Wednesday 10th May 2017
Ground Floor Seminar Room
25 Howland Street, London, W1T 4JG
Mechanisms of adaptive inference
My laboratory studies the computational and neural substrates of adaptive inference; that is, our ability to use experience to shape how we draw conclusions from uncertain data in dynamic and changing environments. In this talk, I will first describe how expectations about environmental dynamics affect the process of evidence accumulation that is central to many sequential-sampling models of decision-making. Second, I will describe ongoing projects that examine how these expectations are learned, focusing on individual variability across human subjects. Third, I will describe recent results that are beginning to shed light on some of the underlying neural mechanisms, particularly those involving interactions between arousal and cognitive systems. I will end with some speculations on possible computational principles that describe the role of arousal in adaptive inference.