The hippocampus, the brain region most associated with memory and spatial navigation, exists in some form across virtually every vertebrate species alive today. Its evolutionary roots stretch back more than 520 million years to the earliest vertebrates, which carried a forerunner of the structure in their primitive forebrains.1PubMed Central. Representational specializations of the hippocampus in phylogenetic perspective From goldfish to homing pigeons to echolocating bats, the hippocampus has been shaped and reshaped by natural selection to meet the spatial and memory demands of radically different lifestyles. What makes it fascinating in a comparative context is how much it can tell us about the relationship between ecology and brain architecture.
An Ancient Map-Making Machine
The hippocampus did not appear suddenly in mammals. Experiments across goldfish, turtles, lizards, monkeys, and humans have converged on a consistent picture: the earliest vertebrates already possessed a hippocampal homologue that housed map-like representations of odors and sights encountered at different locations, including the order and timing in which those stimuli should appear during a journey.2PubMed Central. Representational specializations of the hippocampus in phylogenetic perspective In other words, the brain region that helps you remember where you parked your car was already doing something recognizably similar in jawless fish half a billion years ago.
Research on teleost fish (the bony fish that include most familiar freshwater and marine species) and reptiles reinforces this idea. The medial pallium of ancestral fish appears to have become specialized for processing complex spatial information, and that functional trait has been retained across every independent vertebrate lineage.3PubMed. Spatial memory and hippocampal pallium through vertebrate evolution: insights from reptiles and teleost fish The structure has been modified, expanded, and occasionally shrunk depending on the species, but its core job of encoding space and experience has remained remarkably stable.
Fish and Reptiles Have Their Own Versions
Non-mammalian vertebrates do not have a hippocampus that looks like the familiar curled seahorse shape found in a human brain. Instead, they have brain regions called pallial structures that are thought to be homologous, meaning they descend from the same ancestral tissue and perform overlapping functions.
In goldfish, the lateral pallium appears to be the key player. When researchers damaged this region, the fish lost the ability to solve spatial tasks they had previously learned, while damage to other pallial areas did not produce the same deficit. Training goldfish on a spatial task also triggered increased cellular activity specifically in the lateral pallium’s neurons, mirroring the kind of use-dependent changes seen in the mammalian hippocampus.4PubMed. Selective involvement of the goldfish lateral pallium in spatial memory The finding is striking because fish and mammals diverged hundreds of millions of years ago, yet the functional parallel is clear.
Turtles show a similar pattern. Damage to the medial cortex, the reptilian counterpart of the hippocampus, impairs the ability to use spatial strategies for navigation. Turtles with medial cortex lesions can still learn tasks when a visible cue marks the goal, but they struggle when they need to remember a location based on its relationship to surrounding landmarks. That selective impairment closely matches what happens in mammals and birds after hippocampal damage.5PubMed. Spatial and non-spatial learning in turtles: the role of medial cortex Lizards with medial cortex lesions also show slowed learning and altered navigation strategies, though they rely more heavily on non-spatial approaches than turtles do.6PubMed. Effects of medial and dorsal cortex lesions on spatial memory in lizards
Food-Storing Birds and the Expandable Hippocampus
Some of the most dramatic hippocampal adaptations in the animal kingdom belong to birds that cache food. Species like chickadees, nutcrackers, and jays hide thousands of seeds and nuts across a landscape and then retrieve them weeks or months later. This feat of spatial memory is reflected directly in their brain anatomy.
A study of 52 individuals across 35 species and subspecies of passerine birds found that food-storing species have a significantly larger hippocampal complex, relative to brain and body size, than species that do not store food.7PubMed Central. Hippocampal specialization of food-storing birds Among food-cachers, the difference is even more fine-grained: scatter-hoarding species, which distribute caches across many locations, have a proportionally larger hippocampus than larder-hoarding species that pile everything into one spot.8PubMed. Memory and the hippocampus in food-storing birds: a comparative approach The more locations a species has to remember, the bigger the brain region it devotes to remembering them.
What makes this even more interesting is that the hippocampus in these birds is not a fixed size. In black-capped chickadees, the hippocampal formation is largest in October, when food hoarding is at its most intense, and smaller during other months.9PubMed. Seasonal variation in hippocampal volume in a food-storing bird, the black-capped chickadee This seasonal swelling is accompanied by increased neurogenesis, the birth of new neurons, which peaks in autumn and winter alongside the demands of caching and retrieval.10PubMed Central. Seasonal hippocampal plasticity in food-storing birds
The timing of hippocampal change is not identical across all food-storing birds. In willow tits, relative hippocampal volume peaks in September, and males show a secondary, smaller peak in July that females lack.11PubMed Central. Seasonal changes in the hippocampal formation of hoarding and non-hoarding tits These sex-specific and species-specific rhythms suggest the hippocampus is tuned not just to the general need for spatial memory, but to the particular ecological calendar of each population.
