This web site presents an open source fingerprint match probability model.  The name of the model is the "T-Model" and is summarized as follows:

The T-Model takes into consideration all levels of fingerprint ridge detail, i.e., Level I, II and III, and uses simple statistics to estimate the discriminating values for fingerprint ridge feature "shapes in position".  In general, the discriminating value for ridge feature types are in inverse proportion to the frequency with which they occur, i.e., both in terms of "type" and "position".  These two values are multiplied together to estimate the total value for a fingerprint ridge feature.

The model has "matured past mere counts of corresponding minutiae by taking into consideration ridge feature clarity" (a variable that the National Academy of Sciences states no statistical probability model has addressed [91]). In addition the model takes into consideration ridge edge contour, ridge path, ridge feature reliability and the quality of agreement between latent v. exemplar ridge features. 

The model combines the above variables to produce a "best guess" for the total quantitative and qualitative weighting, i.e., discriminating value, for an arrangement of fingerprint ridge features present in latent v. exemplar fingerprints. This "total" quantitative-qualitative mathematical match probability model for fingerprints is the "T-Model".

T-Model 9.2 Fingerprint Calculator (Link)

Design

The T-Model is designed to conservatively estimate the discrimination power, i.e., match probability, of fingerprint evidence as well as predict the number of friction ridge close matches or “look-alikes” likely to exist in a given population group for a criminal case at hand.  When the model predicts the number of friction ridge look-alikes for an arrangement of fingerprint ridge features to be less than 1, e.g., when the final FMP is less than the reciprocal of the relevant population for the case at hand, then it establishes inference for identification.  The model guards against bias because it removes the decision-making process to make fingerprint identifications from the examiner.  It is the T-Model that infers a match, not the examiner.

The T-Model has been (and continues to be) subjected to the most difficult proficiency tests possible.  It has been tested against the most notable known erroneous fingerprint identifications ever made and it has been pitted against the largest and best amounts of  friction ridge look-alikes ever produced by an automated search, published, or otherwise found during the course of routine casework.  So far the T-Model has not been fooled into making an erroneous identification. 

The T-Model was first published online August 2008.  Since that time it has undergone refinement, continuous updates based on new experimentation and critical scrutiny.  It remains freely available to the broad scientific and fingerprint community for peer review and extended validation study.

T-Model Formula

The below formula is designed to estimate the upper-bound number of close matches or look-alikes present in a given population group.  The number of look-alikes calculated by the formula is conservatively high, yet not significantly so, which make it's application to criminal casework appealing.  The formula is as follows:

L = 120 (R) / T (Log T)

Where,

L = Number of look-alikes or close matches

R = Relevant fingerprint population

T = Total quantitative-qualitative value for the arrangement of ridge features 

The formula is a derivative of the previous 2 formula described in previous versions of the T-Model.  It is testable and available for independent scrutiny by the scientific community.  For further information, see The Formulae 

*          *          *          

“Don't believe, trust or accept this model.  Simply do the same experiments and find out for yourself if it has any validity or not.”

                                                                    Henry Templeman

 

(See Validation Study and Solicitation)      

 

A Challenge to the Fingerprint Community

The T-Model and this web site will be maintained and offered to the fingerprint community for testing and critical scrutiny until another model, method or procedure is shown to be more accurate.  At this time the author considers the T-Model the most accurate tool to make correct fingerprint conclusions, i.e., more accurate than human decision making by the expert latent print examiner using only professional judgment or a fixed point system, and more accurate than any other statistical probability model currently in existence. 

The author challenges the fingerprint community to a "competition" to find out which model, method or procedure is most accurate and reliable to make correct fingerprint conclusions.  Recommended requirements for the competition are as follows:

• Test samples must be distal segments of fingerprints, i.e., no medial or proximal phalanges and no palm prints.
• Test samples of latent fingerprints must be consistent with normal handling and processed using conventional latent fingerprint processing methods, e.g., fingerprint powder, ninhydrin, DFO, etc..  Image manipulation of test samples using digital imaging software is prohibited.
• Test samples of exemplar fingerprints must be consistent with conventional exemplar fingerprint recording methods, e.g., ink or livescan.  Image manipulation of test samples using digital imaging software is prohibited.
• Test samples must include information about the AFIS database fingerprint population if AFIS was used to find fingerprint "look-alikes" which were used in test samples, i.e., the exemplar was found as a result of an IAFIS search of 600 million fingerprints.  
  
• Test conditions, design and administration must be well-controlled, reproducible, and honest.
• Test conditions, design and administration must be carried out by an uninterested third-party team of scientists.
  

Comments and suggestions about the above "competition" are welcome and may be submitted to the author at the following email address:  henry@henrytempleman.com 

"In science, there are no universal truths, just views of the world that have yet to be shown to be false."

