Animacy & Memory

Abstract

Here we asked whether simple visual animacy cues can improve the encoding of spatial information into visual working memory. Subjects played a "matching game" in which they clicked on virtual cards to uncover animate and inanimate displays. In four experiments, the locations of animate displays were remembered better than the locations of any of the control displays. These results may reflect a novel form of adaptive memory .

Experiment 1: Darts

Our initial experiment featured an eight-panel matching game. Panels always contained a disc and five darts, all of which moved randomly. Darts in a given panel stayed oriented toward the disc (0° - the Wolfpack display), or 45°, 90° or 135° to its right. Subjects located Wolfpack panel pairs more efficiently, as evidenced by fewer redundent clicks. It appears that animate motion is prioritized in visual working memory.

Experiment 2: 180° Control

Could the Wolfpack's memory advantage be due to the convergence of the shapes' symmetry axes, rather than to any sort of perceived animacy? To rule this out, we developed a ten-panel matching game which included an extra pair of panels in which the darts stayed oriented 180° (i.e. directly away from) the disc - thus equating 'symmetry axis convergence' while eliminating the perception of animacy. Subjects matched such pairs efficiently, but not as well as Wolfpack pairs!

Experiment 3: Eyes

Is the Wolfpack's memory advantage specific to 'darts' (whose fronts are defined by conspicuous points)? To find out, we next replicated this effect using a very different kind of oriented stimulus - discs with 'eyes dots' on one side.

Experiment 4: A 'Lone Wolf'

How many wolves are required for such 'Wolfpack' effects? In our final experiment, we replicated the effect with single 'lone wolf' (again using a disc with 'eyes').