Bayesian brains and cognitive mechanisms:
Henry Brighton, Gerd Gigerenzer · 2008
Abstract The chapter considers the project of probabilistic rational analysis in relation to a particularly well-studied and simple heuristic, the Take the Best algorithm. The authors relate the tension between ‘rational’ and ‘algorithmic’ explanations of cognitive phenomena to the bias-variance dilemma in statistics. Where the cognitive system must make do with little data in relation to the complexity of the problem, they suggest that simple algorithmic models, rather than full-blown probabilistic calculations, may be the cognitive system's best strategy.