Models Are Stupid, and We Need More of Them

Despite numerous efforts extolling the virtues of formal modeling (Epstein 2008; Schank 2001; Smith & Conrey 2007; Marewksi & Olsson 2009; Farrell & Lewandowsky 2010; Weinhardt & Vancouver 2012; Smaldino et al. 2015), there remains widespread resistance among social and behavioral scientists to adopt formal modeling in their general research approach. In addition to the technical challenge posed by the mathematical and programming skills required to understand and develop models, a common point of resistance appears to stem from the perception of models as crude, overly simplistic, and unrealistic. The conclusion is that models are largely useless as anything but a formal exercise, and unnecessary for most scientists to engage with.