Humans and other bipeds: the evolution of bipedality

Abstract

The papers in this issue arise from presentations at the Anatomical Society mini-symposium on ‘Humans and other bipeds: the evolution of bipedality’ held at the Linnean Society Meeting Room at Burlington House, London, on 5 September 2003. The intention of the organizers was to present new perspectives on the evolution of bipedalism, the oldest characteristic of the human lineage, by placing this locomotor specialization in a broader comparative and biomechanical context, examining its possible antecedents, and trying to identify which features of this behavioral specialization are, and which are not, unique to the human lineage. A common theme to the papers is that bipedalism is not a single or by any means a uniquely human specialization. Nevertheless, mechanically, all forms of bipedalism can be energetically optimized in only two ways: by use of pendular mechanisms, or by using biological springs. The first tend to be more useful at slow speeds, the latter at more rapid rates of locomotion. Even so, how these mechanisms are deployed, singly or in combination, varies very considerably between taxa, and even closely related forms may have very different strategies, and efficiencies, depending on environmental conditions. We should therefore expect the mechanisms to be deployed in varying ways and to varying extent during human evolution, both synchronously and at different geological periods. Neill Alexander began the symposium by reviewing the diversity of bipedalism, and brought us down to earth by pointing out that we share this distinction with cockroaches, although they (sometimes) run bipedally rather than walk. In fact, excluding hopping and skipping, we share walking bipedalism with birds, as well as (sometimes) other apes, while apart from humans, birds and many lizards also run bipedally. Our uniqueness in bipedality is perhaps less in running, where flexed postures, and a single peak in the forces exerted by the ground against the feet are shared by humans and birds, than in walking, where we are characterized by highly extended knee and hip postures and clearly two-peaked ground reaction forces. The former features reflect the mechanics of springs, the latter the mechanics of pendula. Efficiency can be served by both mechanisms, but is not always found. Thus, comparisons with quadrupedal mammals of the same body mass show that human walking is relatively economical of metabolic energy; but human running is expensive. Bipedal walking is curiously economical for wading birds, but expensive for geese and penguins. Hartmut Witte and colleagues’ refreshing comparison, from an engineering perspective of robotic and human bipedalism, showed that the design features of human bipedality are set not only by our more recent common ancestry but also our most ancient affinities. Beyond more familiar energy-saving features such as the upright and inverted pendulum, the efficiency of human walking requires fine-tuning of the resonant frequencies of all body segments. As vertebrates, the mechanics of human bipedalism are based upon the premise of a trunk; and the geometry and muscular anatomy of the trunk is adapted for tuning of resonant frequencies, torsion of the trunk with reference to the lower trunk being controlled and timed by recruitment of elastic structures, including the spine, paravertebral and abdominal musculature. Robotics was also employed in the presentation by Bill Sellers and colleagues, but in this case virtual, evolutionary robotics. They showed how state-of-the-art genetic algorithm optimization of muscle activity patterns can be used to predict energetically optimal gaits for early hominid species that no longer exist, the bipedalism of which, as a consequence of distinctly different body proportions, must have been mechanically distinct from our own. Using forward dynamics modelling, in which motion is predicted from muscle activation patterns, body proportions and mass distribution (body build), they sh