Other regions showed substantial divergence, reflecting the points at which methodological choices dominated over data.

The nasal region showed notable variation across methods — consistent with the well-known difficulty of inferring soft tissue nose shape from nasal bone morphology.

The lip and perioral region also showed divergence.

The overall shape of the midface showed somewhat more agreement, reflecting the clearer constraints imposed by the skull’s zygomatic and orbital architecture.

These quantitative results provide a more honest picture of what facial reconstruction can and cannot deliver than any single reconstruction could.

They specify, in geometric terms, where the evidence is strong — where different methods converge because they are all faithfully reflecting real skeletal information — and where it is weak — where different methods diverge because they are each making different choices in the face of underdetermined evidence.