ASB document 013

[Texas Pat]

From the current ASB draft of document 013 that is out for review:

3.3 disagreement

A single dissimilarity, or an accumulation of dissimilarities, that is deemed to be outside of expected variations in the appearance of impressions from the same source or an accumulation of dissimilarities between two impressions resulting in overall nonconformity.

I've been discussing this with a friend and it seems the OSAC and ASB are lost in the semantical weeds. A couple of questions seem hard to pin down, to wit:

1) Is “resulting in overall nonconformity” a source exclusion (that is the only conclusion and example that mentions ‘disagreement’, inconclusive with dissimilarities mentions having dissimilarities, not disagreement)? And ‘source exclusion’ says you have to have ‘strong disagreement’. What is the difference between disagreement that results in overall nonconformity and strong disagreement?

2) What would be an example of ‘overall nonconformity’ that would not result in a source exclusion?

3) Is disagreement a factor that lead to an exclusion, or are features labeled as disagreement after a conclusion is arrived at?

In SWGFAST, we agreed on definitions basically that a dissimilarity was something that looks different in a print or impression. A discrepancy is something in the friction ridge skin itself. Therefor, a dissimilarity may or may not lead an examiner to conclude that there is an exclusion. But if a determination is made that a discrepancy exists, then it is an exclusion (except for features such as scars, warts, etc.)

Thoughts?

[Michele]

Pat,

I looked at the spreadsheet and previous commenters noted the definitions were not clear enough. I've pasted some examples below. My comments are in parentheses.

-The difference between the terms 3.3 disagreement and 3.5 dissimilarity is not clear.
-Accept. Distinction has been clarified.
( )

-The term "disagreement" is defined in Section 3.3 but then appears to be used in a different context in Section 3.20 i.e. "strong disagreement" and "level of disagreement." The same use occurs in Section 4. 2. This use adds ambiguity and inconsistency which should not be present in a standard.
-Reject. Degrees of disagreement can exist. Disagreement Noted requires only disagreement, but Source Exclusion requires strong disagreement.
(I don't see anywhere in the document that says there are degrees of disagreement)

-Including wording of "single dissimilarity" seems out of date vs. only including "accumulation of dissimilarities"
-Accept with modification. Word "single" deleted.
(To me, the way it's been changed still implies disagreement can be a single dissimilarity, so the change didn't change anything)

-The definition for dissimilarity needs further clarification. Although the definition states that this term is not to be confused with disagreement, the definitions as written do not make the differences between these two terms clear.
-Accept. Definition of "disagreement" clarified.
(Personally, I do not understand the difference between dissimilarity and disagreement. It seems like this may be a new way of stating, if you can explain it then it's dissimilarity. If you can't then it's a disagreement. The problem with this is that if your want to ID a latent then you'll come up with some sort of explanation, even if there's no support behind it, "it's distortion"

I'll write in saying it's still not clear to me. Hope others do the same (public comments are due by Jan. 3rd).

[Mike French]

I did not work on this document at OSAC or the ASB, and I think the definition of disagreement could be simplified. But first I would like to explore this:

Texas Pat wrote:

In SWGFAST, we agreed on definitions basically that a dissimilarity was something that looks different in a print or impression. A discrepancy is something in the friction ridge skin itself.

My experience as an LPE, and later as someone who has observed the profession from the outside, tells me it's too optimistic to assume examiners (at any level) can reliably interpret what the friction ridge skin itself looks like from seeing an impression. I’ve seen this overconfidence repeated by some of the most credentialed trainers in the industry.

This was apparent in an exclusion class I attended where the trainers postulated several causes about the source of discrepancies between different image pairs, and not one of those included the most probable cause, which was most likely the result of livescan artifacts (not mentioned in the training).

Too often I think examiners are pressured to explain more than they are able, and that becomes painfully obvious in the training example above, and this pressure seeps into the standardization process as well.

[Texas Pat]

My experience as an LPE, and later as someone who has observed the profession from the outside, tells me it's too optimistic to assume examiners (at any level) can reliably interpret what the friction ridge skin itself looks like from seeing an impression. . . . Too often I think examiners are pressured to explain more than they are able, and that becomes painfully obvious in the training example above, and this pressure seeps into the standardization process as well.

I don't disagree with you at all, Mike. Your observations are absolutely on point. But let me add a little historical perspective.

The first vocabulary committee for SWGFAST was chaired by Dave Grieve. I was assigned to that committee. Try as we might, for several years we argued over words and definitions without producing even a draft document to present to the whole group, much less to the community for comment. The longer we worked on it, the more convoluted, disorganized, and unclear everything became. A number of us felt like we should just adopt the FBI's vocabulary in the appendix of "The Science of Fingerprints," or endorse Michele's dictionary, and move on. But the whole group did not approve either of those ideas.

Finally, the first committee was disbanded and reassigned. A new committee under Robin Bratton as chair took on the task. I was not on that committee so I don't know what magic Robin worked, but within two sessions of SWGFAST, her committee came up with an acceptable vocabulary document. It was a major loss to SWGFAST and the whole latent print community when Robin was promoted to Director of her lab and resigned from SWGFAST to take on bigger challenges.

