Forensic anthropologists have relied on direct examination of skeletal remains to estimate biological profile. Although digital technologies have transformed documentation and visualization, one important question has remained surprisingly difficult to answer: How well do the observations we make on physical bone translate into virtual models?
That question is the foundation of a student’s work in our laboratory at George Mason University.
The project initially seemed straightforward. We planned to compare biological profile assessments made on physical skeletal elements with assessments performed on handheld laser scan-derived 3D models. Like many validation studies, we expected to determine whether the virtual models were “good enough.”
The results were far more interesting: we found that some traits translated remarkably well into the digital environment. Others did not.
Perhaps the most surprising finding was that agreement and estimation accuracy did not always move together. Generally speaking, skeletal sex estimation maintained comparably accurate despite relatively modest agreement in trait scores, whereas population affinity traits showed comparatively better agreement between physical and virtual observations but lower classification accuracy. This disconnect forced us to think more carefully about what agreement statistics actually tell us, and what they do not.
A more nuanced interpretation of the findings is that traits that depend primarily on overall morphology and shape appeared to be more resilient in virtual environments. In contrast, traits requiring subtle surface texture, fine morphological detail, or tactile assessment proved considerably more challenging.
The review process also shaped the final paper in meaningful ways. In response to reviewers, we took a deeper look into the effects of participant experience, survey attrition, and the limitations of the study. These additions strengthened the manuscript and improved transparency and honesty.
Another aspect that makes this publication particularly meaningful is that it represents the first peer-reviewed publication originating from student thesis research in our laboratory at George Mason University. Seeing a master’s thesis develop into a publication in the International Journal of Legal Medicine has been tremendously rewarding and reflects the dedication of everyone involved in the project.