Predictive Methods and Probabilistic Mapping of Subcortical Brain Components in Fossil Carnivora

Area of Science:

• Paleoneurology
• Comparative Neuroanatomy
• Mammalian Evolution

Background:

• Paleoneurology reconstructs nervous system evolution using fossil and extant species data.
• Fossil data limitations (availability, quality) constrain phylogenetic insights.
• Understanding brain evolution in Carnivora (Carnivora) is challenged by limited fossil evidence.

Purpose of the Study:

• To infer brain structure in fossil Carnivora by translating brain component relationships from extant species.
• To develop predictive models for estimating fossil brain component sizes.

Main Methods:

• Utilized high-resolution magnetic resonance imaging (MRI) on extant canids and felids.
• Employed 3D laser scanning on fossil Carnivora specimens spanning 40 million years.
• Derived predictive equations for cortical (gray matter mass, cortical thickness, gyrification index) and subcortical structures (caudate nucleus, putamen, external globus pallidus mass).

Main Results:

• Regression models demonstrated moderate to high predictability for subcortical masses in fossil Carnivora.
• Exploratory probabilistic mapping allowed reasonable prediction of 3D subcortical morphospace for fossil endocasts.
• Identified allometric departures and established adult species ranges in brain component size for fossil species.

Conclusions:

• The integrative approach successfully estimated brain component sizes in fossil Carnivora.
• This methodology provides a framework for integrating neurobiology and paleontology for understudied fossil groups.
• Facilitates dialogue between neurobiologists and paleoneurologists studying mammalian nervous system evolution.