treatment there used to be therapy excercises called "masturbation satiation tapes," in which an offender is instructed to record his fantasies verbally, while masturbating to a very stubborn deviant fantasy. Even though its called satiation, its a form of aversion therapy, in which the goal is to take the fun out of the fantasy, through verbalization, recording on audiotape, and overuse/depletion/satiation. Its controversial, but was used when offenders could not stop becoming aroused to their victim, or could not stop involuntary/intrusive thoughts or fantasies of a victim while masturbating. The instruction is to continue masturbation to the fantasy of the victim, for 5, 10, 15, or 20 minutes, or longer, after ejaculation or after it is no longer fun. Because its not fun, it is part of the therapeutic technique to audiorecord the activity - which would be reviewed and verified by the therapist. (fun job: listening to an offender masturbate to thoughts of his victim). Therapists would routinely avoid these tasks, and offenders would read novels or newspapers to the recording instead of masturbating. Or, you could listen to these tapes while driving home after work (arriving home feeling thoroughly slimed).

A less controversial version of this therapy is called "covert conditioning." Cover-conditioning is not unlike the visualization excercises which our Olympic hopefuls use in training - they imagine every detail of the event, so vividly and so carefully that its a form of practice (that way when they get to the Olympics, they've already won the 50 yard-dash 100s of times in their mind.) Its not unlike the deviant fantasy rehearsal that predatory serial murderers and sexual homicide offenders go through. In Covert-conditioning, the offender is instructed to visualize the stages before and leading up to a reoffense against a favorite victim or target. The major difference is that this can be done in a therapy session, and when the offender first begins to feel sexually aroused, he is to cue the therapist, or himself if alone, and redirect the vivid fantasy to an aversive scene, involving the legal system, family, police, prison, courts, job, or whatever works. But they do think about reoffending, and it sometimes involves the victim - and its a form of therapy.

Another form of arousal reconditioning is used for offenders who cannot become sexually aroused or sustain sexual arousal to thoughts that are not deviant or don't involve a victim. This method involves instructing to the offender to masturbate to thoughts of the victim/deviant fantasy, except the offender is instructed to begin to substitute a normative fantasy shortly before ejaculation. If done carefully, this treatment uses the principles and mechanisms of classical conditioning to restructure what is arousing to the offender.

Mostly, I think these things are effective at forcing therapists to visualize the same detail of deviancy as themselves.

All of this becomes a very good reason to want to play with motorcycle carbs, and other things that are fun and can actually be fixed.

Another problem with this is that the two RQs in this example have potentially different consequences. Sexual contact with a child is a do-not-pass-Go issue, while attempting to contact Betty is a compliance issue. They have different potential consequences, unless, of course, there is a restraining order violation. But that is still different than a reoffense.

.02

------------------
"Gentlemen, you can't fight in here. This is the war room."
--(Stanley Kubrick/Peter Sellers - Dr. Strangelove, 1964)

posted 07-08-2008 03:33 PM            

Some schools are teaching that if you use the victim's name in the RQs, you must use them in the CQs. Apparently, somewhere along the way, some subject said the name evokes emotion, so this is a form of balancing. It's very common, and, as I said, mandatory for some running tests that are subject to QC with such rules.

posted 07-08-2008 04:37 PM            

I am along the same lines as Barry. I don't like using a name but if I have to in one Q I will use it in the Q as well.

posted 07-08-2008 04:43 PM            

I am a little hesitant to try to estimate signal value between 2 questions that could evoke a response to a relevant/target issue. All we can really do is guess because the examinee actually determines which question has the most "signal value". I don't think we can predict it and that is why we seperate CQ's and RQ's by either subject or time. Isn't this the main reason we don't make split calls on multple issue or multi-facet tests?

posted 07-08-2008 04:46 PM            

------------------
Ex scientia veritas

posted 07-08-2008 07:16 PM            

quote:

---

I don't think we can predict it and that is why we seperate CQ's and RQ's by either subject or time.

