Schedule of Events | Symposia

Sketchpad Series

Investigating the role of neural replay in multi-agent navigation

Poster Session B - Sunday, March 30, 2025, 8:00 – 10:00 am EDT, Back Bay Ballroom/Republic Ballroom

Max A. B. Hinrichs^1,2, Nicholas Menghi², Yangwen Xu², Nicolas W. Schuck^3,4, Christian F. Doeller^1,2,5; ¹Max Planck School of Cognition, Leipzig, Germany, ²Max Planck Institute for Human Cognitive and Brain Sciences, Leipzig, Germany, ³Institute of Psychology, Universität Hamburg, Hamburg, Germany, ⁴Max Planck UCL Centre for Computational Psychiatry and Ageing Research, Berlin, Germany, ⁵Kavli Institute for Systems Neuroscience, Centre for Neural Computation, The Egil and Pauline Braathen and Fred Kavli Centre for Cortical Microcircuits, Jebsen Centre for Alzheimer’s Disease, Norwegian University of Science and Technology, Trondheim, Norway

Replay of hippocampal place cell sequences has been suggested to be involved in memory consolidation and path planning. Recently, methodological advancements enabled the detection of replay-like sequences in humans using fMRI. While previous replay studies investigated individual subjects navigating alone, humans rarely operate in isolation. Instead, adaptive decision-making often requires taking others’ trajectories into account, for instance, to avoid collision in physical space. Nevertheless, whether replay extends to others, prospectively simulating their future trajectories at decision points and retrospectively replaying their past trajectories is unclear. Here, in a novel experimental paradigm, human participants learn how images are related in a state space where possible transitions between states follow an underlying graph structure. Subsequently, pairs of participants navigate the state space in an interactive setting, trying to reach a common goal state while avoiding collisions, while their brain activity is measured using fMRI hyperscanning using two state-of-the-art Siemens 3T MR scanners in parallel. Replay-like events will be identified as the sequential reactivation of state representations by multivariate fMRI classification. Preliminary behavioral results show that participants acquired a representation of the state space and used this knowledge to navigate in the interactive setting successfully. fMRI data will be analyzed to test the hypotheses that the hippocampal formation represents the state space and that neural replay underlies successful multi-agent navigation by prospectively simulating self and other’s future trajectories to avoid collision and retrospectively replaying their past ones to assign credit. Our study will shed light on the role of replay in multi-agent navigation.

Topic Area: THINKING: Decision making