Abstract
The twenty-first century is marked by the need for a paradigm shift within the forensic sciences. Techniques reliant on untested assumptions and subjective judgement should be replaced by robust methods based on relevant data, quantitative measurements, and statistical models. Traditionally, facial approximation has experienced targeted scepticism due to its reliance on visual
arts, conflicting or limited and often fragmented guidelines, history with overconfident claims in accuracy, no error measurements, and insufficient validation testing. Despite these limitations, facial approximation has demonstrated relevance in raising the public profile and supporting successful recognition during unidentified person inquiries. Following a scoping review on current practice, this paper confronts lasting issues with existing guidelines and proposes potential solutions. A streamlined approach to protocol selection is needed, with improved accessibility to relevant supporting guidelines that effectively report method reliability and error margins. Given challenges in biological profile estimation, a call for a globalised approximation model determined by skull morphometrics is needed for adult and subadult individuals. Potential variances in sex, population, and age categories needs to be carefully scrutinised to determine true accounts of practical significance and to ensure influences from study design and sampling error are mitigated. Though developments in facial soft tissue thickness data are recognised, feature prediction protocols have not received the same level of attention and are notably lacking for subadults. There is furthermore a need for more transparent and reproduceable protocols that are intrinsically resistant to cognitive bias that follow a logical framework that has been calibrated and validated.
arts, conflicting or limited and often fragmented guidelines, history with overconfident claims in accuracy, no error measurements, and insufficient validation testing. Despite these limitations, facial approximation has demonstrated relevance in raising the public profile and supporting successful recognition during unidentified person inquiries. Following a scoping review on current practice, this paper confronts lasting issues with existing guidelines and proposes potential solutions. A streamlined approach to protocol selection is needed, with improved accessibility to relevant supporting guidelines that effectively report method reliability and error margins. Given challenges in biological profile estimation, a call for a globalised approximation model determined by skull morphometrics is needed for adult and subadult individuals. Potential variances in sex, population, and age categories needs to be carefully scrutinised to determine true accounts of practical significance and to ensure influences from study design and sampling error are mitigated. Though developments in facial soft tissue thickness data are recognised, feature prediction protocols have not received the same level of attention and are notably lacking for subadults. There is furthermore a need for more transparent and reproduceable protocols that are intrinsically resistant to cognitive bias that follow a logical framework that has been calibrated and validated.
Original language | English |
---|---|
Publication status | Published - 20 Mar 2024 |
Event | 93rd Annual Meeting of the American Association of Biological Anthropologists - Los Angeles, United States Duration: 20 Mar 2024 → 23 Mar 2024 Conference number: 93 https://bioanth.org/meetings-and-webinars/93rd-annual-meeting-los-angeles-california-2024/ |
Conference
Conference | 93rd Annual Meeting of the American Association of Biological Anthropologists |
---|---|
Abbreviated title | 93rd Annual Meeting of the AABA |
Country/Territory | United States |
City | Los Angeles |
Period | 20/03/24 → 23/03/24 |
Internet address |