Chapter 12 Project IMPL: Implementation and Practice

12.1 Demonstrative Evidence and Firearms

Project purpose

1. Examine the use of demonstrative (picture) evidence in examiner testimony
2. Develop and test language for quantitative firearms analysis testimony
3. Validate quantitative language for firearms analysis with and without demonstrative evidence

12.1.1 Previous work

The FRStat program produces a probability that a fingerprint was created by the subject, as opposed to being produced by a different source. This language was investigated in Garrett et al.’s “Comparing Categorical and Probabilistic Fingerprint Evidence” (2018). Garrett et al. found that there was no significant difference in conviction rates between simple match language and the strongest (as well as the weakest) probability language. In terms of the likelihood that the defendant committed the crime, Garrett et al. found that the lowest probability condition (the probability being 10 times greater for the subject versus a different source) was significantly different from the highest probability condition (1,000,000 times greater). However, there was not a significant difference between the lowest probability and the next highest probability (10,000 times greater).

12.1.2 Study Design

This study design is a 2x2x3 factorial. The first factor is the presence or absence of demonstrative evidence. The second factor is the presence or absence of the bullet matching algorithm, and the third factor is the expert’s conclusion: match, inconclusive, or not a match.

12.1.3 Sample Testimony

Two different expert testimonies have been drafted. The first testimony is that of the firearms examiner. This testimony relates specifically to the case used in the study.The second testimony is that of the algorithm expert. This testimony relates to the details of how the algorithm works, as well as when it may be used.

12.1.3.1 Firearms Examiner

The firearms examiner’s testimony includes a description of how bullets are marked, as well as how the expert examines the bullets using a comparison microscope. The examiner then concludes that the bullets either match, do not match, or are inconclusive.

Demonstrative evidence for this section would include: an image of barrel rifling, an image of a fired bullet, and an image of two bullets aligned using a comparison microscope.

When the bullet matching algorithm is included in the testimony, the firearms expert would describe their experience with the algorithm, as well as a simple description of the algorithm and the score it produces. This score should generally correspond to the examiner’s conclusion. Demonstrative evidence would include an image of the land-to-land scores.

12.1.3.2 Algorithm Expert

The algorithm expert’s testimony includes a more in-depth description of how the algorithm works: finding lands, extracting signatures, and comparing signatures to produce a score. It also includes information about how the algorithm has been tested, as well as the validity of using the algorithm for the case in question.

Demonstrative evidence for this section would include: an image of how lands can be located, and an image of the comparison of two signatures.