The hippocampus is often described as the brain’s internal map.
But people and animals do not always navigate using the same kind of map.
Sometimes navigation depends on stable locations in the wider environment. At other times, it depends on instructions relative to the body, such as turning left from where you are standing.
New research suggests that the hippocampus handles these two strategies differently.
A study published in Nature Communications on 24 September 2026 trained male rats on two carefully matched versions of an everyday memory task. One required a world-centred, or allocentric, navigation strategy. The other required a self-centred, or egocentric, strategy.
The researchers recorded activity from hippocampal CA1 neurons using miniature calcium-imaging microscopes while the rats navigated.
They found that the hippocampus represented space in both versions of the task, but the representations were less spatially precise when the rats used the self-centred strategy.
When the researchers temporarily suppressed hippocampal activity during the decision-making phase, performance deteriorated in rats using the world-centred strategy but not in rats using the self-centred strategy.
The findings suggest that hippocampal activity becomes especially important when an animal must use a stable representation of the external world to plan where to go next.
There is more than one way to know where you are going
Navigation can rely on different frames of reference.
In allocentric navigation, locations are represented relative to the external environment.
A person using this strategy might remember that a café is north of a park or that a particular building lies beside a river, regardless of the direction they are currently facing.
In egocentric navigation, locations are represented relative to the navigator’s own body or viewpoint.
A route might instead be remembered as a sequence such as turn left, continue straight and then turn right.
Both strategies can lead to the same destination, but they place different demands on memory.
The new study was designed to compare these strategies while keeping the physical environment as similar as possible.
The rats searched for food in the same large arena
The behavioural task took place in a large square event arena containing several sandwells where food rewards could be hidden.
The rats learned which sandwell was rewarded and then had to use that memory to make later choices.
The important manipulation was how the correct location had to be represented.
For the allocentric task, rats started from different positions around the arena and needed to identify the same goal location relative to the external environment.
Because the starting point changed, a fixed body-centred turning rule was not enough.
The animal had to know where the goal was in the wider space.
For the egocentric task, the training was structured so that the rats could solve the problem using a stable relationship between their starting position and the required action.
This encouraged a self-centred strategy rather than one anchored to distant environmental landmarks.
Miniature microscopes recorded activity from hippocampal CA1 cells
The researchers used miniature microscopes to monitor calcium activity in neurons within the dorsal CA1 region of the hippocampus while the rats moved freely.
Calcium imaging does not measure electrical spikes directly.
Instead, it uses fluorescent signals associated with changes in intracellular calcium as an indicator of neuronal activity.
This allowed the team to follow large numbers of hippocampal neurons during natural movement through the arena.
The hippocampus contains place cells whose activity is associated with particular locations.
These cells are a major reason the hippocampus is linked to cognitive maps and spatial memory.
The study asked whether the quality and structure of these spatial representations depended on the navigation strategy being used.
Spatial representations were less precise during self-centred navigation
The hippocampus encoded spatial information in both groups.
However, the representations differed in precision.
Rats using the egocentric strategy showed less precise spatial representations than rats using the allocentric strategy.
This is important because it argues against a simple interpretation in which the hippocampus participates only in world-centred navigation.
The hippocampus remained active when animals used a self-centred strategy, but the spatial code was organised differently.
The navigation rule therefore influenced the form of the hippocampal representation rather than merely determining whether the hippocampus was active.
The largest difference appeared before the rats entered the arena
The researchers were particularly interested in what happened during planning.
Before each navigation trial, the rat waited briefly in a startbox at the edge of the arena.
During this period, hippocampal neurons associated with locations elsewhere in the arena could become active even though the animal was still physically inside the startbox.
This kind of non-local activity is often interpreted as the brain representing places beyond the animal’s immediate location.
The patterns differed between rats trained to use allocentric and egocentric strategies.
In the allocentric group, remote spatial representations during planning were more closely linked to potential destinations in the external environment.
This raised the possibility that the hippocampus was not simply recalling where the rat had been, but contributing to the representation of where it might go.
Some hippocampal activity appeared to represent future choices
The researchers found evidence that hippocampal activity before movement could be related to the location that the animal was about to choose.
This was especially apparent in rats using the allocentric strategy.
The finding fits with a broader body of research suggesting that hippocampal activity can represent possible future routes and destinations during navigation.
Such activity does not necessarily mean that a rat consciously imagines a route in the way a human might describe doing so.
It does show that neural representations of locations can become active before the animal physically travels to them.
The crucial question was whether this activity was merely correlated with navigation or actually necessary for successful choice.
The researchers temporarily suppressed hippocampal activity
To test causality, the study included a separate optogenetic experiment.
Optogenetics allows researchers to alter the activity of genetically targeted neurons using light.
The team temporarily suppressed hippocampal activity during the period when the animals were preparing to make a navigational decision.
This manipulation made it possible to ask whether the hippocampal representations seen during planning were functionally required.
If the activity were only a by-product of navigation, suppressing it should not necessarily alter performance.
If it were required for choosing the correct destination, performance should deteriorate when the hippocampus was suppressed.
World-centred navigation was selectively disrupted
The effect depended on the strategy the rats were using.
Suppressing hippocampal activity impaired goal identification in rats solving the allocentric version of the task.
The same manipulation did not produce the equivalent impairment in rats using the egocentric strategy.
This provides stronger evidence than brain imaging alone because the researchers directly altered hippocampal activity and observed a strategy-specific change in behaviour.
The result supports a causal role for the hippocampus in planning when successful navigation depends on a world-centred representation of where the goal lies.
