Creases

[kimba325]

Just interested in a little discussion on the following topic-

We have been discussing recently in the office crease detail and in what ways crease detail alone can be used. There seems to be some differences in literature as to what level of detail creases fall within. My main question would be: Can crease detail be reliable enough to exclude (I would never identify solely on crease detail). The more classes I attend and the more fingerprint images that I see with distortion, the more I prefer to have some L2 detail to exclude. Any comments...Thanks for any help with this topic you can provide.

[Gerald Clough]

First, I would make the following point. Identification and Exclusion both depend on determination that some quantity of impressed feature represents specific skin features. In other words, you're observing actual impressions of skin features you can characterize, and in doing so you assign them any qualities known to apply to such skin features, things like permanence. Otherwise, the difference is merely in the thresholds, a structure or complex of structures in the latent that does not appear in the record (Exclusion) or a sufficient complex of structures that appears in the latent and the record (Identification).

The point is that both conclusions depend on interpretation of exactly the same structures. No, you don't use L1 to identify but can use it to Exclude, but that's really because it's simply not a sufficient complex of features for Identification. But let's keep it simple and talk only about L2. Exclusion and Identification can both use L2, and they use it in the same way, which is by comparing observations in the latent and the record. So I think you have to articulate just why you would use another structure - creases - for one conclusion but not another. It is not like using L1 to Exclude but not to Identify. L1 has no "quantity" measure; it's only ever a single "feature." Creases may be many or few. There's a very large range of quantities that may be observable.

Of course, they are not quite like L2D. The relationships among creases are created differently from the relationships among L2 details. But there are indeed chances to observe and reliably characterize crease relationships. I think if you're going to accept the use of creases for one conclusion, you must be prepared to accept them for the other. Or, of course, accept them for neither.

My own belief is that there may exist observable quantities of creases sufficient for me to form an expert opinion that they represent the same source. I can no more tell what quantity that is always going to be than I can tell a quantity of L2 detail that will always form an identification. And I do not have anything like the enormous historical experience that forms our consensus of how much L2 may identify. So I'm probably asking for a lot more crease detail to identify that would be found to make source identity near certain if we had been doing tens of thousands of crease identifications over the years. But in using creases, I would be invoking knowledge of how different creases are formed and why they are formed and what has been observed about how creases tend to appear on human hands and my own experience, etc, all the things I would cite to justify a conclusion from L2D, except the very rich history.

This is not to say that there wouldn't be a lively legal argument over such evidence. "Fingerprint" identification is virtually guaranteed admissible. I wouldn't give crease identification or exclusion much beyond a 50/50 chance in an all-out fight with an astute opponent. But that would be strictly legal issues and doesn't affect my opinion. We always have to remember that distinction between what's admissible and what's valid expert opinion. Is it that you couldn't become convinced by creases or that you would be afraid of the admissibility issue? (At any rate, better tell the D.A. it's shaky ground.) I think though, that if it got in, it would be legitimately convincing. I'd actually want a vigorous defense fight over admissibility. Anything less would probably be ineffective assistance, and we'd have to do it all over again one day.

I actually feel like I would have a more difficult time explaining an Exclusion using creases than I would an Identification. Some creases are not present throughout life (like those ones on my fingers that certainly weren't there 40 years ago), and some creases may appear or not, depending on pressure and flexion. They're not L2D, and there are considerations that are specific and unique to creases.

[kimba325]

Thanks for your insightful reply Gerald. I want to clarify that I am a "newbie" in the fingerprint field (Almost four years of experience - 3 1/2 of those years was spent in training), and it is this limited experience that at this point I state I would not make an identification on creases alone. I am aware that there are known identifications that were made solely on crease detail.

The crease detail that brought up the discussion in the first place is in the palm (the thenar area to be exact). The latent and the known contained crease detail that did not correspond (known had prominent edge creases that the latent did not contain, the latent contained much more cross-hatch creases and cat scratch creases than the known) and like I stated before L2 detail was obliterated in the latent (appeared to have heavy pressure distortion).

