Back

Detailed shape characterization reveals complementary ecomorphological patterns in brain shape and size evolution across the avian radiation

Lowi-Merri, T.; Watanabe, A.

2025-10-14 evolutionary biology
10.1101/2025.10.14.682392 bioRxiv
Show abstract

Birds have exceptionally large brains for their body sizes, which is thought to have facilitated their ability to adapt and survive following the end-Cretaceous mass extinction, and has been associated with many of their diverse ecological adaptations today. However, size metrics are limited in their ability to fully characterize neuroanatomical variation, since the brain and its regions can have divergent morphologies across taxa while maintaining the same volume. Here, we employ high-density geometric morphometric (GM) shape data to capture localized anatomical differences in avian brain endocasts and compare how evolutionary patterns differ between shape and size. We find that the majority of shape variation falls along an elongated-globular spectrum, where terrestrial and aquatic birds generally have anteroposteriorly elongated brains while more aerial birds typically have more rounded, globular brains. Optic lobe shape contained the greatest amount of ecological signal, while rates of shape evolution in the cerebrum were the most uniform across the avian phylogeny. A supermajority (>90%) of endocranial shape variation is independent of evolutionary allometry, demonstrating that a geometric morphometric approach characterizes key aspects of brain anatomy that are missing from volumetric data. Broadly, our results suggest that neuroanatomical diversity in birds is not driven by the dominance of any particular factors, but rather through shape divergence in localized brain structures from the global allometric and ecomorphological patterns.

Matching journals

The top 4 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.