Journal of Experimental Psychology: Human Perception and Performance 1989, Vol. 15, No. 2,372-383 Copyright 1989 by the American Psychological Association, Inc. 0096-1523/89/S00.75 Understanding Collision Dynamics David L. Gilden and Dennis R. Proffitt University of Virginia In two experiments we investigated people's ability to judge the relative mass of two objects involved in a collision. It was found that judgments of relative mass were made on the basis of two heuristics. Roughly stated, these heuristics were (a) an object that ricochets backward upon impact is less massive than the object that it hit, and (b) faster moving objects are less massive. A heuristic model of judgment is proposed that postulates that different sources of information in an event may have different levels of salience for observers and that heuristic access is controlled by the rank ordering of salience. It was found that observers ranked dissimilarity in mass on the basis of the relative salience of angle and velocity information and not proportionally to the distal mass ratio. This heuristic model was contrasted with the notion that people can veridically extract dynamic properties of motion events when the kinematic data are sufficient for their specification. The abilities that people have in reasoning about natural motions are in part determined by the number of independent categories of information that physically describe the event. A theory of dynamic event complexity has been given by Proffitt and Gilden (1989) in which it is argued that people are capable of accurate dynamic judgments only when the event is so simple that a correct judgment can be based on a single category of information. This theory proposes that there are fundamental perceptual limitations that prevent organi- zation of different categories of information into synthetic multidimensional quantities. Categories of information, or dimensions, in Proffitt and Gilden's theory are the collection of object properties (e.g., spatial position, mass, shape, size, and orientation) that deter- mine the way in which the object moves. The dimensionality of an event is determined by the motion context in which objects appear as well as by the objects themselves. An ex- ample illustrates this notion of dimensionality. Figure 1 de- picts a wheel in two motion contexts that differ in their dimensionality. The first context is the free-fall motion of a wheel. In this context there are no object properties of the wheel that are relevant to its motion in a gravitational field. All wheels, regardless of their size, shape, and mass will fall in exactly the same way. The free-fall speed of any wheel depends only on the position of its center of mass, relative to the height from which it is dropped. In this context, the difference between the instantaneous position of the center of mass and the initial dropping height appears as the single This research was supported by Air Force Grant AFOSR-87-0238. The work of David L. Gilden was supported by a National Research Service Award Postdoctoral Fellowship HD-07036 from the National Institute of Child Health and Development. Heiko Hecht, Ellen McAfee, and Sue Whelan conducted the experiments. Stephen Jacquot programmed the stimulus displays. Portions of this article were previously presented at the 1987 meeting of the Psychonomic Society in Seattle, Washington. Correspondence concerning this article should be addressed to David L. Gilden, Department of Psychology, Gilmer Hall, University of Virginia, Charlottesville, Virginia 22903-2477. category of information that specifies the wheel's motion. This event is one dimensional. In the second context, the wheel rolls down an inclined plane. There are now two categories of information that determine the wheel's speed: the position of the center of mass and the moment of inertia that the wheel has about its center of mass. This moment is determined by the ratio of the inner and outer diameters. Wheels that are rimlike roll down the ramp slower than wheels that are solid throughout. Mass and absolute size have no relevance to the motion in either context. The abilities that people have in reasoning about wheels depend critically on the motion context in which the wheel is presented. Most people know how things drop, although terrestrial experience (more massive objects generally have greater terminal speeds in resistive atmospheres) and the failure to distinguish between force and acceleration do induce a bias toward supposing that heavy objects drop faster. Peo- ple's understanding of the rolling context is much different (Proffitt, Kaiser, & Whelan, 1988). People often suppose that the irrelevant dimensions of absolute size and mass have dynamic influences. Moreover, there is a high agreement that mass distribution (moment of inertia), the only physically relevant parameter, is irrelevant. Inability to reason about rolling wheels extends even to bicycle racers, high school physics teachers, and college professors of physics (Proffitt, McAfee, & Hecht, 1989). Although members of the latter two groups can, given sufficient time, solve rolling-wheel problems on the basis of the equations of motion, their formal under- standing does not translate into immediate and accurate impressions of the sort that are encountered for wheels that are dropped. We have tested this theory of dynamic event complexity in a number of physical environments. Assessed performance in reasoning about such objects as balances and tops (Proffitt & Gilden 1989; Proffitt et al. 1989) as well as about Archimedes' principle (Whelan, 1987) has consistently shown that, in general, people reason adequately about objects only in one- dimensional motion contexts. Performance in multidimen- sional reasoning tasks ranges from mere inaccuracy (balances) to amazement (tops). Although each environment has idio- 372