|
Author
|
Topic: gkt use
|
lielabs
Moderator
|
posted 05-03-2005 01:54 AM
  
Does anyone agree that you can use a gkt or pot when the investigators do not know what the murder weapon was for sure but think it was a spanner. They then accuse the subject of using the spanner and then run a gkt later to see if he knew what the weapon used was. This seems absolutely ridiculous to me but Frank Horvath said you can because the investigators did not know for sure it was a spanner. Despite the fact that the subject was accused of using it went to trial was convicted of the murder and then was given a gkt to see if he used the spanner. He did not respond to the spanner on the gkt. Does that mean he is innocent like Frank Horvath is saying even though running a test under these circumstances breaches every guideline about gkt I am aware of. Iacono stated that meant he was incapable of responding to significant items on a gkt and therefore any test on this guy would be worthless. Any thoughts. IP: Logged |
Barry C
Member
|
posted 05-03-2005 07:46 AM
Do you have Kleiner's book? The chapter of the CKT / GKT use in Japan addresses your question.I had always thought (as it just made sense to me) that even the innocent would react to an obvious hey item, but they have not had that problem in Japan. I don't think they discuss a person who has been accused of using a particular item, but I don't recall well enough to say. I would think (and I was wrong as I just said about the similar issue) that accusing a person would direct his attention to the item and make it a worhtless test. It would seem to make more sense to run the SPOT first and interrogate later. IP: Logged |
Ted Todd
Member
|
posted 05-03-2005 08:44 AM
lielabs,Cleve Backster has a "Probing Peak of Tension Test" which is used when the "key" item is not known. It may fit your needs. Ted IP: Logged |
Eric Fiander
Member
|
posted 05-03-2005 08:44 AM
I have always used the GKT with a "False Key", and it seems to work for me. There are of course limited situations in which a GKT can be used, but I find the false key very effective.IP: Logged |
J.B. McCloughan
Member
|
posted 05-03-2005 11:00 AM
In a true GKT, referred more readily to now as a Concealed Information Test(CIT), there must be a known item in each list. You cannot score hits to the key if it is not known.A Searching Peak Of Tension (SPOT), which is not a CIT, is a list of probable items where one is trying to discern a subject's greatest attention but the keys are unkown. If the subject gives their strongest response to a specific item over repeated tests of random item placement, one can determine that the item in the SPOT has significance to the subject. A subject might not respond significantly to any of the items. This could mean a number of things to include that none of the items were correct or they are an unknowledgeable subject to that area of questioning. I would need more information, to see the tests, and to know whether or not any of the items used as keys in the various tests were confirmed information. You may email me if you'd like. IP: Logged |
Barry C
Member
|
posted 05-03-2005 02:08 PM
In Japan, they do use searching CITs, and they never include a catch-all "anything I haven't mentioned" type of phrase.One difference between a SPOT and a searching CIT is that in a SPOT (and POT) the critical item is usually in the middle. In a CIT, the key / critical item is placed in a random position. IP: Logged |
J.B. McCloughan
Member
|
posted 05-03-2005 02:43 PM
Here is the text from Kleiner's book 'The Handbook of Polygraph Testing' on this issue. It was sent to me through email by Makoto Nakayarna and is used for reference only on my CIT research:CHAPTER 2 PRACTICAL USE OF THE CONCEALED INFORMATION TEST FOR CRIMINAL INVESTIGATION IN JAPAN Makoto Nakayarna (In Murray Kleiner (Ed.)The handbook of polygraph testing, Academic Press (London) SEARCHING PEAK OF TENSION TEST (SPOT) AND CQT IN JAPAN Many psychologists accept that CIT is the most scientifically effective method for the detection of deception (Iacono and Lykken, 1997), but the CIT is not always applicable to the event in field situations. When we cannot find appropriate critical questions at the crime scene, SPOT and CQT are also used in Japan.
