Do virtual reality navigation tasks and neuroimaging converge on the same hippocampal and caudate contributions to cognitive map formation?

Why navigation seems to run on two systems at once

The hippocampus is the structure most associated with building a flexible map of space — a representation you can use to find novel shortcuts. The caudate nucleus, part of the brain's habit and reward circuitry, is more associated with learning a fixed route through repeated practice. The useful point for a general reader is that these are not rival theories of navigation; they are two strategies the brain can run in parallel, and the studies describe an interaction between hippocampal- and caudate-mediated spatial strategies [2].

That framing is what makes virtual reality useful. A VR navigation task can be designed so that a person either has to build a map-like sense of the environment or can get by with a learned route, and brain imaging taken during the task can then be compared against which strategy the person appears to be using. Work in older adults with mild cognitive impairment explicitly examines the roles of both the hippocampus and the caudate nucleus in navigation alongside prefrontal cortex activity, which fits this two-system picture [1].

What the scans actually show during navigation

In a study of people with amnestic mild cognitive impairment — a memory-affected condition often preceding Alzheimer's disease — researchers used a virtual reality navigation task inside an fMRI scanner, described as the first such study in this group, and framed the results around the interaction between hippocampal- and caudate-mediated spatial strategies [2]. That is a direct case of VR navigation and neuroimaging being used together to ask about the same two structures.

Other navigation imaging work is broader and less specific. An fMRI study of adolescents and adults navigating a virtual reality environment mapped activity across many regions, including the caudate as well as the cerebellum, frontal cortex and cingulate gyrus, rather than isolating a hippocampal-versus-caudate contrast [3]. A study of people with schizophrenia likewise found activation changes extending to the caudate head during navigation through a virtual town, but its focus was hippocampal function, not the division of labor between the two structures [5]. In other words, the caudate shows up in navigation scans fairly often, but often as one region among many rather than as a tested partner to the hippocampus.

Where VR tasks and imaging line up — and where they don't

The convergence is real at the level of anatomy: multiple studies using VR navigation with brain imaging name the hippocampus and caudate as navigation-relevant structures [1][2], and caudate activation appears in navigation scans even when it is not the main question [3][5]. A scoping review of 29 neuroimaging studies of VR-based cognitive assessment games also reports that navigators showed greater right caudate head activation, tying VR task performance to imaging signals [4].

The gap is in what is being measured. Most of these designs record where activity rises during navigation; they do not directly demonstrate the formation of a cognitive map, and several are framed around clinical groups — mild cognitive impairment or schizophrenia — rather than healthy map-building [1][2][5]. The review connecting VR game performance to hippocampal health is explicitly about hippocampal vulnerability and does not examine the caudate's role or cognitive map formation [4]. So the two approaches agree on the cast of brain regions, but the evidence is thinner for the stronger claim that they jointly explain how a cognitive map is built.

How to read a claim that VR navigation reveals the brain's map

Treat region-level agreement as the solid finding and strategy-level interpretation as the more tentative one. When a study says the hippocampus and caudate mediate different spatial strategies and interact [2], that is a framework the imaging is consistent with, not a demonstration that one structure builds the map and the other stores the route. Similarly, findings in clinical groups tell you about navigation under memory impairment or illness, and should not be assumed to describe healthy map formation without qualification [1][5].

The most defensible summary is that VR navigation and neuroimaging converge on the same hippocampal and caudate contributions at the level of which structures are engaged, while the precise division of labor in forming a cognitive map remains something these studies point toward rather than settle [2][3][4].

Sources used in this answer

Related research questions