Proceedings of the Annual Meeting of the Cognitive Science Society

In his famous address “You can’t play 20 questions with nature and win”, Newell (1973) predicted that experimental psychology’s focus on falsifying theories of individual phenomena (e.g., subitizing, directed forgetting, etc.) would not lead to a unifying theory subsuming several (let alone all) of the phenomena. Instead of seeking to detect effects, and binary searching our way to the explananda of these phenomena, Newell argued psychologists should build predictive theories of behavior by modeling 1) the structure of the task environment (i.e., context), 2) the subjects’ own goals, and 3) the invariant structure of the subjects’ processing mechanisms. Newell concludes that psychology will only make real progress in understanding the mind by 1) targeting tasks that are complex enough to cover the space of naturalistic behavior, 2) developing models that can competently perform the task, and 3) generalizing such models to perform multiple complex tasks. For 1), Newell chose to study chess; we believe he would find video games an even more compelling paradigm (see Gobet, 2017).