Brian Cox and Jeff Forshaw [105]

INTRODUCTION

When a number of friction ridge formations in two fingerprint impressions are found to correspond in shape and position, the probability to find even a very small amount of the most common occurring ridge formations that match is usually in the millions or billions to 1.   As a result, a small amount of highly discriminating corresponding friction ridge formations found in two fingerprint impressions can be defined such that the reciprocal of the match probability exceeds the world fingerprint population many fold.  Once the rarity of a small amount of corresponding ridge formations is estimated, objective criteria can be used to report that with reasonable scientific certainty the amount is significant enough to establish inference for source attribution to a particular individual.

The term “source attribution” should not be confused with “uniqueness” or one and only one to the exclusion of all others that exist, have existed, or will exist in the world.  When deducing source attribution there often is little need to establish that a fingerprint sample is found in only one person in the entire world.  Instead, source attribution is considered in the context of the case, and rarely would the entire world’s population be considered as the pool of plausible contributors of any fingerprint evidence sample. 

In practice, fingerprint match probability is calculated for a number of relevant population groups residing in the geographic area where the crime was committed.  When there is no reason to believe that a smaller population group is relevant, the model sets the human population group to 300 million, roughly equivalent to a total human population in the United States, the same rough population group set by the FBI to match DNA profiles for routine casework [84].*

*Note: The plausible number of people who could be the source of any latent print for any crime is always restricted to a number less than the total number of people on earth based on the time and location for a crime at hand.  It is illegitimate to set [a priori] the size of the relevant population at its maximum, e.g., the total number of persons on earth [34] and therefore is determined, or may be refined, on a case by case basis (see Relevant Fingerprint Population).

Fingerprint Source Attribution

"Professionals forget the following reality.  It is not the estimate or forecast that matters so much as the degree of confidence with the opinion." [88]

Nassim Nicholas Taleb

The "Individualization to the Exclusion of All Others" Opinion

The "individualization to the exclusion of all other sources" opinion traditionally used by latent print examiners has been qualified as inherently subjective and unscientific by the most eminent scientific organization in the United States, the National Academy of Sciences (NAS).  The NAS Report rejects the idea that ACE-V (e.g., the "Analysis-Comparison-Evaluation / Verification" methodology used by latent print examiners) has a zero error rate or can be used to reliably establish fingerprint source attribution.   In addition the report points out the need for scientific research regarding the rarity of ridge features and sufficiency thresholds to infer positive identification, e.g., based on experiment, and the application of the science of probability to justify conclusions [91].

NAS Report

"...merely following the steps of ACE-V does not imply that one is proceeding in a scientific manner or producing reliable results."

Summary Assessment of Friction Ridge Analysis, NAS Report, Pg 5-12

 

The T-Model uses the science of probability to define the rarity of ridge features types in position and sufficiency thresholds to infer positive identification for amounts of corresponding fingerprint ridge features found in two impressions based on the relevant population for the case at hand.  The model calculates the fingerprint match probability (FMP) utilizing the relevant population (RP) for the case at hand such that when the FMP is less than RP, e.g., FMP < RP, then inference for positive identification is justified.  In essence any FMP found to be greater than RP is a qualification for inference to identify since based on probability theory a close match or look-alike would not be expected.  This is the sufficiency threshold required to achieve the acceptable level of reasonable doubt tantamount to a judgment of near certainty [34].   

The Qualified Probability Opinion

Traditional statements for fingerprint source attribution are "identification, exclusion and inconclusive" and have excluded gray-scale probability statements as "highly likely", "likely", unlikely" and so on [34].   The gray-scale probability statement or "qualified opinion" attempts to assess the value of evidence with amounts of corresponding ridge features that fails to exceed a posterior probability of 1, however is deemed significant enough to report if it tends to make inference for positive identification more or less probable that otherwise.

The greatest strength, or weakness, of any probability based fingerprint evidence (PBFE) model is its ability, or inability, to reliably establish a posterior probability greater than or equal to 1 for amounts of matching ridge features in two impressions when utilizing Automated Fingerprint Identification System (AFIS) technology.   All AFIS technology is designed to find the largest and best friction ridge arrangement "look-alikes" that exist in its database.  As a result, any PBFE model that uses likelihood ratios to measure the aggregate weight of friction ridge arrangements present in two impressions must successfully pass validation testing against the largest and best fragmentary friction ridge look-alikes ever recorded in a given fingerprint population.  The model must either demonstrate zero specificity, e.g., a zero false positive error rate, or at the very least show an improved error rate over that found in human decision-making or in a competing model.       

The T-Model discards qualified probability opinion when the Fingerprint Match Probability (FMP) for an amount of corresponding ridge features found in two impressions (based on an AFIS search) is found to be greater then the reciprocal of the relevant population for the case at hand.  Only a model that demonstrates zero specificity, e.g., a zero false positive error rate when pitted against the largest and best friction ridge look-alikes AFIS can find, should be considered reliable enough for use in criminal casework.