Beyond Seed Caches: Navigation and Migration in Birds
Food storing is not the only avian behavior that enlists the hippocampus. Seasonal changes in the hippocampus also occur in brood parasites (birds like cowbirds that must locate host nests) and in migratory species, suggesting the structure responds to any behavior that places heavy demands on spatial cognition.12PubMed. Seasonal change in the avian hippocampus
Homing pigeons provide another window into avian hippocampal function. GPS-tracking studies have revealed that the pigeon hippocampus supports a navigational system based on familiar visual landmarks. While the birds rely on environmental odors for navigation over unfamiliar areas, the hippocampus appears critical for recognizing visual landscapes and using them to guide flight, potentially playing a role in visual-spatial perception itself.13PubMed. The avian olfactory system and hippocampus: Complementary roles in the olfactory and visual guidance of homing pigeon navigation The pigeon hippocampus, in other words, is not just a memory device; it helps the bird recognize where it is in real time.
How Bats Map Three-Dimensional Space
Most research on hippocampal place cells, the neurons that fire when an animal occupies a specific location, has been conducted on rodents running around flat surfaces. Bats upend the picture by flying through three-dimensional space, and their hippocampal activity reflects this.
Researchers recorded from individual hippocampal neurons in freely flying Egyptian fruit bats using a wireless neural telemetry system and found that place cells were active in confined three-dimensional volumes. In over 90% of the recorded neurons, all three spatial axes were encoded with similar resolution, meaning the bat’s mental map was roughly uniform in every direction rather than flattened along one axis.14PubMed. Representation of three-dimensional space in the hippocampus of flying bats The place fields from different neurons spanned different locations and collectively covered the entire available space in the flight room, painting a dense, volumetric map.
A follow-up study with big brown bats, a species that navigates by laryngeal echolocation rather than vision, confirmed 3D place cells in this species as well, without the continuous theta rhythm typically associated with spatial coding in rodents.15PubMed Central. 3D Hippocampal Place Field Dynamics in Free-Flying Echolocating Bats The absence of theta is itself noteworthy. In rats, theta oscillations are so tightly linked to spatial navigation that some researchers consider them essential to the process. Bats seem to manage without them, hinting that the hippocampal “code” for space is more flexible than the rat model alone would suggest.
Collecting these data required a creative setup: tetrode arrays implanted on moveable microdrives, wireless transmission of neural signals during flight, and synchronized behavioral tracking in three dimensions.16Frontiers in Cellular Neuroscience. 3D Hippocampal Place Field Dynamics in Free-Flying Echolocating Bats The technical difficulty of recording from a brain in freefall flight is part of why bat hippocampal data arrived decades after the first rat place cell studies in the 1970s.
The Curious Case of Dolphins and Whales
If a bigger hippocampus tracks with greater spatial memory demands, what happens in animals that have enormous brains but navigate a relatively featureless ocean? Cetaceans, the group that includes dolphins, whales, and porpoises, provide a striking answer: their hippocampus is disproportionately small.
Compared to what you would predict from overall brain size, the cetacean hippocampus is only about 8% to 20% of the expected volume for a mammal of its brain size.17PubMed Central. How the evolution of air breathing shaped hippocampal function The structure also lacks detectable signs of adult neurogenesis, the ongoing production of new neurons in the dentate gyrus that is a hallmark of the mammalian hippocampus in many other species.18PubMed Central. In contrast to many other mammals, cetaceans have relatively small hippocampi that appear to lack adult neurogenesis.
This raises awkward questions. Dolphins are widely regarded as among the most intelligent mammals, with complex social behavior, tool use, and what looks like cultural transmission. Yet their hippocampal anatomy, at least by the metrics we use in rodents and primates, would predict modest spatial and episodic memory. One possibility is that dolphins have offloaded some memory functions to other brain regions. Another is that the open ocean genuinely demands less hippocampal-style spatial mapping than a terrestrial environment rich in landmarks. Either way, the cetacean hippocampus is a reminder that brain size and behavioral complexity do not always align region by region.
Sex, Mating Systems, and Hippocampal Size in Voles
Ecology shapes the hippocampus not only across species but within them, and one of the clearest demonstrations comes from voles. In polygamous meadow voles, males roam widely to find mates and consistently outperform females on laboratory spatial tasks. In monogamous pine voles, neither sex ranges farther than the other, and the spatial performance gap disappears.
The brains reflect this behavioral difference precisely. Only in the polygamous species do males have a larger hippocampus relative to overall brain volume than females. In the monogamous species, male and female hippocampal proportions are equivalent.19PubMed. Evolution of spatial cognition: sex-specific patterns of spatial behavior predict hippocampal size The vole comparison is a clean natural experiment: two closely related species with different mating systems produce predictably different hippocampal anatomy, suggesting that selection for spatial ability during mate-searching is a direct driver of hippocampal expansion.
Hibernation and the Shrinking, Rebuilding Hippocampus
European hamsters demonstrate a different kind of hippocampal plasticity: the structure physically retracts during hibernation and then rebuilds itself with astonishing speed upon waking. During torpor, the dendritic trees of hippocampal neurons shrink, becoming shorter, less branched, and carrying fewer spines (the tiny protrusions where synaptic connections form). The synaptic vesicles inside major hippocampal terminals also thin out dramatically.
