"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

There have been numerous studies showing 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].

Fingerprint experts currently rely almost exclusively on a non-statistical professional judgment or personal opinion approach to define relative weights for fingerprint ridge formations in order to exceed sufficiency thresholds to establish positive identification or exclusion.  However, notable erroneous fingerprint identifications, i.e. [Brandon Mayfield] and [Shirley McKie], and recent fingerprint studies indicate that fingerprint expert opinion can be an inaccurate and unreliable methodology, and in some cases reliably so. 

A recent study performed by Christophe Champod, Cedric Neumann et al, shows fingerprint experts do not agree how much weight to accord fingerprint ridge formations [72].  The study is significant because it reveals that fingerprint expert opinion cannot accurately or reliably measure and define weights for partial and subsequently total amounts of corresponding ridge formations in two impressions.  Without the ability to reliably measure and define total weights for aggregate amounts of ridge formations, fingerprint examiners have no valid basis to establish sufficiency thresholds for positive identification or exclusion.

"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."

  Pierre-Simon Laplace

In terms of estimating the number of close matches or look-alikes for a given arrangement of ridge features for a sized population group, it has been shown decisions by latent print examiners using mere professional judgment are less accurate than statistical probability modeling, and subsequently widespread disagreement results (see Validation Study).  

With exception to formal instruction provided by a small group of forensic scientists and examiners, e.g., Glenn Langenburg, Cedric Neumann, Christophe Champod, the author, etc., in general fingerprint examiners receive little to no training in probability theory, statistical modeling, or proposed ways to measure and define weights for fricition ridge formations.  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 weight to accord individual and aggregate fingerprint ridge formations and at what precise point a fingerprint individualization or exclusion can be inferred, a statistical modeling approach to fingerprint identification based on probability theory is presented. 

The T-Model defines quantitative values combined with qualitative metrics for individual and aggregate amounts of fingerprint ridge formations in two impressions in order to establish valid basis for sufficiency to establish inference for positive identification with a degree of probability that borders on certainty.  The quantitative-qualitative values are called "T-Values". 

 

“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."

Christophe Champod

Frequency of occurrence establishes, in part, the rarity of a ridge formation type and as a result its quantitative weight or value. The T-Model measures the quantitative weight for fingerprint ridge formation types in terms of frequency of occurrence, where quantitative weight is expressed as the reciprocal of the frequency.  The T-Model incorporates the quantitative weight for ridge features in position based on frequency of occurrence of intervening ridge count to the nearest neighbor. The subsequent combined frequency distribution for a ridge feature's "shape in position" or "shaposition" represents the quantitative significance of the particular individual ridge feature.

The T-Model measures the quality of fingerprint ridge features in terms of visual clarity, ridge reliability, and quality of agreement.  Ridge features that display reduced levels of visual clarity and/or ridge reliability and/or quality of agreement are assigned reduction factors.  As a result a total quantitative-qualitative value for amounts of corresponding ridge formations in two impressions are numerically expressed.

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 1/66 billion based on a human population of 300 million (the total population of the United States) multiplied by 10 fingers multiplied by 22 (the estimated number of parts per finger).  300 million is same default human population group used by the FBI in DNA Analysis.