Merve Erdogan¹ (), Wenyan Bi¹, Ilker Yildirim¹, Brian Scholl¹; ¹Yale University

Visual processing seems specialized in several ways for the perception of other agents, and one of the best examples of this is biological motion—as when displays of moving dots (‘point-light walkers’; PLWs) give rise to rich percepts of locomoting agents. Despite thousands of experiments over decades of research, one of the most foundational questions about such phenomena remains unsettled: just how specific are they to *biological* motion? Addressing this question is challenging, largely because of the absence of non-biological comparison stimuli — since the translation or rotation of rigid objects (as in familiar structure-from-motion displays) lacks the rich characteristic *relative* motion of the points from PLWs. Here, to fill this gap, we introduce a novel visual phenomenon: the perception of rich behavior from dynamic point-light *cloths* (PLCs) — as when a sheet on a clothesline is waving in the wind. Across many experiments, we found broad similarities between the perception of PLWs and PLCs — in terms of both experimental results and phenomenological demonstrations. In the first place, people readily perceive the dynamic shape and behavior of cloths from point-light displays, and (as with PLWs) this depends critically on the points’ relative motions, since such rich percepts disappear both in static displays and in dynamic displays with spatially scrambled points. This applies not only to the perception of cloth structure itself, but also to more nuanced properties, such as a fabric’s stiffness. Moreover, the perception of PLCs is also highly robust, persisting (as do PLWs) e.g. even in displays with limited-lifetime points, or when embedded in noise. These results collectively demonstrate how the perception of rich internal structure from dynamic point-light displays transcends biological motion, and also applies to the behavior of other sorts of non-rigid entities: even beyond biology, the visual system extracts rich structure from surprisingly limited input.