---

We don't all do that. The Reid technique works, and they don't make those separations.

Think about it though:

- only thought about it = memory of "bad" thoughts (which may even be normal had he never crossed the line and offended) + "concern" over discovery, consequences, whatever, etc.

- did it = memory of thoughts + memory of act + "concern" over discovery, consequences, whatever, etc.

We do the same with any CQ. If you're not careful, you'll be making the Lykken argument in full next.

posted 07-08-2008 07:44 PM            

2. Proximas--This is a masturbatory fantasy of a person that we were near at one time or another, but was never actually sexual with. Neighbors, friend's wives, or that shapely person who is the cashier at your store---or more common--the "one that got away."

3. Intangible---This is a fantasy of a person who does not exist in the near 3-D world or an object that is not human or is inanimate. Porn actors, pictured sex parts,sex dolls (yes--some people fantasize about statues and other "things"), animals,---and in women--the non-existant "knight in shining armor" who has no identity.

Most men and women have a mixture of all three in thier masturbation matrix. Although a clinician such as Ray has expressed to me a fear of a lack of scientific gravity (data) when I used such---my therapists LOVED it. Repeat--LOVED it. You know an offender who say, masturbates 5 times a week is not being truthful when they have a thin and none robust Masturbation Matrix---unless of course they are absolutely fixated on one fantasy and one fantasy type alone. This would be no less of an attention-getter than a robust or fascinating matrix---either way, to use my matrix is to get valuable info. The trick is to sell it without shaming the offender---to even make it interesting and normal to discuss.

posted 07-08-2008 08:01 PM            

Then after collecting the data over 15 minutes or so---use the question (if you are brave or wreckless) "Have you lied to me today about your masturbation habits." Regardless of the outcome of the test, they will always "remember" some fantasies at either the 11th hour or in the postest.

Think of your own fantasies and determine what types they fall into. Sure there are some ones---such as a local TV anchorwoman---would she be Proxy or Intangible? If the offender has ever seen her in person, than I would say Proxy. [I use that example because we caught a stalker of a TV personality from my mast matrix---child molester who was on parole---became infatuated with a small town local newsgirl.]

posted 07-08-2008 08:13 PM            

I can't help but wonder how some of the examiners who grade PCSOT examiners for APA whether or not they consider it to be a good question under the circumstances given.

------------------
Ex scientia veritas

posted 07-08-2008 08:20 PM            

quote:

---

If the examinee could truthfully give the same answer both questions and still be guilty of the offense OR if he could truthfully give the same answer to both questions and be innocent then the test is defective because it lacks the dichotomy required by a Probable Lie Comparison Question Test.

---

posted 07-08-2008 08:23 PM            

CQ: In your entire life have you ever stolen anything?

RQ: Did you steal any part of that missing money?

Good, bad, or otherwise?

posted 07-08-2008 09:33 PM            

But he is just as likely to react to your CQ "In your entire life have you ever stolen anything?" because the theft of the $500.00 that is the subject of your relevant question would in fact ALSO be a theft that ccurred in his lifetime.

You have no separation between the subject matter (THEFT) contained in the RQ and the subject matter also (THEFT) in your CQ because both questions concern themselves with a theft that occurred within the examiees lifetime.

Unless you have a time bar or a situational bar you are basically asking the same question twice for the guilty party.

Arguably an innocent examinee should react only to the CQ,and not the RQ but what keeps your guilty examinee from reacting to your CQ? Unless you can explain why a guilty person would be expected to fail to react to your CQ or at least have a less significant reaction to the CQ than the RQ,when he is in fact LYING to a question that isn't separated in some fashion from the relevant issue you don't have a valid test.

[Excerpted fro Forensic Psychophysiology Using The Polygraph]
"In Fact Non-Exclusive control questions may not only contain the crime for which the deceptive (guilty) examinee is being tested, but also other similar crimes he
may have recently committed, rendering the Non-Exclusive control question a ignificantly
stronger threat to the deceptive (guilty) examinee than the relevant question against which it is being compared for a determination, thus a formula for false negatives."

