Animal Minds published by Cambridge University Press.

Animal minds are complex and diverse, making them difficult to study. This Element focuses on a question that has received much attention in the field of comparative cognition: Do animals reason about unobservable variables like force and mental states?

The Element shows how researchers design studies and gather evidence to address this question. Despite the many virtues of current methods, hypotheses in comparative cognition are often underdetermined by the empirical evidence. Given this, philosophers and scientists have recently called for additional behavioral constraints on theorizing in the field. The Element endorses this proposal (known as 'signature testing'), while also arguing that studies on animal minds would benefit from drawing more heavily on neuroscience and biology.

Video Abstract

Are current AI as smart as crows or honey bees? The Animal-AI Testbed draws on work in animal cognition research to test the capabilities of AI. Using this platform, we have shown that AI and children perform similarly on basic navigational tasks, but children perform better on more complex cognitive tests like those requiring object permanence. See our recent arXiv paper for details: Animal-AI 3: What's New & Why You Should Care.

Photo by Elsa Loissel