| Jurors are frequently called upon to evaluate DNA evidence in criminal trials. Such evidence is accompanied by statistical information that purports to quantify the likelihood of finding a match by chance. However, when a DNA match is obtained by searching a database, additional questions arise, such as whether the search method should be taken into account in that calculation. The scientific community has not yet settled on the proper approach. Moreover, research has not considered how jurors intuitively reason about DNA "cold hits.";Study 1 provided community-member jurors with a criminal case scenario, and varied how the DNA evidence was obtained (database, suspect confirmation, or unspecified) and the match probabilities provided. Jurors gave equal weight to DNA evidence, regardless of whether or not it was obtained from a database. They did, however, find the defendant guilty with greater frequency when the odds of a coincidental match were 1 in 3 billion, versus 1 in 10,000. When both numbers were presented together, one as a random match probability (RMP) and the other a database match probability (DMP), jurors' judgments largely emulated those in the 1 in 3 billion conditions, indicating that the DMP was not influential for their decisions.;In Study 2, the defendant was either identified by a cold hit, or no information was given. Again, this variable did not affect judgments. Certain subjects also received jury instructions intended to mitigate potential biases, but these proved largely ineffective. A final condition, in which a defense expert presented the DMP (in addition to the RMP), produced a substantial decrease in judgments of guilt, suggesting that the DMP only influences decisions when it is presented via expert testimony.;In addition, jurors who assumed the defendant had a prior criminal record were more likely to find him guilty; however, those in the database conditions were no more likely to think he had prior convictions. Jurors who watched CSI-type programming were also more likely to convict and more likely to think the defendant had committed past crimes, especially violent ones. Implications for the current statistical controversy and legal practice are discussed. |
