The costs of bipedal locomotion through time in hominins

<https://vertpaleo.org/wp-content/uploads/2024/10/2024_SVP_Program_Final3.pdf>
Technical Session 12: Euarchontoglires
(Friday, November 1, 2024, 8:00 AM)
The costs of bipedal locomotion through time in hominins
Bipedal locomotion is a cornerstone of hominin
evolution and is hypothesised to have
contributed to the evolution of large brains and
stone-tool use. Elongated hindlimbs and
specialisations in the feet and hips of later
Homo are considered adaptations to minimise
further the costs of long-distance endurance
running. Many studies have used a variety of
approaches for estimating the costs of walking
and running in select hominin species; however,
it remains untested whether there was a general
trend through time to minimise locomotor costs.
Such a trend would be consistent with persistent
selection pressures to reduce the costs of a new
advantageous mode of locomotion. Here, we apply
Bayesian phylogenetic comparative methods to
biophysical models of bipedal walking and
synthesise estimates on locomotor costs with
morphological characters commonly linked to
efficient bipedal movement. We leverage data
from over 450 hominoid fossils to make
phylogenetically informed predictions about
hindlimb length, stature, and body mass, from
which we estimate variation in the mechanical
and metabolic costs of walking for 25 hominoid
species. Hindlimb length strongly predicts mass
specific locomotor costs across terrestrial
animals. When available, we also collected data
on the presence of characters linked to efficient
bipedal locomotion, such as larger articular
surfaces in the knee, shorter pedal phalanges,
and a rigid plantar arch in the foot. We then used
a Bayesian phylogenetic generalised linear mixed
model to test for a general trend in hindlimb
length, stature, body mass, and the mechanical
and metabolic costs of walking through time. Our
model accounts for the uncertainties in our
morphological predictions, fossil age ranges,
taxonomic assignments of specimens, and
phylogenetic topology. We also test for trends in
locomotor costs and associated traits within
species, such as in the long-persisting and
wide-ranging Homo erectus, and examine
potential deviations from an overall trend. For
example, species like Homo floresiensis
exhibited increased estimated locomotor costs
relative to other late occurring Homo due in
large part to its shorter hindlimbs. Future
analyses will test for an association between
locomotor costs, brain size evolution, and palaeo
environmental change. Our study sets a
benchmark for future studies on biomechanical
evolution that applies to many terrestrial
vertebrate clades.