Double Loop Podcast – Détails, épisodes et analyse

Eric and Glenn are back from a little summer break, prepping for the 2024 IAI Conference in Reno. They do a final “Where in the Whorld?” game. Then they jump into a recent paper on “Inconclusive” decisions (Swofford, et al. (2024) “Inconclusive Decisions and Error Rates in Forensic Science”, Forensic Science International: Synergy (vol 8; 100472)) authored by several members of NIST. The paper proposes a method for computing and communicating error rates when “inconclusive” decisions are made. The paper also focuses on making clear distinctions between “method performance” versus “method conformance”. The guys discuss their views on the method and the implications the paper may have for fingerprint examiners and their agencies. Swofford, H. , Lund, S. , Iyer, H. , Butler, J. , Soons, J. , Thompson, R. , Desiderio, V. , Jones, J. and Ramotowski, R. (2024), Inconclusive Decisions and Error Rates in Forensic Science, Forensic Science International: Synergy, [online], https://doi.org/10.1016/j.fsisyn.2024.100472, https://tsapps.nist.gov/publication/get_pdf.cfm?pub_id=957335 (Accessed August 31, 2024) Link to open source paper here: https://doi.org/10.1016/j.fsisyn.2024.100472

Glenn and Eric are joined by Angela Hilliard to discuss her experiences with a cold case from the Pacific Northwest. Genetic genealogy provided a suspect's name from a cold case double homicide, and a latent palm print was left on the victim's van. Angela walks us through her initial erroneous exclusion, the subsequent identification, court testimony, and the resulting press coverage. A must-listen episode for examiners on how to handle erroneous conclusions, even in high-profile cases. (Apologies for the audio in this episode. We did not realize until later that some of audio was poorer quality than normal.)

Eric and Glenn respond to the recent allegations that a computer science undergraduate at Columbia University, using Artificial Intelligence, has “proven that fingerprints aren’t unique” or at least…that’s how the media is mischaracterizing a new published paper by Guo, et al. The guys dissect the actual publication (“Unveiling intra-person fingerprint similarity via deep contrastive learning” in Science Advances, 2024 by Gabe Guo, et al.). They state very clearly what the paper actually does show, which is a far cry from the headlines and even public dissemination originating from Columbia University and the author. The guys talk about some of the important limitations of the study and how limited the application is to real forensic investigations. They then explore some of the media and social media outlets that have clearly misunderstood this paper and seem to have little understanding of forensic science. Finally, Eric and Glenn look at some quotes and comments from knowledgeable sources who also have recognized the flaws in the paper, the authors’ exaggerations, and lack of understanding of the value of their findings. Gabe Guo et al. ,Unveiling intra-person fingerprint similarity via deep contrastive learning.Sci. Adv.10, eadi0329(2024). DOI:10.1126/sciadv.adi0329 https://www.science.org/doi/10.1126/sciadv.adi0329 https://www.engineering.columbia.edu/news/ai-discovers-not-every-fingerprint-unique https://for-sci-law.blogspot.com/ https://www.cnn.com/2024/01/12/world/fingerprints-ai-based-study-scn/index.html

Glenn Langenburg and Eric Ray return to the annual IAI conference in San Antonio to talk to attendees about the interesting lectures and workshops that they've seen this year. Morgan Hays from Florida talks about how he stays current with the field as a supervisor and how the exclusion and verification topics were covered at the conference. Do you trust your verifier? Linda Manigault from Nevada then joins the conversation and discusses the workshops on FRStat and statistical models that she attended. Glenn ends the episode with a BIG announcement!

Glenn Langenburg and Eric Ray wrap up the three-part discussion on the AAAS Report on the Quality and Gap Analysis of Latent Fingerprint Examination. Surprisingly, the guys agree with many of the recommendations and conclusions in the document with an occasional quibble or clarification. As was hinted in the past two episodes, there are more serious concerns with the report's suggested language and proposed limitation of the word "identification".

Glenn Langenburg and Eric Ray continue the discussion on the pros and cons of the AAAS report. Carey participated with the review of the report and discusses the positive recommendations and the limitations that come from the report not being published in a traditional peer-reviewed journal. After that discussion wrapped up, Carey, Eric, and Glenn get a little off-topic and talk about attending AFIS Users' Conferences. Carey and Glenn presented at the most recent Idemia conference on the benefits of the Case AFIS system. Eric talks about a local AFIS conference and user discussions on how data analysis can lead to improved search practices.

Glenn Langenburg and Eric Ray finally get around to discussing the report on latent fingerprint comparisons from the American Association for the Advancement of Science. John Black (from Black and White Forensics) was on the committee that wrote the report and joins the discussion on some of the merits and limitations of the paper. While many of the research recommendations are the next steps that the latent print field needs, they are also the same recommendations made by other reports. One of the report's main failures is the suggested wording for the identification decision.

Glenn Langenburg and Eric Ray are joined by Della Wilkinson from Canada to discuss the final resolution of the Bornyk case from British Columbia. Originally, Judge Funt (see Double Loop Podcast episdoe 10) ruled that the fingerprint evidence would not be admitted after conducting his own study of latent prints and his own comparison. The Crown appealed and a new trial began with the original examiner, the verifier, Della, and Simon Cole. This episode the Double Loop Podcast also pays tribute to Tony Cantu. He will be missed.

Eric Ray sits down for an interview with Simon Cole from the University of California Irvine. Simon explains how he first became interested in forensics and in latent print comparisons in general, and then outlines the history of his criticisms of the field. Glenn Langenburg later joins Eric for a review of the interview and some of Glenn's perspectives on how Simon's criticisms and the latent print response have both evolved over time.

Glenn Langenburg and Eric Ray continue their interview with Matt Marvin from Ron Smith and Associates about the erroneous identifications and eventual exoneration of Beniah Alton Dandridge. This video (available on YouTube via RayForensics.com) goes through an in-depth and technical review of the latent prints, the errors, the exclusions to Dandridge, and the eventual identifications to the victim's son (who discovered the body).