CLPEX Chat Board

In light of the McKie case, the BBC has interviewed Dr. Dror in their
program Newsnight about LPE expert error and fingerprint reliability in
general.

To view the interview, link to http://www.ecs.soton.ac.uk/~id/bbc.html

I have not been able to listen to the clip of Dr. Dror yet (no sound cards in computers at work) but the SCRO administration and some of the government ministers have taken the position that a fingerprint identification is only a matter of opinion, and different experts are entitled to different opinions. As one Laboratory Superintendent told me earlier this week, if that position is allowed to stand, it could affect not only fingerprint identifications in court the world over, it could be extended to other forensic sciences, as well.

No study has all the answers. Past and current studies lead to future studies. Studies will happen within or without the fingerprint community's support. The community needs involvement in a variety of studies. Science is about explaining what we do, how we do it, and improving our understandings and explanations. If no studies were able to generate concern within the community, why would we desire to move forward? No science currently has all the answers.

I found the closing statements in the interview fascinating. Even with the change of decisions by examiners within this study, Dr. Dror, the author of the study, still comments about and supports the reliability of fingerprint examinations. Criticisms occur on studies that are not published. I know this study has been discussed in numerous forums. I hope to see the actual study published soon. I anticipate Dr. Dror will continue researching, with or without the community's support.

In response to Pat's comments:
"I have not been able to listen to the clip of Dr. Dror yet (no sound cards in computers at work) but the SCRO administration and some of the government ministers have taken the position that a fingerprint identification is only a matter of opinion, and different experts are entitled to different opinions." As Pat and others know, one of the opposing opinions has to be wrong. Do the SCRO and the ministers have the opinion that one of the opposing opinions has to be wrong?

John Vanderkolk

Having now been able to listen to Dr. Dror's television interview, I was very happy to hear him put the contextual bias issue in perspective. Like you, John, I was gratified to hear him stress that fingerprint identification is reliable. His major point, as I understood it, was that we in fingerprint identification need to be aware of context bias and take steps to remove or minimize it.

Since he was being interviewed on a program that for over a week had nightly interviews about the McKie case, I anticipated his appearance was an attempt by the press to stir controversy. Quite the opposite happened. He supported the science as reliable in spite of the results of his study.

Some of the SCRO administration and their higher-ups in the Scottish Government have taken the position that fingerprint identification is merely opinion, and that different experts can legitimately form conflicting opinions. My position in interviews with the press is that we reach scientific conclusions and there is a "right" conclusion and a "wrong" conclusion. The word "opinion" is stated in the "Rules of Evidence" so we are stuck with it, but to confuse the issue by substituting the philosophical definition of "opinion" with the definition under which we operate -- a scientific conclusion -- is improper and does a disservice to all competent fingerprint experts of integrity worldwide. The SCRO administration's reference to "opinion" is already being used against us by defense attorneys in court and, as one laboratory director expressed to me last week, may even be extended to other forensic sciences if the SCRO gets away with discrediting the dissent in the McKie and Asbury cases by dismissing it as mere opinion.

But back to Dr. Dror -- his conclusions regarding contextual bias are valid and, as you point out John, need to lead to additional studies to understand the problem and address it. Dr. Dror deserves the respect and gratitude of the fingerprint community for helping expose contextual bias as an inherent problem in the verification process. Minimal though it is in most cases, it can be a serious problem in some and must be addressed.

Criticisms occur on studies that are not published. I know this study has been discussed in numerous forums. I hope to see the actual study published soon. I anticipate Dr. Dror will continue researching, with or without the community's support.

John, (and all)

Just to clarify, Dr. Dror's study was published. The citation is found below and if you visit his website, you will find a complete list of his publications including links to the articles. You can download a copy of this article in .pdf for free. The link is below.
(John for shame!!! Don't you read your MAFS newsletter?? =) just kidding!)

