The Architecture of Propositional Thought: From Cognitive Maps to Language and Reasoning

One of the most promising routes to understanding the architecture of propositional thought—and cognition more broadly—is to begin with the mammalian navigation system and its associated neural machinery (cognitive maps, place cells, grid cells, path integration, sequence generation, hippocampal replay, and related control mechanisms). Rather than treating navigation as a domain-specific specialization, this approach treats it as an evolutionarily ancient and well-characterized architectural template that may have been replicated, extended, and abstracted to support a wide range of cognitive capacities.

A growing body of work suggests that the hippocampus–entorhinal cortex system is not merely a spatial navigation system, but a general system for managing structured representations and sequences, supporting planning, memory, inference, imagination, and offline simulation. At the same time, complementary research highlights the role of frontoparietal control systems in maintaining, selecting, and flexibly manipulating these representations under task demands. Together, these findings motivate an architectural perspective in which navigation-like mapping and replay mechanisms interact with control, valuation, imagery, and language to support higher cognition.

Accordingly, the guiding idea behind this conference is to explore whether propositional thought itself—across linguistic and non-linguistic forms—can be understood as emerging from the reuse and extension of navigation-like architectures for offline control, simulation, and problem solving.

The conference is also aimed at producing an edited volume on the same topic and foster future research collaborations among participants.