That returns us to signal value and psycholical set which are both determined within the mind of the examinee. If you have found a way to quantify "signal value" I'd love to hear it. If our examinee reacted to your CQ how would you defend an NDI call?

------------------
Ex scientia veritas

posted 07-08-2008 10:36 PM            

It is my opinion, based on both the research that supports non-exclusive questions are superior to exclusive and my field experience in using them, your example questions will work.

ebvan,

All too often we as examiners look for excuses as to why an examination failed (e.g. bad examiner, poor questions, stars were not properly aligned). Sometimes there are notable deficiencies in the examination or test but sometimes the answer is simply that there is no test that is perfect.

Due to the aforementioned and in vein, we as examiners throw away a certain format due to anecdotal experiences and move to others in a quest for perfection (when there really is not such a creature).

The bottom line is, if we do the best with that which is proven through research and conduct a proper examination and test, we will be correct considerably more times than not and make better decisions than if we had not done so (i.e. just interviewed the person and gone on instinct).

posted 07-09-2008 12:48 AM            

For the record. I like stat's masturbation matrix. Only I don't think its a matrix. Matrix implies form of structural organization that does not seem to apply. It is really a form of taxonomy for understanding masturbation fantasy. I also know that a greater portion of females will report masturbating to thoughts of nothing except masturbation, compared with males who tend to report more vivid thoughts of persons, body parts or features, and behaviors. Some offenders will admit to masturbating when they are bored, stressed, angry or frustrated. Others will admit to masturbating when they cannot sleep. This is all worth asking about, because the clinical concern is the tendency to utilize or overutilize sexual behavior to remediate non-sexual needs. stat has created a template for organizing our inquiry and gather of this information. Its not a matrix, its a form of structured interview. Its not unscientific. Better structured interview protocols will improve our science. Structured interviews are always interesting to information junkies. But it is unwise to call it something that it is not (a matrix), just because we like the sound of it.

I don't like "did you lie to me..." questions. Its just bad testing form and bad science. In testing sciences the test stimulus are the test stimulus. The test administrator is the test administrator. Using the test administrator as the test stimulus is inviting a lot of problems with unreliable outcomes, and a lot of excuses such as "the examiner scared me and I was reacting to him not the question." The stimulus is the stimulus. The examiner is not the test stimulus. The examiner merely administers the test and test stimulus. This is a fundamental principle of testing - all forms of testing. You can't give a kid a spelling test, and get a reliable result while you are either feeding him the answers or intimidating him. Test administration is neutral. Of course, with CQs, there is a certain amount of maneurvering, but with the RQs we must ensure that nothing other than the stimulus itself causes a reaction. That means the presentation must be neutral, and all of the dimensions of the stimulus are contained in the stimulus itself. The examiner should not become a dimension of the test stimulus. For an example of the problem: one of my favorite IQ tests is the Slosson (verbal only), which is an administered screening test, with six sub-scales. It is quick and dirty, and clinically rich. Properly used, it can provide a wealth of information in about 20 to 25 minutes. It also has stratified age norms for persons age 8 to 64. The Slosson is a question and answer test, in which the examiner provides no feedback to the test subject. This is odd, and takes a little practice and getting used to. Some subjects will feel like they are doing badly when they are not, simply because they are not getting any cheerleading. Other subjects will not notice they are doing badly because they are not told so. Many will give up more readily with negative feedback, and that would deplete there performance and scores. Most will try harder if they are coached, and most will also attempt to read the evaluator for cues. What we really want to know is how the test subject reacts natively to the stimulus, not the evaluators coaching or discouraging feedback.

Good polygraph science has to be based on the same principles as other testing sciences.

I still don't like the victim's name in the CQ. I haven't done it. Of course, I'm not the final authority on this, or anything else. But if the investigation target issue is the "offending against Judy" then "offending against Judy" is off limits for the CQs.

