Question-driven Explanatory Reasoning about Devices that Malfunction

Abstract

Questions are at the heart of virtually any task an adult performs when using technological artifacts. It could be argued that any given task T can be decomposed into a set of questions that a sailor asks and answers. When a sailor encounters a device that malfunctions, the relevant questions are What's wrong? and How can it be fixed? . When an officer reads a technical document, the relevant questions are Why is this important? and What should I do about it, if anything?. When a young adult reads Navy recruiting material, the relevant questions are What's interesting? , Do I want to join?, and What are the perks?. The cognitive mechanisms that trigger question asking, exploration patterns, and question answering strategies need to be understood in order to design the messages and technological artifacts effectively. In turn, these inquiry strategies map onto cognitive components that are familiar to cognitive scientists, such as search, pattern recognition, comparison, case-based analogical reasoning, knowledge construction, and structure mapping. In projects funded by the Office of Naval Research, we have developed a cognitive computational model of question asking (called PREG) and question answering (called QUEST). Our current ONR grant is investigating relationships among a person's understanding of everyday devices (e.g., dishwasher, cylinder lock), the asking and answering of questions, and general psychometric tests of cognitive abilities. After reading about a device, the participants subsequently receive scenarios in which the device breaks down and they generate questions about the malfunction, eye tracking data are also collected at that time. Adults with high mechanical ability ask good questions that converge on faults. The quality of questions in the center of a breakdown scenario is the fastest way to find out whether an adult has a deep understanding of a device.