Forensic identification science rests upon the notion that certain attributes of inanimate objects, persons, or even the residuum of certain behavior contain so much variability that it is possible to associate evidence from a crime scene with the source of that evidence. But serious scientists have long realized that, for reasons physical, statistical and psychological, such credulousness reflects a fundamental misunderstanding of these techniques. Outside of DNA typing, forensic identification has no random match probability data to share with factfinders, and such experts generally assert that their examination indicates that the crime scene evidence and the defendant's sample share a common source, or some comparable formulation. Our research is concerned with the effects of the kind of testimony that is given, or which could be given, for the numerous other forensic identification sciences that have been in existence much longer than DNA typing (e.g., fingerprints, handwriting, microscopic hair comparison); these fields have no scientific basis of the sort that DNA typing does, and must stand instead on the foundation of the personal experience and judgment of their practitioners, who must support their claims in court not with empirical data but with testimonial assurances. How would such expert testimony be received if it included more complete and candid descriptions of how they reached their findings and the more limited meaning of those findings? In three experiments we examined the impact on judges and jurors of different ways of expressing the meaning of a forensic match. Overall, the data revealed that what factfinders conclude from the evidence varied as a function of the ways in which the expert's findings were expressed and whether or not the limitations of the accuracy of forensic identification science were presented.
Read full abstract