RQs are specific to a behaviorally descriptive allegation and specific time and location (event specific polygraph), or specific to a possible behaviorally descriptive concern that would signal a breach of behavioral compliance or escalating level of threat (screening/monitoring polygraph). RQs should describe an issue of conflict. That is, a behavior that someone has a concern about, while the subject would like to be exonerated.

QCs are broad, ambiguous, probably lie questions that challenge the limits of the subject's integrity and indirectly imply the subject as capable or culpable for the behavior described by the investigation targets.

For the rest of this, its fun, but its important to recognize that: 1) its very easy to get into too much psychologizing, and 2) its tempting to overly rely on linguistic logic in attempt to solve construct validity problems.

Don't forget to keep it simple.

evan wrote:

quote:

---

Arguably an innocent examinee should react only to the CQ,and not the RQ but what keeps your guilty examinee from reacting to your CQ?

---

Well, yes, this is highly arguable. There is nothing about what we actually know about the human brain or human psychology that suggests that an innocent person should react only to the CQ. In fact, it should not be surprising to see some form of orienting reaction, or even defensive reaction, to the RQs also - from an innocent person. It is presently impossible to tell the difference between orienting reactions and defensive reactions by looking at individual reaction segments. The only way can infer that a reaction is due to involvement in the behavioral issue of concern is not by the absence of reaction to one type of question (they will show some reaction to both), but by the aggregate difference in magnitude of reaction to several presentations (under similar conditions - because that is how good testing science works) of several CQs and RQs.

quote:

---

Unless you can explain why a guilty person would be expected to fail to react to your CQ

---

I'm not sure why this would need to be explained.

quote:

---

or at least have a less significant reaction to the CQ than the RQ,when he is in fact LYING to a question that isn't separated in some fashion from the relevant issue you don't have a valid test.

---

This is better. "separated in some fashion" is important - discussion and disagreement often seem to pertain to our (mostly untested) hypotheses about how to ensure that separation.

quote:

---

[Excerpted fro Forensic Psychophysiology Using The Polygraph]
"In Fact Non-Exclusive control questions may not only contain the crime for which the deceptive (guilty) examinee is being tested, but also other similar crimes he
may have recently committed, rendering the Non-Exclusive control question a ignificantly
stronger threat to the deceptive (guilty) examinee than the relevant question against which it is being compared for a determination, thus a formula for false negatives."

---

Except the data do not support all of this fancy psychologizing. The data do seem to consistently support the simpler hypothesis about differences in reaction to behaviorally descriptive, and broader PLCs. Data do not seem to support the hypothesis about time-bars, and FNs without them. Simple wins again. Data also seem to support the hypothesis about differential reactivity (signal value) for RQs, and DLCs.

quote:

---

That returns us to signal value and psycholical set which are both determined within the mind of the examinee. If you have found a way to quantify "signal value" I'd love to hear it. If our examinee reacted to your CQ how would you defend an NDI call?

---

Which is essentially another caution about psychologizing and sticking to basic polygraph.

In actuality, we do have a method for quantifying differences in signal value. We do this by evaluating differential reactivity among a plurality of RQs and CQs, after presenting each of the stimulus several times, in the same manner. Then we use inferential statistics and probability theory to evaluate whether that difference is significant, and whether our observed score is a poor fit or good fit with our known models for truthful or deceptive persons.

I'll go further and suggest that if we are still using test methods or scoring methods for which we cannot determine and report the level of statistical significance or p-value, then we, as a field of science, have not been doing enough homework and study.

.02

r

------------------
"Gentlemen, you can't fight in here. This is the war room."
--(Stanley Kubrick/Peter Sellers - Dr. Strangelove, 1964)

posted 07-09-2008 03:01 AM            

------------------
Ex scientia veritas

posted 07-09-2008 08:36 AM            

Horvath, F. S. (1988). The utility of control questions and the effects of two control question types in field polygraph techniques. Journal of Police Science and Administration, 16, 198-209.

Horvath, F. S. (1991). The Utility Of Control Questions And The Effects Of Two Control Question Types In Field Polygraph Techniques. Polygraph, 20, 7-25.

