A Vector Navigation and Inference Architecture can Construct Universal Cognitive Maps for Abstract Reasoning

The idea of cognitive maps in the hippocampal formation has its origins in decades of spatial cognition research. However, increasing evidence points to cognitive maps also supporting non-spatial tasks. It appears that the hippocampal-entorhinal system can map any combination of systematically varying stimulus dimensions onto the same neural manifolds that support spatial representations - grid cells. Here I propose a model that shows how spatial navigation architectures can iteratively construct universal cognitive maps. The model is neurally plausible and exhibits noise-tolerance that implies specific behavioral predictions. The mapping process accommodates discontinuous stimulus spaces and the model can be used to support several types of abstract reasoning, including analogy making, subspace construction, and perspective taking. Importantly, these capabilities are supported by the same neural processes that are used to build a map. Thus, the combination of mechanisms at play suggests how spatial navigation architectures can function as domain-general substrates of cognition.