This talk is based on ongoing work with Torben Brauner and Irina Polyanskaya at the University of Roskilde, Denmark. We have experimental results for children with Autism Spectrum Disorder (ASD) which show that mastery of linguistic recursion is a significant predictor of success in second-order false belief reasoning tasks. We also have experimental results which suggest that children with ASD acquire the type of reasoning induced by the Danish discourse particles “Jo”, “Da” and “Vel” rather differently. This is somewhat surprising, as the two types of reasoning at first glance seem very similar: both involve sophisticated perspective-taking skills of the sort that underlie what cognitive and developmental psychologists call Theory of Mind.

In this talk (which touches on logic, cognitive science, and pragmatics) I will first discuss what first- and second-order false belief tasks are, and what we know about them logically and empirically. I will then go on to discuss the JoDaVel reasoning task (introduced by Elisabeth Engberg-Pedersen and Ditte Boeg Thomsen) again from both logical and experimental perspectives. I won’t assume any particular background in either logic, cognitive science or pragmatics; my aim in this talk is to give a coherent interdisciplinary overview of our work, and to clarify some of the conceptual issues suggested by the results.