Slowly at first, then faster, and never in a neat straight line. More than 3.5 million years ago, our relatives were already walking on two legs, with brains about a third the size of yours (Smithsonian). Over the long run of human evolution, brain size roughly tripled (Smithsonian). About 300,000 years ago, early Homo sapiens in Africa already had brains inside today’s size range. The rounder, globe-shaped brain most people have now came later, between roughly 100,000 and 35,000 years ago (Neubauer et al., 2018). Why brains grew is still argued over. Cooking, energy, social life, long childhoods and climate swings are all serious suspects, and they may have worked together. And a skull can tell us a brain’s size and shape, but not what anyone thought or felt.
How do you measure a brain that rotted away millions of years ago?
For these ancient humans, researchers study the space inside the skull rather than a preserved brain. Sometimes sediment fills it and hardens into a natural cast. Scientists also make casts, called endocasts, or measure the space with CT scans (Smithsonian). Analogy: it’s like working out the size of a cake that’s long gone by measuring the pan it was baked in.
That gives us a ladder of confidence. The volume and shape of a measured skull are Documented. What that brain let someone do is an inference. What that person thought or felt is Unknown. The whole collection runs on this habit. And the samples are small. Some species are known from only a few skulls, or even a single one (Villmoare & Grabowski, 2022).
Who had which brain, and when?
| Who | Roughly when | Fossils found in | What the braincases show |
|---|---|---|---|
| Australopithecus afarensis (“Lucy’s species”) | 3.85–2.95 million years ago | Ethiopia, Kenya, Tanzania | Usually under 500 cc, about ⅓ of ours (Smithsonian) |
| Homo habilis | 2.4–1.4 million years ago | Eastern and southern Africa | A “slightly larger braincase” than earlier species (Smithsonian) |
| Homo erectus | about 1.89 million–110,000 years ago | Africa; Dmanisi, Georgia; China; Indonesia | Eight measured endocasts: about 790–1,130 ml (Neubauer et al., Table 1) |
| Homo heidelbergensis | about 700,000–200,000 years ago | Europe, eastern and southern Africa, possibly China | Two measured skulls: about 1,160 and 1,250 ml (Neubauer et al.). Whether these fossils form one species is Disputed (Smithsonian). |
| Neanderthals | about 400,000–40,000 years ago | Europe; southwestern to central Asia | Eight measured: about 1,210–1,750 ml (Neubauer et al.; dates: Smithsonian) |
| Homo sapiens | about 300,000 years ago to today | Evolved in Africa; now worldwide | Jebel Irhoud, Morocco (~315,000 years): about 1,375 and 1,470 ml. Present-day average about 1,300–1,330 ml (Smithsonian; Neubauer et al.). |
| Homo floresiensis | bones about 100,000–60,000 years ago | Flores, Indonesia | About ⅓ the size of ours (Smithsonian) |
(A milliliter and a cubic centimeter are the same volume.)
Walking came first. About 3.6 million years ago, three early humans walked across wet volcanic ash at Laetoli, in Tanzania. Mary Leakey’s team found animal tracks there in 1976. In 1978, Paul Abell, newly on the team, found the human trail, about 70 footprints (Smithsonian). Donald Johanson and Tom Gray found “Lucy,” a 3.2-million-year-old skeleton, at Hadar, Ethiopia, on November 24, 1974. The original is kept at the National Museum of Ethiopia in Addis Ababa (Institute of Human Origins; Smithsonian).
Tools came before big brains. The oldest stone tools are a bit older than the oldest fossils of our genus, Homo (Smithsonian). At Nyayanga, Oldowan tools occur with cut-marked hippo bones at about 3 million years ago (Smithsonian research report). Here’s a fun wrinkle. When Leakey, Tobias and Napier named Homo habilis in 1964, the definition of Homo had to be rewritten, including lowering its minimum brain size (Smithsonian). Brain size was literally part of the rulebook.
