Cognitive science as a source of forward and inverse models of human decisions for robotics and control

Those designing autonomous systems that interact with humans will invariably\nface questions about how humans think and make decisions. Fortunately,\ncomputational cognitive science offers insight into human decision-making using\ntools that will be familiar to those with backgrounds in optimization and\ncontrol (e.g., probability theory, statistical machine learning, and\nreinforcement learning). Here, we review some of this work, focusing on how\ncognitive science can provide forward models of human decision-making and\ninverse models of how humans think about others' decision-making. We highlight\nrelevant recent developments, including approaches that synthesize blackbox and\ntheory-driven modeling, accounts that recast heuristics and biases as forms of\nbounded optimality, and models that characterize human theory of mind and\ncommunication in decision-theoretic terms. In doing so, we aim to provide\nreaders with a glimpse of the range of frameworks, methodologies, and\nactionable insights that lie at the intersection of cognitive science and\ncontrol research.\n