- Welcome to AntiPolygraph.org Message Board.
A forum for free and open discussion of polygraph-related topics
Posted by Marty
- Aug 03, 2003, 04:47 AMQuote from: Public Servant on Aug 01, 2003, 12:28 PMMarty,
Great info you provided on GKT analysis. However, statistical analysis and error rate were not part of my assertion here. I do not refute the advantages of a well planned and executed GKT, nor the promise of CNS methodology advocated by Drew.
My point was merely that GKT depends on the same physiological response as CQT, POT, R/I, etc. For this physiological response to be a valid indicator of guilty knowledge, there must be a psychological process initiating the response to given stimuli. Thus GKT also relies on the existence of a lie response, deception response, fear of detection...whatever you want to call it. This contrasts with the argument (often asserted by George) that no such response exists (or is it that this response can not be proven to exist beyond a doubt in every case, so we must assume it does not exist).
The point is: GKT is based on the same psychology-physiology connection, so for this phenomenon to exist for GKT, it has to exist for other formats as well. What is monitored is not different, only the type and format of stimuli is changed.
.....
Public Servant,
I certainly don't believe there is no response associated with lies, fear of consequences, etc. In fact, I think the GKT itself can be used to better establish this. There are issues with controls and it is remarkably easy for well meaning people (and I do believe the vast majority of polygraphers are well meaning) to insert bias but the GKT lends itself to a properly done, double blind, test protocol. So why isn't this happening?
BTW, just got Ian Rowland's "Full Facts of Cold Reading", which illuminates the ways of psychics, palmreaders, etc. Much of what he discusses has application to both interrogation and establishing "setting" in compliance situations
Fascinating insight about an active group of fraudsters as well as vulnerabilities we all share.-Marty
Posted by Mark Mallah
- Aug 01, 2003, 07:33 PMQuoteOn another thread, Mark Mallah conceded that a blind stim led him to believe there is a physiological response to a lie.
It wasn't the blind stim test that led me to believe it. I already believed that a lie would produce, as you said, cognitive dissonance. I knew from personal experience that if I lied, my physiology would change.
The blind stim test convinced me of nothing, other than the examiner was missing the point (sure he could recognize a lie, but could he recognize the truth?)
This won't affect your above point, Public Servant, but I just wanted to set the record straight.
Posted by Public Servant
- Aug 01, 2003, 12:28 PMMarty,
Great info you provided on GKT analysis. However, statistical analysis and error rate were not part of my assertion here. I do not refute the advantages of a well planned and executed GKT, nor the promise of CNS methodology advocated by Drew.
My point was merely that GKT depends on the same physiological response as CQT, POT, R/I, etc. For this physiological response to be a valid indicator of guilty knowledge, there must be a psychological process initiating the response to given stimuli. Thus GKT also relies on the existence of a lie response, deception response, fear of detection...whatever you want to call it. This contrasts with the argument (often asserted by George) that no such response exists (or is it that this response can not be proven to exist beyond a doubt in every case, so we must assume it does not exist).
The point is: GKT is based on the same psychology-physiology connection, so for this phenomenon to exist for GKT, it has to exist for other formats as well. What is monitored is not different, only the type and format of stimuli is changed.
On another thread, Mark Mallah conceded that a blind stim led him to believe there is a physiological response to a lie. However, he said other psychological phenomena could also cause such physiological response. I conceded on that issue but said I believed these other phenomena rarely occurred consitently on the same question for a series of charts. But I did also concede that any false positives would be as a result of such phenomena as he cited.
Now, since GKT stimuli would likely be more benign than a CQT relevant question, the exchange I had with Mark, would support assertions of a lower false positive rate for GKT. However, it does not change the fact that a lie response / fear of detection response has to exist for GKT to be valid.
In that vein, does your research of literature on GKT compare false negative rates to CQT? The above would make one think that CQT would be the one to fair better there (though that would be of little comfort to those concerned with screening).
My understanding of "brain fingerprinting" is that memory recall patterns within the CNS are analyzed. Thus my above arguments would not be enhanced (nor hindered) by this technique.
In regard to balancing the other psychological phenomena which might cause false positive to CQT, it is important for the examiner to carefully apply the theory of psychological set to ensure any anxiety regarding relevant issues (other than actual deception) are properly re-directed. One who is truly deceptive to the issue will have no problem responding to relevant questions (especially in a criminal specific exam). Giving the innocent person something else on which to focus their anxiety is the main purpose of a good pre-test. I know it succeeds at a high rate to produce corroborated NDIs when relevant questions are about felony offenses. Why wouldn't it succeed at an equally high rate for exams with lesser consequences?
Great info you provided on GKT analysis. However, statistical analysis and error rate were not part of my assertion here. I do not refute the advantages of a well planned and executed GKT, nor the promise of CNS methodology advocated by Drew.
My point was merely that GKT depends on the same physiological response as CQT, POT, R/I, etc. For this physiological response to be a valid indicator of guilty knowledge, there must be a psychological process initiating the response to given stimuli. Thus GKT also relies on the existence of a lie response, deception response, fear of detection...whatever you want to call it. This contrasts with the argument (often asserted by George) that no such response exists (or is it that this response can not be proven to exist beyond a doubt in every case, so we must assume it does not exist).
The point is: GKT is based on the same psychology-physiology connection, so for this phenomenon to exist for GKT, it has to exist for other formats as well. What is monitored is not different, only the type and format of stimuli is changed.
On another thread, Mark Mallah conceded that a blind stim led him to believe there is a physiological response to a lie. However, he said other psychological phenomena could also cause such physiological response. I conceded on that issue but said I believed these other phenomena rarely occurred consitently on the same question for a series of charts. But I did also concede that any false positives would be as a result of such phenomena as he cited.