ANd yes I agree, I hope he will continue his work, and I think it would be encouraging for us to show support by agreeing to participate in the research. We may not like all the answers we get, but if we participate, at least we will have a dog in the race (and no one to blame but ourselves if we aren't as good as we think we are!!!)


http://www.ecs.soton.ac.uk/~id/biometrics.html

Dror, I.E., Charlton, D., & Peron A. (2006). Contextual information renders experts vulnerable to making erroneous identifications Forensic Science International, 156 (1), 74-78.

Glenn/Pat and everyone, thanks for the kind words about the research. Everyone on this forum knows my background and I for one was sceptical and I told Itiel so. I said that while I wanted to work with Itiel on this research that I would never believe that bias and top down processing would be a factor in our profession. Then Mayfield happened along to confirm our research findings. We have a long way to go. Itiel and I have only just started to scratch the surface. I embarked upon this line of study as an academic exercise. Now I am more convinced than ever that this work is new and can help deliver a better profession. Glenn, Alice, John and many more friends have demonstrated support for this work. Thanks again. Glenn is doing my marketing job for me, but I echo his words, support the research and help to provide a novel insight into our world and help bring empirical understanding to what it means to be a fingerprint expert or any other expert witness.

cheers

Dave Charlton

I get this feeling like the context/bias issue is an elephant in the room that everyone is trying to ignore. Kudos to Itiel and Dave for finding such a unique, although somewhat painful, way to expose the issue. It's not like we all weren't aware at some level . . . how many times have you gone back to review an old case and wondered just how much coffee you drank that morning to find and identify THAT latent print! You are not necessarily less confident, sometimes just surprised at yourself. And if for some reason you are struggling with it - what is going on? What are the variables that could cause your confidence in a difficult identification to waiver?

These questions aren't as terrible as they seem . . . and unless we ask . . . we can't move forward. All sciences progress, as soon as one question is answered, ten more pop up. Even mathematics, believed to be the purist of all, is not perfect, cannot answer all questions, is still engaged in the act of discovery, still requires creativity of its practitioners to solve problems.

Science does not exist without the scientist. Scientists live in their time and place, guided by the paradigms of their world. These things aren't static. Most science plods along making discoveries, looking for answers to address practical problems. Occasionally it hits on something that causes a monumental shift in how the world is viewed. How it is VIEWED, which is not necessarily the same as how it IS - we all look at the world through our own eyes, in our time and place. What is it that effects our views?

So as we exist in the forensic world, with 100 years of progress behind us, let us keep on going. Its a journey, not a destination.

What about the next day when you`ve had no coffee and you still identify THAT LATENT print? That`s something else. Especially if you see it different without the coffee and you decide not to tell anyone. Especially if the person you decide not to tell is the one you already told about the ident and you don`t want to look a fool! And especially if the consequencies affect the outcome of a murder case. Who are the fools now?
I cannot imagine a situation where an expert could make an ident one day and not make it on another. On rare occassions a `missed` match is looked at again and matched on another day but not the other way round.

One thing we are learning about bias is the effect of being influenced in our judgments by someone or something else. If I can be influenced or biased from external sources, Can I be biased about determinations I have previously made? I would have to think, Yes. After all, I will most likely be defending my previous determinations. That should qualify as a bias. Learning about external sources of biases and internal sources of biases should give us some strategies for better dealing with bias. Can examiners have influences or biases, external or internal, and still reach the correct determination? Most definitely, Yes. Can influences or biases influence us negatively? That could happen. I at least have to recognize it could happen. I have to learn how to deal with influences and biases. Learning about influences or biases should help us deal with them, whether that influence is an excessive amount of coffee that I had or a peer I have worked with my whole career.

How do we ever challenge ourselves? At a later time or date, Re-start the examination without previous markings or documentation of what we had determined corresponds. Start from scratch. Some images in my notes have marked details on them, but for a re-examination, start with no markings of details in the images, even with my own previously examined images. When I prepare to go to court, or for some other reason I decide to re-examine evidence, I do my best to start over on the examination. Often in the initial stages of the re-examination I ask myself, where did I start on this one? There is always some concern, doubt, or waivering confidence if I get off to a slow start on determining sufficient details on the re-exam. I always start with doubt when re-examining some body else's determinations or my previous determinations. The early stages of the re-exam should always have doubt or waivering confidence, until doubt is eliminated and confidence is set. If there is no doubt or waivering confidence in the initial stages of the re-exam, how biased is the re-exam?

