![Fingerprint in blue and green pigment on a white background.](https://blogs.iu.edu/sciu/files/2018/06/prints_featured-1m3bhjf-e1529694374548-200x300.jpg)
We’ve all seen it on a TV crime series – fingerprints taken from a crime scene are put into a computer, instantaneously matched to a person of interest, and “Voila!” The crime is solved. As with many aspects of TV crime drama, however, fingerprint analysis is more complicated in real life. The reason? According to IU Psychology and Brain Sciences (PBS) professor Tom Busey, a lot of the work for fingerprint identification and matching is actually done by human experts. Although the process is much more difficult in real life than on TV, in the real world the result of a trial can still hang on fingerprint analysis. With this in mind, let’s explore the ins and outs of a fingerprint analyst’s work.
Busey, whose research generally relates to visual perception, has been working with crime scene analysts since the early 2000’s. His lab explores how fingerprint analysts become experts, how they differ from novices, and the process of learning fingerprint analysis in general. One of the basic things fingerprint analysts learn is which visual features of prints are useful for identification. Many of us know that fingerprints are unique to each person, and trained eyes can often tell fingerprints apart by weighing similarities and differences in ridge patterns to determine whether or not they are from the same person. But as Busey explains, in many cases, the actual task is more difficult, since each and every print we leave is also unique — even those left behind by the same person! Fingerprints from just a single finger on my hand can differ, depending on how wet or dry my hand is, what material I’m touching, and what I’ve just touched.
Since fingerprints left behind in the real world are quite variable, and also not as high in quality as intentionally placed ones, much of Busey’s research relates to the problems that fingerprint analysts face in determining whether or not multiple fingerprints are from the same person. Experts are trained for two years to identify and match fingerprints, and they each complete a supervised apprenticeship before finding work with law enforcement or in the private sector. Once trained, they can classify prints in question–which can come from a single crime scene, multiple scenes, or from fingerprint samples taken explicitly–as matching, non-matching, or inconclusive.
![Two different fingerprints are shown side by side, with matching features between the two prints highlighted in various colors.](https://blogs.iu.edu/sciu/files/2018/06/prints_busey-1vpsbbz-e1529694629981.png)
Aside from their rigorous training, Busey notes that fingerprint analysts have a more difficult job than we typically acknowledge. In addition to the issue of working with incomplete information and noise – like partial or smudged prints – the stakes of fingerprint analysis can be quite high because an analyst’s classification can send an innocent person to jail, or let a guilty person go free. Consider the consequence of claiming that two similar prints match when they actually do not. This problem has a simple logical solution: analysts should be cautious when matching samples and should rate prints as inconclusive more often. This too, however, can have its downside. If two prints, which actually match, are classed as inconclusive, a guilty person can go free.
This trade-off between “false alarms” and “misses” can be summarized in one question: How willing are we, as a society, to implicate an innocent person in order to catch a guilty one? The Busey Lab is currently working to understand how members of the public view this trade-off, with hopes of communicating this to officials. You can participate directly in this work by completing the survey found at https://www.indiana.edu/~buseylab/fingerprintvalues/.
![Image of a hand overlying a fingerprint sheet. Black ink is still visible on the fingers.](https://blogs.iu.edu/sciu/files/2018/06/prints-rl2vis.jpg)
The Busey Lab is also working with expert fingerprint analysts to identify which features are deemed most useful for matching prints; these observations may one day be used to train a deep-learning computer algorithm to perform fingerprint analysis, taking some of the pressure off of human experts in the case of ambiguous or high-profile cases. For now, however, there is no substitute for the well-trained eyes of fingerprint analysts! Work from the Busey lab shows us how basic aspects of human visual perception directly relate to practical problems in law enforcement, and this has broad impact and social relevance to us all.
Edited by Jennifer Sieben and Kerri Donohue
Leave a Reply