Rava Azeredo da Silveira, Head, Theoretical & Computational Neuroscience Group, Institute of Molecular and Clinical Ophthalmology Basel, will give a presentation via Zoom in this week’s Social, Economic, and Decision Psychology research seminar (Thursday 3 June).

Cognitive biases and costly mental representations

While the faculty of rational thinking defines, at least to an extent, our human nature, it suffers from a remarkably long list of so-called “cognitive biases”—systematic deviations from rational information processing. Human behavior is also variable, even when an optimal observer would behave in a deterministic fashion. Biases and variability are particularly salient in situations in which humans update their beliefs as a function of a stream of stochastic observations, and cannot be explained on the basis of a sound, Bayesian manipulation of probabilities. In this talk, I will introduce a theoretical framework in which biases and variability emerge from a trade-off between Bayesian inference and the cognitive cost of carrying out probabilistic computations. After discussing the theoretical implications of this framework, I will present data from behavioral experiments in humans and their analysis in the context of our framework. If time allows, I will contrast this approach and predictions with those of an alternative framework in which cognitive costs impose limits on the precision of memory. The various theoretical approaches I will outline point to the fact that biases and variability in human cognition can reflect optimality under constraints—”resource-rational cognition”—rather than ad hoc erroneous beliefs or heuristics.