Abstract
Is there more to ‘good science’ than explaining novel facts? Social interaction within the scientific community plays a pivotal role in defining acceptable research. This paper develops an agent-based computational model of scientific communities to explore the connection between research outcomes and the socio-cultural environment they are constructed in. Agent-to-agent interaction is added to the notion of scientific research programs (SRPs) developed by Lakatos (1969, 1970) as an important factor guiding the practice of researchers. Early in scientific inquiry, when there are many novel facts to explain, the research community is fragmented and researchers prefer to use their own talents (or innovation) when conducting research. Over time, consensus emerges in the form of a dominate SRP which utilizes intra-disciplinary approaches (or adaptation). This result illustrates the rise and fall of scientific paradigms as described by Kuhn (1970). The history of Keynesian macroeconomic theory provides colloquial support for the simulation model.