Perception, 27(suppl.): 68. [ECVP 1998; Oxford, England.]

We examined biological-motion processing as a function of motion quality (smoothness) and attentional load. Observers viewed a display of randomly moving dots and tried to determine the direction of heading (left versus right) of a point-light walker embedded within the dot mask. Motion quality was manipulated by inserting blank frames of either 0, 40 or 80 ms between successive walker frames. Attentional load was manipulated by presenting a concurrent task in which observers detected a changing rectangle among a flickering array of nonchanging rectangles. This task can only be performed with the use of focused attention (Rensink, 1996 Perception 25 Supplement, 2).

In the full-attention condition, observers were told to ignore the flickering rectangles and to concentrate on the walker task. Here, direction discrimination remained above 80% across all ISIs, consistent with previous findings (Thornton & Shiffrar, 1996 Investigative Ophthalmology & Visual Science 37 S742). In the divided-attention condition, observers performed the change-detection task and the walker task concurrently. When the motion of the walker was smooth (0 ms ISI), direction discrimination was relatively unaffected (94%). At 40 ms ISI, however, accuracy dropped dramatically (66%) and reached chance levels by 80 ms (53%). These results suggest a shift in the nature of biological-motion processing, from passive detection at very short ISIs to some form of active tracking at longer ISIs (Cavanagh, 1991 Spatial Vision 5 303-309).