How Things Appear to Work: Predicting Behaviors from Device Diagrams

Proceedings of the AAAI Conference on Artificial Intelligence, 12

Volume

Qualitative and Model-Based Reasoning

Model-Based Reasoning

This paper introduces a problem solving task involving common sense reasoning that humans are adept at, but one which has not received much attention within the area of cognitive modeling until recently. This is the task of predicting the operation of simple mechanical devices, in terms of behaviors of their components, from labeled schematic diagrams showing the spatial configuration of components and a given initial condition. We describe this task, present a cognitive process model developed from task and protocol analyses, and illustrate it using the example of a pressure gauge. Then the architecture of a corresponding computer model and a control algorithm embodying the cognitive strategy are proposed.