Reasoning versus association in animal cognition Current controversies and possible ways forward

The study of animal cognition is rife with controversy, and among the most long-standing and most intensely debated controversies in the field is the question of the extent to which the behavior of nonhuman animals can be fully understood on the basis of purely associative principles, or whether some behaviors exhibited by animals necessitate the assumption of inferential capacities in animals that defy an associative explanation. Remarkably, the continuing debate on the topic seems to be spawning little genuine progress in terms of substantial accumulation of new, generally accepted insights. As an introduction to a special section of the Journal of Comparative Psychology on the topic, the present article outlines a number of reasons for the stalemate and suggests ways to refertilize the debate. In particular, we claim that progress will not come from the adoption of general principles like Morgan’s canon or the primacy of prediction over postdiction. Instead, emphasis should be placed on a careful analysis of what it is that different sides in the debate do and do not agree on and an increased willingness to engage in adversarial collaboration, in the spirit of a shared interest in furthering our understanding of animal behavior.