Scientific discovery is the process of bringing together previously known facts to create an idea or theory that explains those facts. This new explanation must be either a new result or a reconceptualization of existing results. Scientific discoveries can also be new techniques or devices for obtaining data or for doing science, or they can be novel methods for interpreting experimental or observational results. Scientific discoveries often have profound repercussions and are considered one of the key features of science that distinguishes it from other activities such as artistic creation, business innovation or engineering design.
Until the late 20th century, philosophers generally accepted that there are at least two ways to conceptualize discovery: the eureka moment narrowly construed as a non-analyzable leap of insight; or a pragmatist theory of knowledge generation that treats the idea of a discovery as an extended reasoning process that can be analyzed. More recently, many philosophers have embraced the pragmatist view of discovery and have developed elaborate theories of the mental and physical-practical operations involved in generating and articulating scientific ideas.
The most prominent philosophical theory of discovery is that a scientific hypothesis follows a distinctive logical pattern that is different from inductive logic or deductive logic. Philosophers working in this tradition, including Norwood Hanson, have argued that this logic of discovery is revealed through the method of abductive inference. This theory of discovery has been influential in research on artificial intelligence, where it is interpreted as a form of problem-solving activity that involves heuristic selective search.