Dinosaur brains: what fossils can and cannot tell us

A dinosaur brain was rarely preserved as soft tissue. What researchers can usually study is the space inside a skull, reproduced as a natural cast or reconstructed with CT scans. These endocasts reveal the brain's approximate shape and some sensory structures, but they are not brains, and they do not provide a simple score for intelligence.

The contrast between a small braincase in a large animal and the more compact skull of a bird has encouraged dramatic claims about dinosaur intelligence. The useful evidence is more specific: skull anatomy, inner-ear form, sensory pathways, growth and behaviour preserved indirectly in trackways, nests and other fossils.

Endocasts preserve the shape of the cavity that housed the brain, not the brain's full soft-tissue anatomy. Brain size can be estimated, but intelligence, personality and exact behaviour cannot be read directly from a skull.

How a brain becomes visible in a fossil

The brain sat inside the braincase, protected by the skull. In some fossils, sediment or mineral deposits filled the space after decay and formed a natural endocast. More often, palaeontologists use computed tomography to scan the surrounding bone, then digitally separate the brain cavity from the rock. The result is a three-dimensional model of the space and the structures that left impressions on it.

An endocast can preserve the broad outline of the cerebral hemispheres, the olfactory region, the cerebellum and the hindbrain. It may also show the positions of nerves and blood vessels where they passed through the skull. The inner ear, especially the semicircular canals, can be reconstructed from its bony housing and offers clues about balance and head movement.

These structures vary in how closely they follow the soft tissue. The brain did not fill every part of the cavity in the same way across all species, and membranes, fluid and blood vessels occupied space too. A digital endocast is therefore an anatomical proxy. The method provides valuable comparisons, but it does not expose every fold, connection or function of the original nervous system.

Brain size is not an intelligence meter

Absolute brain volume mostly reflects the size of the animal. A useful comparison must account for body mass, evolutionary relationships and the way the brain changes during growth. Researchers have used measures such as relative brain size and encephalisation quotients, but these depend on the comparison group and statistical model. Different methods can produce different rankings.

A small value relative to body mass does not mean an animal was incapable or “stupid”. A large body requires extensive sensory and motor control, while brain tissue is energetically costly. Animals can also specialise in different senses and behaviours. For that reason, a single ratio cannot fairly compare a large herbivore, a small active theropod and a modern bird.

The bony skull can mislead in another way. A thick roof, an enlarged sinus or a large crest may make a braincase look spacious or compact without corresponding directly to neural tissue. Comparisons must distinguish the brain cavity from adjacent structures and use well-preserved specimens. Juvenile and adult skulls may also differ as the brain and surrounding bones grow at different rates.

The “second brain” myth

The best-known story concerns Stegosaurus. An enlarged space near the hips was once described in popular accounts as a second brain that controlled the hindquarters. The structure is part of the spinal canal and associated sacral anatomy, not evidence for a second brain. Similar enlargements occur in other dinosaurs and can relate to the size of nerves, blood supply or a glycogen body, though the exact interpretation of the soft-tissue space is not settled.

The dinosaur's movements were coordinated by its brain and spinal cord, as in other vertebrates. Spinal circuits can organise rapid reflexes without a separate brain. Because a very large animal has long nerves, signal travel times and local processing are useful questions, but they do not turn a sacral cavity into an independent control centre.

Senses, behaviour and what fossils can show

Endocasts and inner-ear anatomy can help compare smell, balance and hearing across dinosaurs. These are clues, not complete sensory biographies. The size of an olfactory region may suggest that smell mattered, but it does not tell us exactly what an animal could detect in its environment. Trackways can show movement and spacing; nests and juveniles can inform reproduction and development. None records a dinosaur's private thoughts or provides a direct IQ test.

Birds are living theropod dinosaurs, and crocodilians are their closest living archosaur relatives. Their anatomy and behaviour help researchers frame hypotheses about extinct nervous systems. The comparisons must still respect millions of years of evolution: a modern bird is not a stand-in for every non-avian dinosaur, and a crocodile is not an unchanged ancestor.

The fossil record supports a varied picture. Some small theropods had relatively large brains and well-developed sensory regions, while giant herbivores had different proportions and demands. That diversity is more informative than a single ladder ranking all dinosaurs from clever to dull.

What we can conclude

Skull cavities, CT scans and comparisons with living archosaurs let scientists reconstruct important parts of dinosaur neuroanatomy. They can estimate the size and shape of the braincase, examine nerve openings and compare the inner ear. They cannot recover a complete brain or translate a volume into a precise level of intelligence.

Claims about a “second brain”, exact mental ability or a universal ranking go beyond the fossils. The strongest account keeps anatomy, functional inference and speculation separate. For related evidence, see how researchers study other internal structures and compare teeth and jaw mechanics. The dinosaur catalogue links these anatomical questions to individual taxa.