Building and Using a Cognitive Map
The hippocampus is thought to build a cognitive map that supports navigation, memory, and planning, but what defines such a map and how it is used remain debated. In this talk, I will present computational models in which hippocampal-like representations emerge in recurrent neural networks trained to predict sequences of sensory observations. While spatially tuned units reliably arise, they are not sufficient to form a cognitive map. Instead, map-like representations emerge when recurrent dynamics support multi-step prediction, yielding a population-level encoding of environmental geometry. Once learned, these representations can autonomously generate offline trajectories biased by recent experience, capturing key features of hippocampal replay. I will then show how these representations guide behavior in navigation tasks. In a hippocampal–striatal model facing visual ambiguity, access to hippocampal activity enables rapid learning and flexible adaptation. Place-like coding supports self-localization, while population-level hippocampal states can be used to derive intrinsic learning signals that estimate progress toward a remembered goal, improving performance beyond full sensory observability. Together, these results suggest that cognitive maps arise from predictive recurrent dynamics and support behavior through both localization and internally generated learning signals. Presented in the van Vreeswijk Theoretical Neuroscience Seminar series (formerly WWTNS) on 2026-01-21. Recording duration: 00:46:20.
A recurrent network model of planning predicts hippocampal replay and human behavior
When interacting with complex environments, humans can rapidly adapt their behavior to changes in task or context. To facilitate this adaptation, we often spend substantial periods of time contemplating possible futures before acting. For such planning to be rational, the benefits of planning to future behavior must at least compensate for the time spent thinking. Here we capture these features of human behavior by developing a neural network model where not only actions, but also planning, are controlled by prefrontal cortex. This model consists of a meta-reinforcement learning agent augmented with the ability to plan by sampling imagined action sequences drawn from its own policy, which we refer to as `rollouts'. Our results demonstrate that this agent learns to plan when planning is beneficial, explaining the empirical variability in human thinking times. Additionally, the patterns of policy rollouts employed by the artificial agent closely resemble patterns of rodent hippocampal replays recently recorded in a spatial navigation task, in terms of both their spatial statistics and their relationship to subsequent behavior. Our work provides a new theory of how the brain could implement planning through prefrontal-hippocampal interactions, where hippocampal replays are triggered by -- and in turn adaptively affect -- prefrontal dynamics.
A recurrent network model of planning explains hippocampal replay and human behavior
When interacting with complex environments, humans can rapidly adapt their behavior to changes in task or context. To facilitate this adaptation, we often spend substantial periods of time contemplating possible futures before acting. For such planning to be rational, the benefits of planning to future behavior must at least compensate for the time spent thinking. Here we capture these features of human behavior by developing a neural network model where not only actions, but also planning, are controlled by prefrontal cortex. This model consists of a meta-reinforcement learning agent augmented with the ability to plan by sampling imagined action sequences drawn from its own policy, which we refer to as 'rollouts'. Our results demonstrate that this agent learns to plan when planning is beneficial, explaining the empirical variability in human thinking times. Additionally, the patterns of policy rollouts employed by the artificial agent closely resemble patterns of rodent hippocampal replays recently recorded in a spatial navigation task, in terms of both their spatial statistics and their relationship to subsequent behavior. Our work provides a new theory of how the brain could implement planning through prefrontal-hippocampal interactions, where hippocampal replays are triggered by - and in turn adaptively affect - prefrontal dynamics.
Canonical neural networks perform active inference
The free-energy principle and active inference have received a significant attention in the fields of neuroscience and machine learning. However, it remains to be established whether active inference is an apt explanation for any given neural network that actively exchanges with its environment. To address this issue, we show that a class of canonical neural networks of rate coding models implicitly performs variational Bayesian inference under a well-known form of partially observed Markov decision process model (Isomura, Shimazaki, Friston, Commun Biol, 2022). Based on the proposed theory, we demonstrate that canonical neural networks—featuring delayed modulation of Hebbian plasticity—can perform planning and adaptive behavioural control in the Bayes optimal manner, through postdiction of their previous decisions. This scheme enables us to estimate implicit priors under which the agent’s neural network operates and identify a specific form of the generative model. The proposed equivalence is crucial for rendering brain activity explainable to better understand basic neuropsychology and psychiatric disorders. Moreover, this notion can dramatically reduce the complexity of designing self-learning neuromorphic hardware to perform various types of tasks.
Efficient reuse of computations in planning
Solving complex planning problems efficiently and flexibly requires reusing expensive previous computations. The brain can do this, but how? I present a new theory that addresses this question and connects planning to hitherto distinct areas within cognitive neuroscience, such as entorhinal representation of cognitive maps and cognitive control.
Peril, Prudence and Planning as Risk, Avoidance and Worry
Risk occupies a central role in both the theory and practice of decision-making. Although it is deeply implicated in many conditions involving dysfunctional behavior and thought, modern theoretical approaches to understanding and mitigating risk in either one-shot or sequential settings, which are derived largely from finance and economics, have yet to permeate fully the fields of neural reinforcement learning and computational psychiatry. I will discuss the use of dynamic and static versions of one prominent approach, namely conditional value-at-risk, to examine both the nature of risk avoidant choices, encompassing such things as justified gambler's fallacies, and the optimal planning that can lead to consideration of such choices, with implications for offline, ruminative, thinking.
Geometry of Neural Computation Unifies Working Memory and Planning
Cognitive tasks typically require the integration of working memory, contextual processing, and planning to be carried out in close coordination. However, these computations are typically studied within neuroscience as independent modular processes in the brain. In this talk I will present an alternative view, that neural representations of mappings between expected stimuli and contingent goal actions can unify working memory and planning computations. We term these stored maps contingency representations. We developed a "conditional delayed logic" task capable of disambiguating the types of representations used during performance of delay tasks. Human behaviour in this task is consistent with the contingency representation, and not with traditional sensory models of working memory. In task-optimized artificial recurrent neural network models, we investigated the representational geometry and dynamical circuit mechanisms supporting contingency-based computation, and show how contingency representation explains salient observations of neuronal tuning properties in prefrontal cortex. Finally, our theory generates novel and falsifiable predictions for single-unit and population neural recordings.