The concept of scientific discovery has attracted much attention among philosophers. Various approaches have been developed, ranging from sociological theories of knowledge construction to detailed models of how scientific concepts or hypotheses are generated and analyzed in the course of scientific practice. Some approaches have emphasized the role of random variations in thought processes that generate ideas or hypotheses but fail to clarify how new ideas are first formed. Others have emphasized the importance of the initial, non-analyzable, intuitive leap that leads to a scientific discovery.
The most widely accepted view of the process of discovery is a paradigm-driven paradigm: Scientific discoveries happen when anomalous phenomena are observed, they are tentatively conceptualized as a theory, and they lead to experiments and observations that can be tested to see whether they can confirm the predictions of the theory. This approach has a number of important advantages, but it cannot provide a precise explanation of how new ideas or hypotheses are first formed.
Another approach to the logic of discovery is based on a general theory of problem solving. These approaches treat the adoption, articulation, and preliminary evaluation of a scientific hypothesis as a problem-solving activity and analyze the heuristic strategies used in this activity. This approach is often linked to pragmatist ideas and to Peirce’s version of the logic of abductive inference. In this context, scientific research is viewed as an extended, iterative, and risky activity in which the goal is not to find an objectively “true” but empirically compelling solution (Wimsatt 2007). It has also motivated studies of artificial intelligence at the interface between philosophy of science and cognitive science.
About the Author
