Preschoolers have a quite sophisticated sense of the sort of mechanical causality that is intrinsic to the motion of simple physical solids. For example, when two events precede another one, they will usually correctly sense which is more physically plausible and then prefer it as the cause (Bullock, Gelman, and Baillargeon, 1982; Gelman and Lucariello, 2002). When preschoolers’ spontaneous explanations of various entities are examined in large transcriptions of everyday speech, the children flexibly and easily employ causal reasoning, using different kind of explanations depending on whether the events are thought of as physical, psychological, or biological (Hickling and Wellman, 2001). They show similar distinctions in more experimental tasks (Heyman, Phillips, and Gelman, 2003). Indeed, when asked to explain anomalies in physical regularities, children use very different patterns of reasoning than when explaining anomalies in social conventions of moral rules (Lockhart, 1981).
Preschoolers are also adept at inferring hidden causes. Thus, they assume that similar external motions of animate and inanimate objects are governed by radically different internal causes (Gelman, Durgin, and Kaufman, 1995). They understand that unseen factors must be linked to observable ones in systematic ways that are mechanistically mediated (Yoachim and Meltzoff, 2003). Moreover, preschoolers are quite sophisticated at using complex patterns of covariation over time to infer hidden causes and not just correlations (Gopnik et al., 2004), although often such inferences may be constrained by prior mechanistic theories that they are applying to those tasks (Griffiths, Baraff, and Tenenbaum, 2004). Finally, preschoolers will track a sequence of events occurring in causal chains and infer that the first event in that chain is most likely to be the most important cause (Ahn et al., 2000), a strategy frequently used by adults as well.
A vast literature on science “misconceptions” argues that erroneous beliefs about the physical world are held by many, ranging from preschoolers to adults. And many of these beliefs are highly resistant to change by instruction (Chi, 2005). Much of that literature, especially in the area of mechanics, has focused on high school and college students (e.g., Brown and Clement, 1987; Carmazza, McCloskey, and Green, 1981; Minstrell, 1983, 1988; Clement, 1982); there have been many fewer studies of younger preschool or elementary schoolchildren (Doran, 1972; Ioannides and Vosniadou, 2002; Viennot, 1979). This literature makes clear, however, that the elegant theoretical construction of Newtonian mechanics (including its three primary laws of motion) is by no means obvious even to high school or college students who have had courses in introductory mechanics.
Student misconceptions are sometimes revealed in tasks in which they are asked to predict the trajectories of objects or evaluate whether an observed trajectory is possible or impossible, but even more often when they are asked to identify and explain the forces acting on an object in a given