I noticed in Alice's comments she said "You are not necessarily less confident, sometimes just surprised at yourself. And if for some reason you are struggling with it - what is going on? What are the variables that could cause your confidence in a difficult identification to waiver?" I did not read this to be she changed her conclusion. I read this to be she re-examined the images. Challenge others' conclusions and challenge our own previous conclusions. I teach that doubt is involved, or needs to be involved, in the initial stages of all examinations, including re-examination of our own evidence. If any examiner starts an exam, or re-examination, without doubt about a previous determination, I would wonder about the biases of that examiner. Alice mentioned confidence waivering in the re-exam. That is what doubt is all about. She had worked her way through the re-examination and passed through doubt and reached the same conclusion. I did not read anything in her response that she changed a conclusion after a particular re-examination. To me, she did a good job of incorporating doubt into the initial stages of a re-examination, worked her way through waivering confidence because of the initial phases of doubt, and reached a conclusion that was not necessarily less confident than the initial examination. It seems to me she challenged herself without letting internal biases cloud her judgments on the re-examination.

To address a few comments from "redlion62" and Mr. John Vanderkolk:

John - your interpretation of my response is correct. I never look at any of my marked images when I re-examine for just that reason - I do not want to be tainted by what I looked at before, what do I see now? Was my past judgement correct based on a "fresh" look?

However I am glad "redlion62" raise the point further and I would like to see how fast my foot finds my mouth on this one:

If ever, I was to change my conclusion on an examination - I would not hesitate to notify my supervisor, the prosecutor, whomever - because I have ETHICS. Scientists re-evaluate their data, re-run statistics, re-do published experiments, and sometimes CHANGE their interpretation of the data before them. A scientist catching their own mistake and dealing with it still has a career because their peers understand that they are humans and can make mistakes.

What if 5-10 years have gone past and your eyes have deteriorated to the point you simply can't see the identification? What if you have become more conservative with your criteria and feel the comparison should have been inconclusive? What if you catch your own bad identification? I know what I would do, I think most would do the same. However there is no cure for egos so run amok they cannot admit mistake even at so great a cost. And these egos exist in all realms . . . . the latest I read was a cell biologist who falsified data in order to report the successful cloning of embryonic stem cells . . . he was a hero, soon he'll be parking cars at a casino (and probably making more money).

And there can be days when you are simply exhausted because kids have been up sick all night (or too much wine) and your eyes just aren't having it. Most examiners recognize their "good days" and "bad days" and generally have the luxury and wherewithal to wait for another day - to go back to it under proper rest /Starbucks levels. But what if you didn't?

Do I go back and look at my own comparisons with a suspicious eye? YES! Someone may spend the rest of their life in jail or be put to death in part due to my conclusions. Conversely someone may be set free and continue to prey on society in part due to my conclusions. I don't want to be wrong, even at the "embarrassment" of changing my conclusion. I don't know if it ever will happen, but it could happen. I am not perfect. The latent may be "the latent" and the known may be "the known" but it takes an expert to determine if they came from the same origin. As long as you have experts, there will always be the possibility of re-examining the data and changing your conclusion. You don't want it to happen often, it certainly can become an indication of a problem, but you have to allow for it, it happens in all sciences. Even judges reverse themselves on occasion (like when Judge Pollack realized Steve Meagher was a real person)!

All sciences have grey areas because it is practiced by people - people who don't know everything, who make decisions based on best practices of the time, who constantly make decisions and use creativity when designing research and crunching data, who sometimes make mistakes, who are sometimes unethical.

Looking foolish by changing my mind is something I am not afraid of redlion62 . . . having to live with that nagging feeling for the rest of my life because I believe I came to an improper conclusion and kept quiet - that would be the handbasket to my personal hell.