Krapohl, D., Stern, B. & Ryan, A. (2003). Exclusionary vs. nonexclusionary: A review of the evidence. Polygraph, 32 (4), 245-250.

Palmatier, J. J. (1991). Analysis of two variations of control question polygraph testing utilizing exclusive and nonexclusive controls. Masters Abstracts, 30, 380.

r

------------------
"Gentlemen, you can't fight in here. This is the war room."
--(Stanley Kubrick/Peter Sellers - Dr. Strangelove, 1964)

posted 07-09-2008 08:55 AM            

MREs are nourishing and work to prevent starvation but most grunts won't eat them if anything better is available.

Going back to the original question concerning whether or not, the CQ was a good one, If our examinee' strongest reaction was to this CQ that covered essentially the same activity and time period as your RQ, how would you explain and defend an NDI call to someone who raises the activity and time period argument?

------------------
Ex scientia veritas

posted 07-09-2008 09:05 AM            

quote:

---

Going back to the original question concerning whether or not, the CQ was a good one, If our examinee' strongest reaction was to this CQ that covered essentially the same activity and time period as your RQ, how would you explain and defend an NDI call to someone who raises the activity and time period argument?

---

Exactly. Except that the part about time-period is moot.

r

------------------
"Gentlemen, you can't fight in here. This is the war room."
--(Stanley Kubrick/Peter Sellers - Dr. Strangelove, 1964)

posted 07-10-2008 10:57 AM            

I was being to harsh and concrete when I criticized the word "matrix." In its generic form it simply means "substance" or "mixture." So, I'll concede that it is not wrong.

It is a good idea to excavate offenders' masturbatory fantasy maxtrix, or take inventory of it, if you like it that way. "Inventory" strikes me as an odd word here, but I get it.

Eric has created a typological or taxonomical framework for interviewing the range and types of masturbatory fantasies which offenders employ.

The thing to be careful about is in attempting to interpret the meaning of the information - that is a clinical matter. Masturbation issues also include large components of value judgment, and shame. Mostly though, we really don't know much about the masturbatory behaviors of non-offenders - so making inferences about this information is difficult and best left to the clinicians and evaluators. Our role, as polygraph examiners, is to gather the information and give it to the professionals whose job it is to know what to do with it.

I've used some of these questions, and will continue to do so from time to time. There is a point at which our effectiveness at doing this becomes enabling for avoidant therapists who don't themselves ask enough intrusive questions about masturbatory behaviors. I do not believe it serves the treatment fields needs to enable that. Basically, if talking with me, or stat, or Donna, at the time of a PCSOT polygraph is the first time the offender is asked to discuss this stuff, then one of two conditions exist: 1) the information is irrelevant to the treatment and supervision process, and only services polygraph voyeuristic or busy-ness needs, or 2) someone else ain't doin' their job ahead of the polygraph.

With all that said, it has been my experience that masturbation question are very difficult questions for PSCOT test subjects (offenders). First, it is very normal to underreport and keep secrets about masturbatory behaviors. Second, therapists get weary of the slime pit, and tend to become avoidant of continuously asking intrusive questions. Third, offenders are very good at creating other forms of present chaos and topical problems to soak up therapy time and distract from the offense-specific-ness of there therapy.

Polygraph can help with this problem by leveraging masturbation problems back into the clinical radar scope, and back into the therapeutic conversation. We do this by getting the offender to role over during the pretest that they have been underreporting some specific kinds of problems, testing their denial, and by sometimes testing the limits of their admissions.

Be ware that testing the limits of admissions is not well respected by ATSA folks - despite the fact that we polygraph examiners get "gold-nugget" experiences from them, in the form of additional admissions after failing the questions when we attempt to test the limits of their admissions.

In practical reality, the test results are meaningless when we attempt to test the limits of these admissions (i.e., how many times, besides what you told me, are you witholding any information regarding..., or worse - have you lied to me today...). Think about it: if the offender passes, do you really believe you know EVERYTHING? Doubtful. There is still more. Then: if he fails: what do you know now that you didn't already know before he failed the question? Nothing. You already knew that you don't, and won't ever, know everything.

