Date: Wednesday, 17.05.23 17:00 CET
Location: Building S1|03 Room 223
Abstract:
Agent-based behavioural modelling (ABM) is a current research paradigm that combines concepts from AI and operations research with behavioural experimentation in a probabilistic inference framework. The fundamental aim of ABM is to understand the computational and algorithmic underpinnings of human cognition, primarily in the context of sequential decision making under uncertainty. In this talk, I will review our recent work on the conceptualization and instantiation of the ABM framework. I will introduce the framework’s central partition into task, agent, and behavioural models and exemplify the framework’s application in the context of perceptual choice, state-action-reward contingency learning, and, time permitting, simulated goal-directed spatial navigation. Throughout, I will highlight open questions and challenges facing the framework and point to its applicability in AI research and computational psychiatry.
