Development of biological motion perception: Insights from late-sighted children

Abstract

Our visual system exhibits an exceptional capacity for processing the movement patterns of other humans (i.e., biological motion), which starts to emerge with very little visual experience. Even inexperienced individuals (human neonates, dark-reared chicks) can detect biological motion patterns. Patients treated for congenital cataracts can recognize human walking patterns in the first moments of unobstructed sight after a lifetime of near-blindness, not only from upright displays typical to real-life settings, but also from up-side-down displays. These demonstrations naturally lead to questioning whether visual exposure is central for acquiring this visual capacity, or if it is rather acquired phylogenetically or supported by non-visual factors (such as feedback from the observer's motor system). More intricate sensitivities, including walking direction discrimination and action recognition, are in place months after removal of congenital cataracts. However, it is not known if these capabilities are acquired after surgery or spared altogether. To probe this, twenty-five children with early-life visual deprivation and late sight onset participated in six tasks testing multiple aspects of their perception, starting prior to sight restoration, and continuing periodically for up to five years after. Our results indicate a gradual onset of abilities, modulated by the quality and extent of available visual input. Recognition of human walking patterns soon after sight onset is followed by a progression of other proficiencies: discrimination of walking direction, detection of coherent action patterns, and finally recognition of these actions. Perception of finer-resolution action patterns, including the ability to recognize fine-motor and face-movement displays, is not gained even over a period of five years. Intriguingly, recognition of complex two-person interaction sequences is exceptional, surpassing that of basic simple actions of a single individual. Our studies offer the opportunity to place different biological motion perception capabilities on a common developmental timeline, and reveal a stratification of their onset.