For the offender who's taken 10, 20, or 30 polygraphs - he already knows he's not going to prison for failing - so long as he admits something. He also already knows where he can make additional admissions without getting his arse kicked too awful hard. He also knows the numbers game that polygraph examiners play. We polygraph examiners have to be professional enough to refrain from being so so eager for gold-nuggets, in the form of additional admissions to affirm our own competence and worth. We have to pause and think through the tendency to assume that we have accomplished something on behalf of the risk assessment process or risk management plan when get a "confession" to an issue for which we should assume we will never know everything. The messages from ATSA seem to be that they want to regard these additional admissions as not all that additive. In part, they just don't like the polygraph - perhaps because it challenges their own god-complexes, but that's another matter.

Polygraph examiners can undeniably help the risk management and treatment processes by leveraging the initial admissions and topics into the conversation, when the offender denies any involvement in the behavior of concern. Testing the limits of admissions is specious. It would be more efficient to just take their admissions and multiply them by 2 or 10 - then advise the risk evaluators and risk managers to regard that as the size of the problem, with the understanding that we will never know everything. Then continue to test the issues of denial - secret girlfriends, new victims, pornography, accountability in the community, boundaries or contact with minors etc.

For example, an offender just this week admitted to searching for and masturbating with his adult step-son's pornographic magazines and movies. His treatment program does not inherently prohibit or pathologize masturbation activity, but he has had trouble with masturbatory thoughts of a former step-daughter/victim, and related polygraph questions, throughout 8 years of therapy and 20-some polygraphs. So, he attempts to refrain from all masturbation, in order to pass polygraphs. He has failed a couple of exams regarding pornography - duh - and has now admitted to withholding information about masturbating with those materials on 2 occasions. He was previously denied any pornography use, and then denied any masturbation while viewing pornography. Do we think there are not more than two incidents - try 4 or 20 (that''s better). We could test the limits of this but why bother. We know already. Of course he lies. I would rather his file include the unresolved question - that way, when he's off probation a few years and wants to petition the court for removal from the registry - it is evident to the new evaluator or prosecutor that he can barely manage his masturbatory fantasies.

I routinely ask about frequency of masturbation and fantasy material during the pretest. If they deny any masturbation, I make sure and document whether they are required or instructed to refrain from that activity - or do they explain its a personal choice. This is a natural opportunity to review and re-document any admitted history of masturbatory thoughts of children or victims, and whether that was voluntary or involuntary. Keep in mind that voluntary = belligerent, and involuntary = inept self-control.

Hopefully, a future prosecutor, judge/magistrate, and evaluator can someday get the picture that the guy can barely control himself and can barely maintain abstinance from deviancy - before removing him from the registry, just because he completed treatment, was discharged from probation, and earned his wizard-of-oz-empathy-certificate.

So all of this can be important: proximas, intangibles, recall, and nothingness. It may also be helpful to ask about masturbating to relieve stress, boredom, frustration or sleep problems. The labels themselves don't really matter - it's just a structural framework for gathering and organizing the information. It is the information that matters - possibly in a narrative sense. It is a clinician's job to determine the meaning, including deviancy and normal-ness of the information and behavior. Some people are a lot more creative and energetic than others. Some are creatures of habit and routine.

I'm not feeling warm to the idea of masturbatory deviancy as a comparison issue though. There are times when masturbation has to be a relevant investigation target - including times when it is useful to the team to test the limits of the offenders admissions (simply for the opportunity to leverage the conversation).

CQs are supposed to be broad, somewhat ambiguous probably lie questions. Masturbation is a specific behavior - and it matter in the way an RQ matters.