It is significant to note that the T-Model has shown to be robust in its ability to calculate the fingerprint match probability (FMP) for the largest and best arrangement of fingerprint ridge feature AFIS look-alikes ever recorded as greater then the reciprocal of the relevant population for the case at hand, and therefore is able to correctly identify such look-alikes as insufficient to infer positive identification (see Error Rate in Terms of Best Look-alikes and  Error Rate in Look-alikes Calculated).

The Justified Opinion

The T-Model sets the demarcation for sufficiency to infer positive identification at 1, where if the Fingerprint Match Probability (FMP) divided by the Relevant Population (RP) for the case at hand is greater than 1, then there is valid basis to infer positive identification.  As a result, the inference for positive identification is considered a justified opinion. 

The justified opinion to infer positive identification may be considered to bridge the gap between the individualization to the exclusion of all others opinion and the qualified opinion by setting the demarcation for sufficiency to absolute 1. 

The justified opinion for fingerprint source attribution can be made on partial, fragmented parts of whole, complete fingerprints.  For purposes of performing routine latent print casework in a timely manner, the human population group of 300 million people (e.g., approximately the total United States human population) is recommended for use as the default T-Value (or Likelihood Ratio) that needs to be exceeded in order to establish inference for positive identification.

*It is significant to note here that 300 million people is the same default human population group used by the FBI to match DNA profiles for routine criminal casework [84].  

NAS and IAI Recommendations

As a result of the National Academy of Sciences (NAS) report released on 2/18/2009 titled Strengthening Forensic Science in the United States:  A Path Forward [Link], the International Association for Identification (IAI) made the following recommendations to members on 2/19/2009 [Link]:

"It is suggested that members not assert 100% infallibility (zero error rate) when addressing the reliability of fingerprint comparisons." 

and,

"Although the IAI does not, at this time, endorse the use of probabilistic models when stating conclusions of identification, members are advised to avoid stating their conclusions in absolute terms when dealing with population issues." 

The IAI acknowledgment of a non-zero error rate for ACE-V methodology and that fingerprint identification conclusions should not be stated in absolute terms supports the use of a probabilistic approach to fingerprints. 

It is significant to note here that the use of subjective probability in fingerprint identification was supported by forensic scientists Joshua Bergeron, Glenn Langenburg and Cedric Neumann during the State of Minnesota v. Jeremy Jason Hull Frye-Mack hearing [Link] in 2008.

IAI Resolution 2010-18 

On July 16, 2010 at the International Association for Identification Conference in Spokane, WA, members voted to adopt Resolution 2010-18 which in effect rescinds Resolution 1979-7 and Resolution 1980-5 forbidding latent print examiners to provide oral or written reports, or give testimony of possible, probable or likely friction ridge identification.  The adoption of this resolution ushers in a new paradigm for the field of fingerprints by allowing the significance of a fingerprint match to be expressed in terms of probability theory (see Probability Theory).

The full resolution is scheduled to be published in the Journal of Forensic Identification.  For subscription information, see International Association for Identification.

NEXT PAGE >>>

Historically, latent print examiners have testified to conclusions of identification or exclusion with absolute certainty and a confidence level of 100%.   However, notable mistakes have been made in the field of fingerprints, e.g., Madrid Error,  which means a confidence level of 100% cannot exist.

As a result, latent print examiners who testify to their conclusions with a confidence level of 100% or absolute certainty exaggerate the strength of the evidence and therefore mislead the trier of fact.

Although the T-Model has a tested zero error rate to make erroneous conclusions, the theoretical possibility for error always exists.  Therefore statements regarding the significance of a fingerprint match, or non-match, are expressed as an inference and with a confidence level that borders on certainty.

On February 18, 2009 the National Academy of Sciences (NAS) publicly released its report to Congress recommending fundamental scientific research and standards to strengthen the forensic sciences in the United States, including fingerprints (see below links).

Information on NAS Committee Members  1 

Executive Summary of the NAS Report  2 

Summary of NAS Report's Recommendations   3

The T-Model is grounded in experimental research that estimates the discriminating value of the various ridge formations and clusters of ridge formations in a flat fingerprint, e.g., the area of friction skin most often found at the scenes of crimes.

The T-Model speaks to the following concern raised in the NAS Report:

"More research is needed regarding the discriminating value of the various ridge formations and clusters of ridge formations.”

The T-Model provides standard values for the most frequently occurring ridge formations used in fingerprints, e.g., ending ridges, bifurcations and dots, and formula to estimate the discriminating value for clusters of ridge formations and the number of close matches likely to be present in a given population group.  Based on these estimates, the T-Model forms demarcations to establish inference for identification to a single source.