Out of Africa, early. Homo erectus is generally considered the first species to spread beyond Africa, and it’s linked to the earliest Acheulean handaxes, by about 1.76 million years ago (Smithsonian). Hominins (members of the human family tree) reached Georgia by about 1.8 million years ago, and the spread may have begun even earlier (Plausible; Gowlett, 2016).
A small brain on an island. On Flores, Homo floresiensis made stone tools with a brain about a third the size of ours. Its small body and brain “may have resulted from island dwarfism.” Some scientists are now considering whether its ancestors were already small when they arrived (Smithsonian). Which story is right is Disputed.
Did brains grow smoothly, and are they shrinking now?
The pace is argued over. The Smithsonian describes slow growth until 2 million years ago and the fastest growth between 800,000 and 200,000 years ago. A 2021 analysis instead found that the growth rate jumped about 2.1 million years ago, then slowed about 1.5 million years ago to a steady rise (DeSilva et al., 2021). The tempo is Disputed.
The same 2021 team reported that brain size dropped within the last 3,000 years, roughly since 1000 BCE (DeSilva et al., 2021). A 2022 reanalysis found the pattern disappeared once the very large recent samples were adjusted for, and called brain size “remarkably stable” over the last 300,000 years (Villmoare & Grabowski, 2022). Recent shrinking is Disputed.
Why did brains get bigger?
Start with the energy bill. In an average adult, the brain accounts for about 2% of body weight but about 20% of oxygen and calorie use (Raichle & Gusnard, 2002). Analogy: it’s like a phone app that drains a fifth of your battery all day. In a 2016 comparison, humans used about 400 more calories a day than chimpanzees and bonobos after adjustment for body size and activity (Pontzer et al., 2016). That suggests our ancestors raised their total energy budget, not just cut costs elsewhere (Plausible).
The main explanations:
- Cooking and fire. Wrangham and colleagues argue that cooking, which makes food easier to digest, began with Homo erectus about 1.7 million years ago. A 2012 model suggests that a raw diet caps how many neurons a primate can afford (Fonseca-Azevedo & Herculano-Houzel, 2012). But direct evidence of fire is rare before about 1.5 million years ago, and recognizable hearths come later. Pinning cooking to one early moment is “far harder to demonstrate” (Gowlett, 2016). The timing is Disputed.
- Social life. Humphrey proposed that competitive social lives demand brainpower. Byrne and Whiten developed this as “Machiavellian intelligence,” and Robin Dunbar built the “social brain hypothesis” (Dunbar & Shultz, 2017). A 2022 meta-analysis of 44 published analyses found that social behavior tracks absolute brain and neocortex size, while diet and life history (how fast a species grows up and reproduces) track brain size relative to body size (Shultz & Dunbar, 2022). Plausible.
- Long childhoods. Human newborns have brains less than 30% of adult size. Chimpanzee newborns start at about 40%. The old “obstetrical dilemma” says hips narrowed for walking forced early birth. But human pregnancy is about 37 days longer than expected for a primate our size, and one 2012 study argues the mother’s energy limits set the timing (Dunsworth et al., 2012). Disputed.
- Climate swings. The Smithsonian links the fastest growth, 800,000 to 200,000 years ago, to a time of dramatic climate change (Smithsonian). Plausible.
These don’t have to compete; one review treats them as causes, results and constraints of one process (Dunbar & Shultz, 2017).
Is a bigger brain a smarter brain?
Not simply. For its body size, the human brain is roughly seven times bigger than expected for a mammal in one comparison and about three times bigger than expected for a primate or ape (Herculano-Houzel, 2012; Woods et al., 2005). Yet elephant brains are two to three times heavier than ours (Herculano-Houzel, 2012). Suzana Herculano-Houzel’s lab, then in Rio de Janeiro, counted brain cells directly by dissolving tissue into a soup of cell nuclei. They found about 86 billion neurons in the human brain. She argues that ours is a scaled-up primate brain that stands out mainly because it has so many neurons. Whether neuron count best predicts thinking ability is Plausible, still under study.
What do genes add?