In regard to these two terms, however, you are absolutely correct. But there may be some features appearing in a latent print that are clearly dissimilarities not due to skin structure, and there may be some dissimilarities that clearly have their origins as discrepancies in the skin itself. But you are, of course, quite right in saying that you just can't always tell. That ability to know when you can't tell comes with experience.

You are also spot on with your observation that we are often pressured to explain things that we simply can't know for sure. Unfortunately, too many LPEs pretend to know the answers and try to bluff it through, rather than admit they can't tell or don't know, not only in this, but in a lot of things. But that's an argument for a different thread, not one on what should be simple vocabulary terms.

My point was that it appears OSAC and ASB have fallen into the complex mire that the first SWGFAST vocabulary committee had fallen. We overcame that by starting over fresh with a different group of people unaffected by the mindsets of the first group. Perhaps that would be a good approach here, too.

[Michele]

Mike wrote, "I think the definition of disagreement could be simplified."

It's very hard to try to give a suggestion when I'm not sure of the intent. This is what I think they are trying to say, but I’m not really sure. With these definitions, it is not asking LPE’s to over interpret the reason, it’s just noting the difference in appearance and labeling if it is an important difference or not.

DISSIMILARITY: a difference in appearance between two impressions that is normal, expected, or transitory; dissimilarities are not factors of a comparison or evaluation (e.g., the width of ridges, the width of furrows, the size of the impression, the exact distance between features, temporary wounds, etc.)

DISAGREEMENT: a difference in appearance between two impressions that is unexpected if the two impressions originated from the same source; disagreement is a factor of the comparison and evaluation (e.g., the ridge count between features, the absence of a feature(s), etc.).

With these definitions, the question then becomes, what is distortion considered, a dissimilarity or disagreement?

If there is a self-evident reason for the difference, then it is a dissimilarity. If there is no self-evident reason, then it is disagreement and must be weighed appropriately. Perhaps that can be added in, then again, perhaps that's out of scope for this document but still an important concept. Maybe add examples:

DISSIMILARITY: a difference in appearance between two impressions that is normal, expected, or transitory; dissimilarities are not factors of a comparison or evaluation (e.g., the width of ridges, the width of furrows, the size of the impression, the exact distance between features, temporary wounds, self-evident distortion, etc.)

DISAGREEMENT: a difference in appearance between two impressions that is unexpected if the two impressions originated from the same source; disagreement is a factor of the comparison and evaluation (e.g., the ridge count between features, the absence of a feature(s), the interpretation of distortion, etc.).

Thoughts?

[Mike French]

I went to the ASB website to see the definition for myself. The most recent document I could find has these definitions:

3.3
disagreement
A dissimilarity, or an accumulation of dissimilarities, that is deemed to be outside of expected
variations in the appearance of impressions from the same source resulting in overall
nonconformity.

3.4
dissimilarity
An observation that two impressions have a general difference of appearance when comparing an
individual feature or detail. Not to be confused with “disagreement”.

These are what I had in mind when I suggested the definition could be simplified.

[4n6Dave]

Departing from Support for Same Source to Inconclusive with similarities seems like it is going in the wrong direction.
Inconclusive implies that the conclusion is neutral and by definition Inconclusive with Similarities is not a neutral conclusion.
I know there is a segment of the community that is opposed to the use of support for same source as it feels to probabilistic.

[Michele]

The spreadsheet of changes indicates why these titles were changed.

Line 339 states, “Perhaps correctly examiners are cautious of testimony sounding like a “possible ID” or “almost an ID” to the jury and this being misleading and prejudicial”.
Line 91 states, “A conclusion of support for same source will intentionally introduce bias.”

The resolutions for both comments say that the ASB accepted both of these comments and made modification by changing the titles.

[Shane Turnidge]

Optical artifacts are some of the most challenging dissimilarities of all. Live scan systems can create these features and there will be little in the way of tell tale signs that they are false features.
I like the SWGFAST approach as a means of addressing them via the abundance of similarities route.
If someone comes up with something better I'd sure like to know about it.
Shane

[Michele]

Shane,

I've always had an issue with the statement:
A competent examiner is trained to recognize the difference between distortion within the impression and a true discrepancy and will be able to explain this distinction in court (SWGFAST 2002c).
This was published by the FBI in 2009.

It feels as though the OSAC/ASB has changed the terms but kept the same intent (distortion is now dissimilarity; and discrepancy is now disagreement)

From the ASB recent public comment period for 016 - Terminology Related to Friction Ridge Examination, the current definitions are:
disagreement
A dissimilarity, or an accumulation of dissimilarities, that is deemed to be outside of expected variations in the appearance of impressions from the same source, resulting in overall nonconformity.
dissimilarity
An observation that two impressions have a general difference of appearance when comparing an individual feature or detail. Not to be confused with “disagreement.”

I have 2 issues with these definitions.
1) they do not tell examiners how to determine the difference between a dissimilarity and disagreement.
2) it appears that features are labeled as a difference or a dissimilarity based on the end conclusion (if an examiner excludes then it's disagreement, if they decide to ID then it's dissimilarity). To me, that's circular reasoning.

It'll be interesting to see if this is recognized and resolved.