In forensic anthropology, personal identification is mainly performed through a qualitative assessment and comparison of morphological bone and dental features between antemortem and postmortem data. Although non-metric traits have been traditionally considered as individualizing features, their potential has been limitedly investigated. Indeed, frequencies of variants can represent an additional tool to provide probabilities and likelihood ratios that an individual presents a combination of traits, hence quantifying a possible identification. This study investigates the potential of cranial non-metric traits as individualizing features in personal identification, and it describes the application of this probabilistic approach to a sample from a mass fatality which occurred in 2015. 119 crania of males were assessed for scoring 35 non-metric traits by presence and absence. For each cranium, the compound frequencies of independent traits, probabilities and likelihood ratios that a cranium presents a specific blend of traits were calculated. Over 70% of the likelihood ratios exceeded 1,000,000, providing extremely strong evidence that a specific set of traits corresponds to a cranium. Probabilities to find an individual with the set of traits within a group of 528 people (corresponding to the recovered bodies for this case) were extremely low (e.g., 0.006 people out of 528). The considerably high likelihood ratios and low probabilities suggest that combinations of cranial non-metric traits are extremely specific to the single individual, hence they represent valuable individualizing features. Despite this approach does not seem immediately applicable for the resolution of this case because of the dearth of appropriate antemortem images, collecting cranial non-metric frequencies may be worth of further investigation as a supplementary tool to screen potential identities and provide quantitative evidence to the investigators and the judge.
From traditional to innovative: implications of cranial non-metric traits in personal identification
Caccia G.;Alemanno S.;Campobasso C. P.;De Angelis D.;
2025
Abstract
In forensic anthropology, personal identification is mainly performed through a qualitative assessment and comparison of morphological bone and dental features between antemortem and postmortem data. Although non-metric traits have been traditionally considered as individualizing features, their potential has been limitedly investigated. Indeed, frequencies of variants can represent an additional tool to provide probabilities and likelihood ratios that an individual presents a combination of traits, hence quantifying a possible identification. This study investigates the potential of cranial non-metric traits as individualizing features in personal identification, and it describes the application of this probabilistic approach to a sample from a mass fatality which occurred in 2015. 119 crania of males were assessed for scoring 35 non-metric traits by presence and absence. For each cranium, the compound frequencies of independent traits, probabilities and likelihood ratios that a cranium presents a specific blend of traits were calculated. Over 70% of the likelihood ratios exceeded 1,000,000, providing extremely strong evidence that a specific set of traits corresponds to a cranium. Probabilities to find an individual with the set of traits within a group of 528 people (corresponding to the recovered bodies for this case) were extremely low (e.g., 0.006 people out of 528). The considerably high likelihood ratios and low probabilities suggest that combinations of cranial non-metric traits are extremely specific to the single individual, hence they represent valuable individualizing features. Despite this approach does not seem immediately applicable for the resolution of this case because of the dearth of appropriate antemortem images, collecting cranial non-metric frequencies may be worth of further investigation as a supplementary tool to screen potential identities and provide quantitative evidence to the investigators and the judge.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
