Frugal preference formation

Most theories explaining how animals form preferences for their actions agree upon a basic outline: animals discover what is preferable through interactions with the world, store this information in memory, and recall it to help them decide what to do in a new situation. However, no single theory currently explains both how preferences are learned, and how they are recalled in a way that is compatible with empirical data. We advance precisely such a proposal in the form of a stochastic choice model where the decision agent learns what to do based on scale-free comparisons between options it observes in the world and at each decision instance recalls a subset of these comparison experiences in a manner that minimizes the metabolic costs of memory recall. In simulation, this model makes qualitatively accurate predictions connecting agent choices with various dynamic choice correlates documented in the literature on choice process models.