Analysis of children' mechanistic reasoning about linkages and levers in the context of engineering design

Reasoning about mechanisms is one of the hallmarks of disciplined inquiry in science and engineering. Despite the central importance of mechanistic reasoning, its origins are not well understood. Numerous curricular efforts involve simple machines and related physical systems, but these do not yet build toward a systematic and longer-term vision for promoting the development of reasoning about mechanisms. The research we describe here was developed in partnership with a team of engineers and science educators who aim to support the early development of mechanistic reasoning through a curriculum that challenges children to design kinetic toys called MechAnimations. Our research aims to characterize the intellectual resources available to children as they engage in design challenges and to describe the process by which these design activities may promote development of mechanistic reasoning. This paper provides an in-depth look at children's prior understandings of a key aspect of MechAnimation design -the mechanics of linkages and levers. In a flexible interview, 9 children at grades 2 and 5 were asked to predict and explain the motion of mechanical linkages. Children explored contrasting pairs of mechanisms, chosen to highlight components of the system important to its functioning (such as the location of the fulcrum in relation to the input). As one might expect, many student responses focused on aspects of the mechanical system that were not oriented toward its function. For example, "it looks like a plus sign." However, children also exhibited more sophisticated thinking, such as describing the parts and structure of the mechanisms. The most sophisticated student responses included mechanistic descriptions of how the parts and structure worked to: constrain motion, affect the direction of rotation, coordinate the direction of motion for input and output levers, coordinate the movement of lever arms, and affect the magnitude of motion. Overall, children who more readily tended to relations between input and output seemed better able to predict mechanism motion. All children demonstrated at least some elements of mechanistic thinking, but many of their responses lacked coordination of multiple elements. When children coordinated multiple elements, they were also more likely to successfully predict the motion of one or more outputs, given an input. Children who predicted incorrectly tended to exhibit mechanistic reasoning only after observing the mechanisms move, if at all. Although designed to ascertain children's' naive ideas about mechanisms, certain aspects of the interviews seemed to support the development of elements of mechanistic reasoning. For example, comparing the motion of contrasting mechanisms helped some children notice relevant variables not apparent before the contrast. These results suggest methods for characterizing mechanistic reasoning as well as potential resources for supporting its development. We anticipate that the latter may profitably be incorporated into design challenges.