- MCPH1 and ASPM. Some studies found signs of selection on them in apes and humans. A 2005 review urged caution. For MCPH1, the strongest signals weren’t on the lineages where brains grew most. For ASPM, the signal was highest on the human line but showed up in other apes too. Neither signal was strong enough to clearly tie these genes to human brain growth (Woods et al., 2005). A role in human brain expansion is Plausible, not established.
- FOXP2, a gene linked to speech. A 2018 reanalysis of hundreds of genomes found “no evidence” for a recent burst of selection reported in 2002. Neanderthals and Denisovans carry the same two protein changes once thought to mark our species alone (Atkinson et al., 2018).
- SRGAP2C, a partial gene copy found in humans and Neanderthals but not in chimpanzees, gorillas or orangutans. In mice it slowed the maturation of dendritic spines, the tiny bumps where synapses form (Charrier et al., 2012). A role in human evolution is Plausible, not settled.
How we know
What evidence exists: fossil braincases and endocasts, dated rock layers, stone tools, burnt material, footprints and ancient DNA. DNA shows that Neanderthals and Homo sapiens interbred. A 2014 genome study estimated Neanderthal ancestry at 1.5–2.1% in the non-African populations it examined (Prüfer et al. 2014, Gene flow).
What it can’t show: thoughts, feelings, beliefs, or when language began. All of those are Unknown from skulls alone.
Whose voices are missing: none of the prehistoric individuals discussed here left a written account, and the people who did the finding haven’t always been credited. At Zhoukoudian, near Beijing, local villagers showed J. G. Andersson’s team “Dragon Bone Hill” soon after he began work in 1918. Chinese paleontologists made nearly all the major finds there. Three of them, Pei Wenzhong, Yang Zhongjian and Jia Lanpo, became founders of Chinese paleontology and paleoanthropology. Yet foreign accounts often minimized the Chinese researchers’ central role (Roberts et al., 2021, pp. 98–99).
How power shaped the record: war did. The original Peking Man fossils vanished in December 1941, and what happened to them is still disputed (Roberts et al., 2021). So did the question of who measured skulls and why (see Tiedemann and Morton below).
Then vs. now
What people believed then:
- 4th century BCE. In Parts of Animals, Aristotle called the brain cold. He said its job was to balance the heat of the heart (Parts of Animals II.7).
- 1836–1839. Friedrich Tiedemann measured 248 skulls, found wide, overlapping ranges between groups, and argued for equality and the abolition of slavery. Samuel George Morton, with similar numbers, argued for a racial hierarchy (Mitchell, 2018).
- 1891–1940s. Eugène Dubois, a Dutch surgeon, found the first Homo erectus fossil on Java in 1891, before any early human fossils had turned up in Africa (Smithsonian). Raymond Dart described the Taung child from South Africa in 1925, but it took more than 20 years for scientists to widely accept Australopithecus (Smithsonian). Through the 1940s, most anthropologists worldwide thought humans had originated in Asia (Roberts et al., 2021, p. 99).
- 1949–1990. Paul MacLean, who began studying what he called the limbic system in the 1940s, later proposed a “triune brain”: three layers evolving in sequence, a “reptilian” core for basic functions such as movement and breathing, a limbic layer for emotion, and a cortex for language and reasoning. His book on it appeared in 1990 (Cesario et al., 2020, p. 258).
How that compares with current evidence:
- The brain is one of the body’s biggest energy users, not a cooling unit (Raichle & Gusnard, 2002).
- A 2018 study argued that Morton’s measurements may have been unbiased, but his race science was not. The same kind of data led two men to opposite conclusions (Mitchell, 2018).
- Ancient genomes support interbreeding between modern humans and archaic groups, complicating a simple replacement story (Prüfer et al. 2014, Gene flow).
- Your brain is not an onion with a lizard inside. All vertebrates share the same basic brain regions, and all mammals have a prefrontal cortex. Neurobiologists rejected the triune model long ago, but it lives on in textbooks (Cesario et al., 2020).
One more boundary. Evolution explains how this organ came to be. It doesn’t diagnose anyone, and it doesn’t explain any one person’s mental health.