The evolutionary background to (mis)understanding an uncertain world

Abstract

Misunderstandings of causality are often referred to as superstitions. More formally, superstitious behaviours can be defined as actions (or inactions) that are performed in order to increase the probability that a beneficial outcome arises when there is no causal relationship between the action and the outcome. While superstitious behaviours are common in humans, they also arise in non-human animals. Although behaving superstitiously may on first reflection appear always maladaptive, recent models have shown that superstitions will readily arise as a by-product of adaptive learning, in which individuals seek to balance gaining new information about the world with exploiting their current information. In short, if a behavior appears associated with a beneficial outcome, it may not be worthwhile experimenting and losing out on this benefit to determine whether the association has arisen by chance. The models help explain why superstitions get started, and indicate the types of superstitious behaviours that are likely to persist. In support, empiricists have widely observed that superstitions are more likely to develop when the perceived benefit of adopting a behaviour is high compared to the cost of not adopting it and when the number of opportunities to test one’s understanding is low. Collectively, therefore, while superstitions are commonly presented as entirely irrational behaviours, they can actually represent a smart strategy, promoted by natural selection, in situations where causal relationships are uncertain.

Keywords:

• Superstition
• evolution
• animal behaviour
• Bayesian learning
• environmental risk

Acknowledgements

We thank Esther Eidinow for the invitation to contribute to this special issue, along with suggestions on an earlier draft of this paper. Kevin Abbott made extensive suggestions which greatly contributed to this work.

Disclosure statement

No potential conflict of interest was reported by the authors.