Models of Scientific Explanation

EXPLANATION Explanation of why things happen is one of humans' most important cognitive operations. In everyday life, people are continually generating explanations of why other people behave the way they do, why they get sick, why computers or cars are not working properly, and of many other puzzling occurrences. More systematically, scientists develop theories to provide general explanations of physical phenomena such as why objects fall to earth, chemical phenomena such as why elements combine, biological phenomena such as why species evolve, medical phenomena such as why organisms develop diseases, and psychological phenomena such as why people sometimes make mental errors. This chapter reviews computational models of the cognitive processes that underlie these kinds of explanations of why events happen. It is not concerned with another sense of explanation that just means clarification, as when someone explains the U. S. constitution. The focus will be on scientific explanations, but more mundane examples will occasionally be used, on the grounds that the cognitive processes for explaining why events happen are much the same in everyday life and in science, although scientific explanations tend tobe more systematic and rigorous than everyday ones. In addition to providing a concise review of previous computational models of explanation, this chapter describes a new neural network model that shows how explanations can be performed by multimodal distributed representations. 2 Before proceeding with accounts of particular computational models of explanation, let us characterize more generally the three major processes involved in explanation and the four major theoretical approaches that have been taken in computational models of it. The three major processes are: providing an explanation from available information, generating new hypotheses that provide explanations, and evaluating competing explanations. The four major theoretical approaches are: deductive, using logic or rule-based systems; schematic, using explanation patterns or analogies; probabilistic, using Bayesian networks; and neural, using networks of artificial neurons. For each of these theoretical approaches, it is possible to characterize the different ways in which the provision, generation, and evaluation of explanations are understood computationally. The processes of providing, generating and evaluating explanations can be illustrated with a simple medical example. Suppose you arrive at your doctor's office with a high fever, headache, extreme fatigue, a bad cough, and major muscle aches. Your doctor will probably tell you that you have been infected by the influenza virus, with an explanation like: People infected by the flu virus often have …