[Ernie Hamm]

Forget the finger, palm, friction ridge aspects; you are involved in a pattern transfer comparative examination. Basically, could this object make this mark (opinion of origin)? The “object” can be an area of a finger, palm, sole, footwear, tire, rubber stamp, fabric or any object having a surface with some pattern that can be transferred through contact or pressure to another surface. Any comparative examination will evaluate features and marks found in questioned and known with consideration to time and circumstances (if known) of transfer, method of transfer and substrate to establish an opinion of origin. Latent prints are a form of pattern transfer involving friction ridge details, but features resulting from age and flexion creases are in a different area of comparative examination. Now, L1D and observable but limited L2D features associated with friction ridges can certainly influence an opinion of origin when crease markings are present and in agreement.

[David Fairhurst]

Ask yourself this question: what is a crease?
Well, on the skin it's a place where the skin is folded or depressed, but what about on the fingerprint form or in the latent print?
It's nothing. It is the absence of ridges. The only reason you can see a crease is that you can't see ridges. If you took the ridges away you wouldn't have a crease.
Where I am leading here is that creases ARE ridge detail.

Try looking at it like this. Creases, like ridges, exhibit all three levels of detail.
Level one crease detail is simply the type of crease; major flexion crease, chequer creases on the thenar area, E-lines, etc.
Level two crease detail is the specific path of the crease and the pattern of branches or chequers they form.
Level three crease detail is their shape; that is the shape of the ends of the ridges at thier margins.

I beleive that it was this realisiation that crease detail could be described in the same way as ridge detail that gave me the confidence to start calling identifications on creases alone.

Good luck.

[Gerald Clough]

I think I like the analysis of multiple levels of crease detail. It's not, nor was it intended to be, really analogous to ridge detail levels, but it imposes some structure on interpreting creases.

And since we've recently been talking about using probabilities and the arguments for and against using every possible bit of information, I suppose we would have to think about a a single transverse palmar crease (simian crease) and its known one in 30 incidence (varies with the study), it preference for males, and the preference for Asians/Native Americans.

Kimba:

As I said, I have a notion that the relationship between Exclusion and Identification may be somewhat different for the two modes, using ridge detail and using creases. I suppose I feel this way because creases are impressed as voids. So you can extend and enhance lesser crease details by flexing and extending the hand. You can't really do anything like this with friction ridges. I can't alter the fundamental character of L2D or create newly visible L2D in that way. But some of this with regard to creases depends on what you're looking at. Obviously, the larger patterns of multiple crease paths or the clear location and direction of creases may be so different as to be a clear Exclusion.

Interestingly, your actual case problem displays what I see as both situations. In one part, there's a lack of detail in the latent, and you have to rule out the possibility that it wasn't impressed on account of pressure, etc., and in the other part, there's detail that doesn't appear in the record but that may be finer details that evolved with age, so that has to be ruled out, although you often have the luxury of getting a current inked set.

[Neville]

I see it like this, creases do change with time, as we get older the crease lines become more and more obvious until they almost obliterate the level two and three detail in some people, this takes many years and the fine lines will come and go occasionally (possibly just the recording), however once there, they appear to remain and I for one have made identifications on these creases but there has always been some L2D to support these findings, that's just my nature.

I have identified foot prints on size, L1D, shape and creases, and this was accepted by the high court, I prepared charts with grid lines drawn across known and unknown impressions. There was no other forensic evidence as I remember it, though I was not called to court so I can't be sure of that. One thing I can say is that the O/C case said the crown prosecutor was very pleased with my submission.

I like Ernie's thoughts on this one and I can't really disagree with David's thoughts.

I tend to look at creases as being similar to tool mark and shoe mark identifications

My main issue is with how old are the known impressions, the time of recording must be as close as possible to the unknown impressions for make a comparison.

[kimba325]

Thank you everyone for this discussion. It has helped me. Some of my co-workers actually did some experiments in the lab leaving impressions on a similar surface and varied pressure. Their outcomes have helped me become more confident in what happens to creases as pressure increases (L2 became obliterated, but the crease detail remained/became more prominent). I think I will do some of my own trials just to see it first hand and increasing the number of subjects having corresponding results will only increase my confidence level of what happens to crease detail. I also appreciate the suggestion of having as close in date as possible of knowns (which I think mine were pretty close) because I don't know if I would have thought about that as a concern.

This is why I love this field....there is never a time to stop learning.

[Neville]

excellent and this web site means you do not have to reinvent the wheel.

Thanks Kasey, what a difference this site would have been to me when I was training. But that was pre Amstrad.