Self-centred navigation can rely on other brain systems
The fact that hippocampal suppression did not disrupt the egocentric task does not mean the task required no memory.
It suggests that the required information could be supported sufficiently by other neural systems when the animal could rely on a body-centred response strategy.
Previous research has associated habitual response strategies and body-centred action rules with structures including the striatum.
Navigation therefore appears to be distributed across several interacting brain systems rather than controlled by one universal spatial circuit.
Which system becomes essential depends partly on how the problem is framed.
The same physical destination can require different neural computations
One of the most important implications of the study is that behavioural success alone can hide differences in the strategy used to reach that success.
Two animals may reach the same sandwell and collect the same reward.
One may solve the problem by locating the target within a stable map of the external environment.
The other may solve it using a learned relationship between its own starting position and the direction it should travel.
From the outside, the behaviour can look similar.
Inside the brain, the computational problem is different.
The new findings show that this difference is reflected in the hippocampal spatial code and in how necessary the hippocampus becomes for making the correct choice.
This could help explain why navigation strategies vary between people
Humans also use a mixture of world-centred and self-centred navigation.
Someone familiar with a city may build a flexible mental map and find several routes to the same place.
Someone else may rely on a memorised sequence of turns that works well until the usual route is blocked.
These strategies are not inherently good or bad.
A self-centred route can be efficient and cognitively simple when the environment is familiar.
A world-centred representation becomes especially valuable when a person needs to take a shortcut, approach a destination from a new direction or adapt when a familiar route changes.
The rat study provides a mechanistic framework for understanding why those strategies may depend differently on the hippocampus.
The study may also matter for memory research
The hippocampus is important for more than navigation.
It also plays a major role in episodic memory, which allows experiences to be remembered with information about what happened, where it happened and when it happened.
The everyday memory task used in this study was developed to capture some of these features in rodents.
The finding that future-relevant spatial representations change with navigation strategy suggests that memory retrieval is not simply a fixed replay of stored information.
How a memory is going to be used may influence which parts of a spatial representation become important.
This supports a view of memory as a flexible system that contributes to planning as well as remembering.
The work was conducted in male rats
The animal model is an important limitation.
The experiments were conducted in male rats.
Rodent hippocampal systems share important organisational principles with mammalian spatial-memory systems more broadly, but rat behaviour is not a direct model of every aspect of human navigation.
Human navigation is influenced by language, conscious strategies, maps, digital navigation tools and extensive cultural learning.
The study therefore cannot show that temporarily reducing hippocampal activity in humans would produce exactly the same pattern.
It provides evidence about a neural mechanism that can be tested further across species.
The study compared trained strategies rather than spontaneous preferences
Another limitation is that the researchers deliberately trained the animals to rely on different reference frames.
In everyday life, animals and humans can switch between strategies or combine them.
A person may initially follow a memorised route, then use landmarks when the route becomes unfamiliar.
The controlled design was useful because it allowed the researchers to isolate the effect of navigation strategy.
However, it simplifies the more fluid way spatial strategies are selected outside an experiment.
Calcium imaging is an indirect measure of neural firing
The imaging method also has trade-offs.
Miniature calcium microscopes allow many neurons to be recorded while an animal moves freely, which is a major advantage for naturalistic navigation research.
But calcium signals change more slowly than individual electrical spikes.
This means the technique provides a broad picture of neuronal activity rather than the millisecond-level timing available from electrophysiological recordings.
The optogenetic manipulation strengthens the study because it provides a separate causal test of whether hippocampal activity was necessary for behaviour.
The bigger lesson is that a cognitive map is task-dependent
The hippocampus is often described as if it contains one stable internal map of the environment.
The new findings suggest a more flexible picture.
The way space is represented depends partly on what the animal needs to do with that information.
When navigation required a stable relationship between locations in the external world, hippocampal representations during planning became functionally necessary for choosing the correct goal.
When the task could be solved using a self-centred action rule, the hippocampus still represented space, but suppressing it did not produce the same behavioural failure.
The brain’s map of space therefore appears to be shaped not only by where an animal is, but also by the strategy it is using to decide where to go next.
Source Information
Study Title: Navigational strategy dictates hippocampal representation of space in an everyday memory task
Authors: Francesco Gobbo, Rufus Mitchell-Heggs, Adrian J. Duszkiewicz, Elena Faillace, Dorothy Tse, Nuria Garcia-Font, Omer Hazon, Kathrine Clarke, Patrick A. Spooner, Adana Keshishian, Mark Schnitzer, Simon R. Schultz and Richard G. M. Morris
Journal: Nature Communications
Published: 24 September 2026
Sample: Male rats trained on two controlled versions of an everyday memory task using either allocentric, world-centred navigation or egocentric, self-centred navigation. Separate experimental components used hippocampal calcium imaging and optogenetic inactivation.
Method: The researchers recorded dorsal CA1 activity with miniature calcium-imaging microscopes while freely moving rats performed matched spatial-memory tasks in the same arena. They compared the precision and planning-related structure of hippocampal representations across navigation strategies. In a separate experiment, hippocampal activity was temporarily suppressed with optogenetics during navigational decision-making to test whether it was required for successful goal choice.
Main finding: Hippocampal spatial representations differed according to navigational strategy and were less precise during egocentric navigation. Temporarily suppressing hippocampal activity during planning impaired goal identification in the allocentric task but not the egocentric task, supporting a causal role for planning-related hippocampal representations in world-centred navigation.
DOI: 10.1038/s41467-026-77885-3