IP: Logged |
Barry C
Member
|
posted 05-03-2005 04:45 PM
A searching POT and a searching CIT are the same test, in essence. The difference is, in a searching test of any type, the critical item can be in any place. In a POT or a SPOT the critical item, or more appropriately, the educated guess (the spanner above) would be placed in the center of the list. In a searching CIT the order is random (as it is in a true unknown item SPOT).Does anybody recall where searching CITs are discussed? Apparently it's not the chapter in Kleiner as I thought it was. And now that I think of it, I think only the EDA channel was looked at in the article I'm thinking of - unlike in Japan where they evaluate all channels. IP: Logged |
lielabs
Moderator
|
posted 05-04-2005 04:49 AM
Are you saying that you can accuse the subject of using an item (spanner)and then test him with a searching gkt/cit and the results will still have value.I was always taught that an innocent subject would respond to the key if they knew details of the crime from news releases etc or were informed of the investigators suspicions. You would think in a list of alternatives the item the subject is accused of using would be the most salient if innocent and therefore the results rendered useless. If the subject is informed of details of the crime via interrogation they then have knowledge so how can they be tested reliably. Paul W. IP: Logged |
Barry C
Member
|
posted 05-04-2005 08:17 AM
No Paul. I agree with you. What the Japanese examiners found (I think it was the Japanese anyhow) is what we would think is an obvious item anybody would respond to isn't always the case. For example, I would expect everybody to react to a gun or knife as a weapon in a murder case, which would make using those items meaningless in a CIT / POT, but that's apparently not the case according to the Japanese (?). Innocent people don't necessarily respond to what we would think is the most obvious choice.Giving away the item would preclude conducting a good CIT / POT because to both the innocent and guilty, the information is known. In the spanner case above, I don't know the details, so it's possible it's a good item (but I would doubt it). If the investigators accused the suspect point blank of using it, I would say it's a worthless test. If it was one object of many, and they didn't over do it, maybe, it's good. IP: Logged |
lielabs
Moderator
|
posted 05-04-2005 11:50 PM
Barry,The item was used point blank by the investigators and in his subsequent trial. This was used and attempted to be submitted as evidence for his appeal. It was rejected by the appeals court based on the comments of Iacono. I agreed with his comments tottally in this case because of the fact he had been accused several times of using the spanner by investigators and during his trial. Horvath argued that because the investigators did not know for sure it was a spanner then the test would still be ok. I have been trying to find something other than his opinion that supports that position. I can't find anything only support to say that would be an unreliable test. What you are saying about what the japanese have discovered is different from the above case. I wanted to see what other examiners thought. I have been arguing against the inappropriate application of this test. The trouble I am having is that because Horvath said it should be ok it must be. I see from my own research on the subject that any gkt,cit,pot applied in a case like this would be worthless and Horvath made an error here. He offered an opinion with out any factual evidence to support his position. The courts rejected the test which has set a precidence here even though the test should never gave been submitted as evidence because of its improper application. The chance it would have been accepted here was small anyway but given the circumstances it had no chance regardless. I tend to agree with the scientific literature on the subject not one mans opinion which in this case was wrong. IP: Logged |
Barry C
Member
|
posted 05-05-2005 08:14 AM
Have you talked with Dr. Horvath since? That sounds crazy. Dick Arther uses a false key in which he casually mentions a particular item to distract the innocent, and he's had very good luck with that. To outright accuse the subject and then throw the item in guarantees (virtually) he'd respond to it.You might try contacting Dick Arther. He's written on it in his newsletter. IP: Logged |
J.B. McCloughan
Member
|
posted 05-05-2005 02:51 PM
Gershon Ben-Shakhar, Nurit Gronau, and Eitan Elaad conducted a recent research titled “Leakage of Relevant Information to Innocent Examinees in the GKT: An attempt to Reduce False-Positive Outcomes by Introducing Target Stimuli”. The results were similar to those of previously done studies (Bradley et al., 1996; Bradley & Rettinger, 1992, Bradley & Warfield, 1984). In the discussion it is indicated that the informed innocent had larger responses to the key information than uninformed but smaller responses than guilty. Ben-Shakhar, Gronau, and Elaad’s study however used the Guilty Actions Test (GAT) format and was scored using both SCR and RLL. It is thought that this may be true in the CIT but I am unsure of just how a field examiner would go about setting the cutoff for a response being that of the magnitude to be considered as informed innocent or informed guilty. IP: Logged |
Barry C
Member
|
posted 05-05-2005 03:09 PM
I take it they were still the largest responses in each series? (Which doesn't help most of us as you point out.)Did they compare the guilty to the innocent informed, or did the examiners "inform" everybody alike? (I realize you can't "inform" the guilty, but I'm curious to know if talking about the item - or accusing - makes the guilty more, less or equally responsive than just running a normal test.) Do you have an e-version of the study? I'd like to read it. IP: Logged |
Barry C
Member
|
posted 05-05-2005 03:12 PM
Nevermind. I found a copy. Thanks for the info!Here's the link for anybody who wants a copy:
http://websites.mscc.huji.ac.il/zerolab/BG&EJAP99.PDF [This message has been edited by Barry C (edited 05-05-2005).] IP: Logged |
Polygraph Place Bulletin Board
