ABSTRACT
Allen Newell's primary research goals were so fundamental, and his accomplishments so prodigious, that two fields—cognitive psychology and artificial intelligence (AI)—trace their ancestry to his pioneering work with Herbert Simon in the late 1950s. Newell and Simon started investigating the nature of intelligence by using very simple domains: closed-form games and puzzles. Forty years later, their successes are indicated by the fact that we are able to use the current versions of the methodologies and theories that they invented to investigate the cognitive processes that support scientific discovery: a domain that represents one of the pinnacles of human intelligence. The early tensions and mutual interactions between psychological approaches and artificial intelligence approaches remain in the studies of scientific discovery: In psychology, the research goal is to determine just how people manage to do science, whereas in AI the goal is to build systems that can make discoveries. This work has produced an accumulating body of evidence that there can really be a “science of science.” As a result, the old view of scientific discovery—that it is mystical, ineffable, transcendent, unknowable—is giving way to both a descriptive and a syn-thetic science of discovery. The descriptive side is mainly from cognitive psychology, and the synthetic side is mainly from machine learning. Early interest in the psychology of science can be traced to Bruner, Goodnow, and Austin (1956), Wason (1960), and Simon (1966, 1973), among others. The state of the art as of a dozen years ago is summarized in Tweney, Doherty, and Mynatt (1981). The more recent resurgence of interest in the “cognitive science of science” can be attributed to Simon and his colleagues (Cheng & Simon, 1992; Kulkarni & Simon, 1988; Langley, Simon, Bradshaw, & Zytkow, 1987; Qin & Simon, 1990; Valdez-Perez, Simon, & Murphy, 1992). But psychologists were not the first, nor the only, scientists to argue for the ultimate knowability of the process of scientific discovery. More than 50 years ago, Einstein wrote: “The whole of science is nothing more than a refinement of every day thinking” (Physics & reality, 1936, reprinted in Einstein, 1950, p. 59). He also wrote, “The scientific way of forming concepts differs from that which we use in our daily life, not basically, but merely in the more precise definition of concepts and conclusions; more painstaking and systematic choice of experimental material, and greater logical economy” (“The common language of science,” 1941, reprinted in Einstein, 1950, p. 98).
