"It is very certain that, when it is not in our power to determine what is true, we ought to act according to what is most probable."

René Descartes

Numerous studies have shown that simple statistical methods are more accurate than humans and that the more complicated the problem the less likely an “expert” will beat a statistical model.  One reason why humans perform poorer than statistical models is because they fail to assign the right weights to the overall “equation”, but they think they do [53][74][75][76].

At this time, fingerprint experts rely exclusively on professional judgment to estimate the discriminating value to of fingerprint ridge features.  However, notable erroneous fingerprint identifications, i.e. Brandon Mayfield, and recent fingerprint studies show that fingerprint expert opinion can be an inaccurate and unreliable tool, and in some cases reliably so.

August 2010, Bruce Bodowle gives a talk on "Perspectives on Error Rate Reporting in Forensic Casework and Testimony" and expresses the need for latent print examiners to learn statistics (click below link to view video.)

“The recognition of identification fields as scientific domains seems deeply related to the capability of the field to provide reliable statistical estimates (either objective or subjective) of the rarity of identification features."

A recent study performed by Christophe Champod, Cedric Neumann et al, shows fingerprint experts do not agree how much weight to accord fingerprint ridge features [72].  The study is significant, because it reveals that fingerprint expert opinion does not accurately or reliably measure the discriminating value for fingerprint ridge features.

In general fingerprint examiners receive little to no training in probability theory, statistical modeling, or proposed ways to measure and estimate discriminating values for fingerprint ridge features.  As a result fingerprint examiners are for the most part unaware of how much it could likely help them make more accurate, reliable judgments.

Based on works by Meehl [53], Ayres [74][75], Trout and Bishop [76], Champod and Neumann, et al [72], there is overwhelming evidence for the need of fingerprint experts to apply probability theory and statistical models when making decisions about fingerprint identifications or exclusions, especially if the amount or volume of ridge detail present in two impressions is unusually distorted, complex, or “borderline”.

In an effort to make more accurate and reliable decisions about how much discriminating value to accord individual and aggregate fingerprint ridge features and at what point a fingerprint individualization or exclusion can be inferred, a statistical probability model, e.g. The T-Model, was designed.

"The theory of probabilities is at bottom nothing but common sense reduced to calculus; it enables us to appreciate with exactness that which accurate minds feel with a sort of instinct for which of times they are unable to account."

NEXT PAGE >>> 

The T Model has been subjected to validation testing and shown ability to correctly identify with zero error rate (so far) the largest and best amounts of corresponding ridge formations ever recorded in a non-match, including the most notable erroneous fingerprint identifications, as insufficient to infer individualization (see Clark Non-Match, Chesapeake IAFIS Non-Match, and Error Rate in Look-alikes Calculated).

For purposes of "routine" casework the T-Model defines the terms identification (or individualization) as a match probability less than or equal to 1/34.2 billion based on a human population of 300 million (the total population of the United States) multiplied by 10 fingers multiplied by 11.4 (the estimated number of parts per finger)(see Pre-Determined Minimum to Individualize).  Note:  The total United States population is same default human population group used by the FBI in DNA Analysis.