Now, since GKT stimuli would likely be more benign than a CQT relevant question, the exchange I had with Mark, would support assertions of a lower false positive rate for GKT. However, it does not change the fact that a lie response / fear of detection response has to exist for GKT to be valid.
In that vein, does your research of literature on GKT compare false negative rates to CQT? The above would make one think that CQT would be the one to fair better there (though that would be of little comfort to those concerned with screening).
My understanding of "brain fingerprinting" is that memory recall patterns within the CNS are analyzed. Thus my above arguments would not be enhanced (nor hindered) by this technique.
In regard to balancing the other psychological phenomena which might cause false positive to CQT, it is important for the examiner to carefully apply the theory of psychological set to ensure any anxiety regarding relevant issues (other than actual deception) are properly re-directed. One who is truly deceptive to the issue will have no problem responding to relevant questions (especially in a criminal specific exam). Giving the innocent person something else on which to focus their anxiety is the main purpose of a good pre-test. I know it succeeds at a high rate to produce corroborated NDIs when relevant questions are about felony offenses. Why wouldn't it succeed at an equally high rate for exams with lesser consequences?
Posted by Marty
- Jul 30, 2003, 02:01 PMPublic Servant,
The advantages of a GKT are well known and I'm surprised, Public Servant, you are not up on them. A principal advantage is that it is subject to statistical analysis. Of course this assumes that proper controls are established to mitigate examiner bias but this is easily done for GKTs. Especially appealing is that the polygraph in GKT mode has the potential to be more easily approved for evidentiary purposes. This is also one of the appealing characteristics of Drew's "brain fingerprinting." As for the CQT in screening, most of the polygraph community concurs that CQT has a higher false positive rate for SI CQTs, let alone what can be achieved with GKTs. My problem is the "P" part of the PLT, often polygraphers interpret the "P" as "A", ie: Assumed. It produces an undeniable bias against the exmainees that are truthful on the controls.
http://www.nationalpolygraphconsultants.com/gkt.htm
Here's a key paragraph:
The real beauty of the GKT, in addition to its simplicity, is that the test permits the user to calculate the precise probability of a false positive error. The capability to calculate the error of a test is a factor many jurisdictions find important in forensic testing. One method of error estimate for a version of the CQT has shown promise (Krapohl & McManus, 1999), however at this writing the GKT has the larger body of supporting data. Probability tables for GKT scores are found in Dr. Lykken's recent book (1998, pp 290, 292).
BTW, for those interested in the way CQT polys are scored, there is a Java script that can be looked at on one of the other pages on this site. It is consistant with the description in Kleiner.
-Marty
The advantages of a GKT are well known and I'm surprised, Public Servant, you are not up on them. A principal advantage is that it is subject to statistical analysis. Of course this assumes that proper controls are established to mitigate examiner bias but this is easily done for GKTs. Especially appealing is that the polygraph in GKT mode has the potential to be more easily approved for evidentiary purposes. This is also one of the appealing characteristics of Drew's "brain fingerprinting." As for the CQT in screening, most of the polygraph community concurs that CQT has a higher false positive rate for SI CQTs, let alone what can be achieved with GKTs. My problem is the "P" part of the PLT, often polygraphers interpret the "P" as "A", ie: Assumed. It produces an undeniable bias against the exmainees that are truthful on the controls.
http://www.nationalpolygraphconsultants.com/gkt.htm
Here's a key paragraph:
The real beauty of the GKT, in addition to its simplicity, is that the test permits the user to calculate the precise probability of a false positive error. The capability to calculate the error of a test is a factor many jurisdictions find important in forensic testing. One method of error estimate for a version of the CQT has shown promise (Krapohl & McManus, 1999), however at this writing the GKT has the larger body of supporting data. Probability tables for GKT scores are found in Dr. Lykken's recent book (1998, pp 290, 292).
BTW, for those interested in the way CQT polys are scored, there is a Java script that can be looked at on one of the other pages on this site. It is consistant with the description in Kleiner.
-Marty
Posted by Public Servant
- Jul 30, 2003, 11:58 AMMany anti-poly posters seem to like (prefer to other techniques?) the GKT. However, many anti-poly contributors use the argument, when discussing polygraph in general, that response to a question could be due to many emotional/psychological phenomena other than deception or lying (sounds redundant but I think we found a difference when we discussed stim tests on a previous thread).
The point I am making is not an attempt to discredit GKT. The argument that multiple (perhaps random?!) phenomena can cause a consitent response to a particular question(s) would have to apply to GKT as well. Yet the same posters often assert that if polygraphers were more responsible, they'd use GKT. I'd argue that the credibility given to GKT negates this anti-poly assertion of random/irrelevant sources of physiological response, applied to other formats.
So, is there a belief that GKT is somehow immune to these other phenomena, though it is montoring the same physiological activity? Perhaps it is favored here because it is "nicer" to an examinee than CQT and takes much forethought and planning from the beginning of the investigation. Or perhaps it's because GKT's application to screening is highly limited at best.
Game on.
The point I am making is not an attempt to discredit GKT. The argument that multiple (perhaps random?!) phenomena can cause a consitent response to a particular question(s) would have to apply to GKT as well. Yet the same posters often assert that if polygraphers were more responsible, they'd use GKT. I'd argue that the credibility given to GKT negates this anti-poly assertion of random/irrelevant sources of physiological response, applied to other formats.
So, is there a belief that GKT is somehow immune to these other phenomena, though it is montoring the same physiological activity? Perhaps it is favored here because it is "nicer" to an examinee than CQT and takes much forethought and planning from the beginning of the investigation. Or perhaps it's because GKT's application to screening is highly limited at best.
Game on.