Probable Cause

Statistical model may unlock use of fingerprint evidence in court.

fingerprintPearson Scott Foresman, via Wikimedia Commons

Fingerprints have been used for over a century as a way of identifying criminals. In current practice, however, fingerprint evidence is not permitted in court unless examiners claim absolute certainty that a mark has been left by a particular suspect. The problem is, this courtroom certainty is based purely on expert opinion, which—however well-informed—must in the end be subjective.

“Forensic scientists do not deal with exact reproduction,” notes Cedric Neumann, assistant professor of forensic science and statistics at Penn State, ”but have to draw conclusions from imperfect prints found at crime scenes.“ Yet less-than-certain fingerprint evidence is not allowed at all, which means that some less-than-perfect fingerprints that could be key pieces of evidence in court are simply not being considered.

“It is unthinkable that such valuable evidence should not be reported, effectively hidden from courts on a regular basis,” says Neumann. “Such is the importance of this wealth of data, we have devised a reliable statistical model to enable the courts to evaluate fingerprint evidence within a framework similar to that which underpins DNA evidence,” he adds.

That is, Neumann has come up with a way to express the weight of fingerprint evidence in quantitative terms.

To do so, he and his team first developed a method to assign numerical values to the fine details, or “minutiae,” typically used to define a fingerprint. After mapping these details on both a control print from a given suspect and a print obtained as evidence at the crime scene in question, the researchers then tested two hypotheses.

fingertipFrettie, via Wikimedia Commons

The first test—designed to establish the probability that the crime-scene print was made by the suspect—compared the control print with a range of other prints made by the suspect. The second test—to establish the probability that the crime-scene print was made by someone other than the suspect—compared the crime-scene print with a set of prints in a reference database. Neumann’s research team then calculated the likelihood ratio between the two probabilities. The higher the ratio, the stronger the evidence that the suspect was the source of the crime-scene print.

To validate the model, Neumann compared 364 crime-scene prints against reference prints whose owners could not possibly have committed the crimes in question. The new method, he reports, produced no misleading false matches.

“Current practice allows a state of certainty to be presented, in criminal-justice situations, which is not justified scientifically or supported by logical process or data,” Neumann says. “We believe that the examiner should not decide what evidence should or should not be presented. Our method allows all evidence to be supported by data, and reported according to a continuous scale.”

Cedric Neumann, Ph.D., is assistant professor of forensic science and statistics,

Last Updated March 14, 2012