To be confirmed
Venue: centre Broca
Speaker: Laura Grima,
Dudman Lab at HHMI’s Janelia Research Campus in Ashburn, Virginia
Invited by Mathieu Wolff
Title: Dynamics of learning in many option foraging
Brief bio: Laura Grima is a Research Scientist in the Dudman Lab at HHMI’s Janelia Research Campus in Ashburn, Virginia. Prior to this position, she completed both a Master’s and DPhil (PhD) at the University of Oxford, with her PhD focusing on the role of mesolimbic dopamine and activity at striatal receptor subtypes on reward-driven action. She became interested in foraging behaviours during a brief postdoctoral position in Oxford, where she found that slow changes in ventral striatal dopamine levels reflected changing environment quality in mice making decisions in a novel operant chamber foraging task. In 2019 she moved to Janelia to study the neural mechanisms underlying foraging decisions in large spatial environments, particularly in the context of learning ‘from scratch’.
Current research interests: I am broadly interested in discovering the neural basis of the computations that underlie animals’ ability to rapidly learn and make effective decisions in complex, changing environments. My approach is to develop scalable, interactive environments that leverage the natural behavioural repertoire of mice, together with development of ethologically-inspired computational models to enable detailed characterization of how complex decision strategies emerge and change through learning. These models can also guide interpretation of neural data, and I am particularly interested in dopamine-hippocampal interactions that may support the formation of structured representations for decision-making over learning. Importantly, this environment and computational framework are readily translatable across species, providing a powerful avenue for advancing our understanding of human decision-making disorders such as ADHD.
Abstract: In natural environments, animals must effectively allocate their choices across many concurrently available resources when foraging. This is a complex decision-making process not fully understood, nor well captured by existing models. In this talk I will describe a novel paradigm that we developed in which completely naïve mice were free to sample and learn about six options of differing quality positioned around the walls of a large (~2m) arena. Mice exhibited rapid learning, matching their choices to integrated reward ratios across all six options within tens of minutes. To develop a mechanistic description of this learning, we constructed a reinforcement learning model inspired by foraging theory. In combination with a dynamic, global (across all options) learning rate, this model was able to accurately reproduce mouse learning and decision-making. Finally, I will discuss results of fiber photometry recordings and optogenetic manipulations of dopamine levels in the nucleus accumbens core (NAcC), revealing a unique role of this signal in implementing the global learning rate. Altogether, our results provide insight into the neural substrate of a learning algorithm that allows mice to rapidly exploit multiple options when learning to forage in large spatial environments.