I've recently heard that Eric Holden is now teaching a form of PCSOT in which "treatment issues" are used as CQs when the RQs involve "probation issues," and "probation issues" are used as CQs when the RQs involve "treatment issues." As I've heard it, the subject is always told he failed the CQs. If this is accurate, then it is troublesome to me. Our jurisdiction will always refer a retest when the subject fail - so we have a form of "everlasting-gobstopper" for referrals (that is: Willy Wonka's famous candy that never melts and never looses its flavor), which is prolly good for bid'ness but will become concerning to all but the most duped and dupable POs and clinicians. I also suspect that it may increase cynicism on the part of offenders - which cannot possibly help them to be safer (not “safe”) people.

I could be wrong about this. If so, someone please straighten me out.

I agree with Eric Holden and others that the RQs are best if they imply similar potential consequences. Holden's time-of-reference/frame-of-reference is also interesting, but time-of-reference really requires proof - in consideration of what we know about time-bars. So, the frame-of-reference construct is perhaps more important. But there is the assumption that probation issues and treatment issues are different. In some jurisdictions they might be, in others they might not. This construct requires some form of factor analysis which has not been done. In our jurisdiction, probation and treatment issues are virtually equivalent - boundaries and behavioral requirements, accountability requirements, incidental reporting requirements. Of course, laws are laws, and new offenses present somewhat different potential consequences than non-compliance. As I've heard Dan Sosnowski say on more than one occasion: probation issues are treatment issues, and treatment issues are probation issues. Holden's "situational relevant question" (i.e., did you lie to me today) is an indefensible fabrication which he offered in 2000, referencing his own 1971 publication in the Psychophysiology Journal. I've contacted him for that publication, with no response. I contacted the Psychophysiology Journal, and they have no Holden & Uno (1971) reference or citation in their archives or table of contents. There is no page number on the citation in the 2000 publication. There is a collection of abstracts from the Psychophysiolgy conference, in which Holden & Uno presented data on prison inmates EDA responses to 20 minutes of differing film content - which has little, if anything except the use of an EDA, to do with "did you lie to me today..." during a polygraph test. I've already ranted about the problems these questions present in terms of good basic testing principles, and why it is wrong to turn the test administrator into the test stimulus. There are field practice implications as well (more later).

So, we're back to good-ole' basic polygraph, with RQs and CQs.

RQs are specific behaviors, which the offender might deny or minimize, and which hold useful signal value for professionals involved in risk assessment, risk management, and treatment activities.

CQs are broader, somewhat ambiguous, probably lie questions. If it is a signal issue/behavior for risk evaluators, or risk managers, then it is an RQ, not a CQ. QCs create doubt within the subject, and challenge the subject's integrity and confidence in general. CQs give the truthful person a place to react, and provide a comparison measurement with which to evaluate the significance of reactions to the RQs. In signal detection theory, CQs are the background field of noise, against which we evaluate whether the RQs produce significantly different (read: statistically significant) measurement values. If CQ reactions occur in response to specific behavioral concerns (masturbatory fantasy/deviancy) and must be sometimes evaluated for significance themselves, then they are NOT CQs.

If a question describes a specific behavior that is of concern to the referring agent's decision process, then it is an RQ. CQs do not become relevant simply because of the scope of some verbal Venn logic – and its silly to say they “become relevant,” but I've heard it said that way several times. Similarly, if an RQ does not describe a specific behavior, if it is a broad concern, with value-laden implications that are of no real decision-support value to the referring agent, then is is a CQs. (So, one could argue that those famous questions: “are you withholding any information about your involvement with ...” might be a good CQ. Who knows. Hiliard (1979) reported that the overall truth question (2T) worked better as a control question than in its originally intended usage.

.02

r

------------------
"Gentlemen, you can't fight in here. This is the war room."
--(Stanley Kubrick/Peter Sellers - Dr. Strangelove, 1964)

posted 07-10-2008 01:49 PM            

In Texas (Baja Oklahoma) as I understand their structure probabtion issuse are quite separate from treatment issues, only converging at the commission of a new crime or entering a NO LURK zone that might be spelled out at sentencing.

------------------
Ex scientia veritas