The remarkable part is what happens when the hamster wakes up. Within roughly two hours of arousing from torpor, dendritic length, branching patterns, and spine density return to levels seen in fully active animals. Synaptic structure also bounces back to normal.20PubMed Central. Rapid and reversible changes in intrahippocampal connectivity during the course of hibernation in European hamsters This rapid rewiring suggests that the hippocampus has built-in mechanisms for protecting itself during metabolic shutdown and snapping back into full connectivity almost instantly. Understanding how hamsters pull this off could eventually have implications for neurodegenerative disease research, where the loss of synaptic connections is a central problem.
How Neurogenesis Rates Differ Across Species
Adult hippocampal neurogenesis, the birth of new neurons in an already-mature brain, is often discussed as though it is a universal mammalian trait. It is not equally distributed. Across species and orders, the rate of new neuron production in the dentate gyrus declines with longer life expectancy. Long-lived primates and foxes have significantly fewer proliferating cells at equivalent life stages (weaning, first reproduction, average lifespan) than rodents do.21PubMed. Comparing adult hippocampal neurogenesis in mammalian species and orders: influence of chronological age and life history stage
When researchers normalized proliferating cell counts to total granule cell numbers, they found an overall exponential decline in neurogenesis that was chronologically similar between species, regardless of lifespan. In other words, the hippocampus appears to have a built-in schedule for winding down new neuron production, and long-lived animals simply reach that wind-down point while they still have decades of life ahead of them. Cetaceans sit at the extreme end of this spectrum, apparently lacking detectable neurogenesis altogether, as noted earlier.
For aging dogs, the pattern is familiar to anyone who has watched a pet slow down. Aged dogs show select neuron loss and reduced neurogenesis in the hippocampus, and these changes track with declines in learning and memory that resemble, in some ways, early cognitive decline in humans. Dogs have become a valuable natural model for studying hippocampal aging precisely because they share a domestic environment with humans and develop age-related cognitive impairment spontaneously rather than through laboratory manipulation.
Social Recognition and the Hippocampus
The hippocampus is not only about maps and landmarks. Social recognition memory, the ability to remember and identify familiar members of one’s own species, also depends on hippocampal circuits. This kind of memory is essential for daily social life: recognizing allies, avoiding rivals, maintaining pair bonds, and navigating dominance hierarchies all require remembering who is who.22PubMed Central. Regulation of Social Recognition Memory in the Hippocampal Circuits
Research into the hippocampal basis of social memory has explored how social isolation affects the strength and persistence of social recognition. Animals kept in isolation show altered social recognition, suggesting the hippocampal circuits underlying this ability are sensitive to social experience. The finding makes intuitive sense: a brain region that builds contextual representations of the environment would naturally incorporate information about who was present at a given time and place.
When Invertebrates Reinvent the Wheel
The hippocampus is a vertebrate structure, but evolution has a habit of arriving at similar solutions through unrelated paths. Octopuses offer a compelling example. The vertical lobe system of the octopus brain, particularly the connections between the median superior frontal lobe and the amygdala-like lobes, shows robust long-term potentiation of synaptic transmission that resembles what happens at hippocampal synapses in mammals.23PubMed Central. Neurotransmission and neuromodulation systems in the learning and memory network of Octopus vulgaris
Octopuses and vertebrates last shared a common ancestor roughly 600 million years ago, long before anything resembling a hippocampus existed. The fact that octopus learning circuits use a similar synaptic strengthening mechanism is a case of convergent evolution: two lineages, facing similar computational challenges around learning and memory, arrived at overlapping molecular solutions independently. It suggests that long-term potentiation is not just one possible mechanism for memory storage but something close to an inevitable one, given the biophysical toolkit available to neurons.
Place Cells in Rats and What They Revealed
Much of what we know about hippocampal function in any species traces back to work in laboratory rats. The rat hippocampus contains place cells, neurons whose firing is tightly coupled to the animal’s location in an environment.24PubMed. Evidence for a relationship between place-cell spatial firing and spatial memory performance When a rat moves through a maze, different place cells activate in sequence, creating an internal representation of the route. These cells were first described in the early 1970s and have since become one of the best-studied examples of how a neural population encodes an abstract variable.
The rat model established the link between place cell firing and actual spatial memory performance, making it the foundation on which comparative studies in bats, birds, and fish were built. Yet there is a persistent tension in the field about how far rat findings generalize. Rats are nocturnal, largely two-dimensional navigators with excellent whisker-based touch and a strong reliance on olfaction. Extending their hippocampal physiology to three-dimensional fliers, open-ocean swimmers, or scent-guided migrators has sometimes produced surprises, like the absence of theta in bat place cell firing. The rat hippocampus remains indispensable as a reference point, but comparative work has made it clear that it is one data point, not the template.