These kinds of correlation have supported the social brain hypothesis, which says that large brains have evolved because intelligence is a vital component of social life. The hypothesis nicely explains how animals that live in groups can benefit from being clever by outwitting their rivals in competition over mates, food, allies, and status. It also explains why species with bigger brains tend to have more complex societies, and the hypothesis suggests that if a species has limited brainpower, its social options may be constrained as well: small-brained monkeys may be too dim to handle many social relationships.
The social brain hypothesis is very important in explaining a major benefit of being intelligent. Indeed, the advantages are so strong that we might expect all social primates to have developed big brains and high intellect. Yet there is wide variation. Lemurs are as small-brained as typical mammals. Apes have bigger brains than monkeys, and humans have the biggest brains of all. The social brain hypothesis does not explain these variations. It sets up this problem: if social intelligence is so important, why do some group-living species have smaller brains than others?
Diet provides a major part of the answer. In 1995 Leslie Aiello and Peter Wheeler proposed that the reason some animals have evolved big brains is that they have small guts, and small guts are made possible by a high-quality diet. Aiello and Wheeler’s head-spinning idea came from the realization that brains are exceptionally greedy for glucose—in other words, for energy. For an inactive person, every fifth meal is eaten solely to power the brain. Literally, our brains use around 20 percent of our basal metabolic rate—our energy budget when we are resting—even though they make up only about 2.5 percent of our body weight. Because human brains are so large, this proportion of energy expenditure is higher than it is in other animals: primates on average use about 13 percent of their basal metabolic rate on their brains, and most other mammals use less again, around 8 percent to 10 percent. As expected from the importance of maintaining energy flow to our many brain cells (neurons), genes that are responsible for energy metabolism show increased expression in the brains of humans compared to the brains of nonhuman primates. The high rate of energy flow is vital because our neurons need to keep firing whether we are awake or asleep. Even a brief interruption in the flow of oxygen or glucose causes neuron activity to stop, leading rapidly to death. The constant energy demand of brain cells continues even when times are tough, such as when food is scarce or an infection is raging. The first requirement for evolving a big brain is the ability to fuel it, and to do so reliably.
Given that large brains need large amounts of energy, Aiello and Wheeler asked themselves what special features of our species enable us to apportion more glucose to our brains than other animals do. One possibility is that humans might have a uniquely high rate of energy use. After all, human food is exceptionally calorie-dense and we routinely take in more energy per day than a typical primate of our body weight, so maybe extra energy running through our bodies gives us the calories we need to feed our hungry brains. But basal metabolic rates are well known in primates and other animals, and they are unremarkable in humans. A resting person supplies energy to their body at almost exactly the rate predicted for any primate of our body weight. Since nothing about basal metabolic rates is special to humans, Aiello and Wheeler were able to rule out the idea that our big brains are powered by inordinate amounts of energy passing through the body.
The elimination of the overall high-energy-use theory was a critical breakthrough because it left only one solution. Among species that have the same relative basal metabolic rate, such as humans and other primates, extra energy going to the brain must be offset by a reduced amount of energy going elsewhere. The question is what part of the body is shortchanged. Among primates, the size of most organs is closely predicted by body weight because of inescapable physiological rules. A species whose body weighs twice that of another needs a heart that weighs almost exactly twice as much. Hearts have to be a certain size to pump enough blood around a body of a certain size. No trade-off is possible there. Similar principles apply to kidneys, adrenals, and most other organs. But Aiello and Wheeler found a provocative exception to this tendency. They discovered that across the primates there is substantial variation in the relative weight of the intestinal system. Some species have big guts and some have small. The variation in gut size is linked to the quality of the diet.
Anyone who has handled tripe or cleaned a deer knows that mammals have a lot of gut tissue. Mammalian intestines have a high metabolic rate, and in large, mostly vegetarian species like great apes, intestines tend to be busy all day, starting with the postdawn meal and continuing ceaselessly until hours after the animal goes to sleep. All this time the guts are engaged in several energy-intensive functions, such as churning, making stomach acid, synthesizing digestive enzymes, or actively transporting digested molecules across the gut wall and into the blood. Active guts consume calories at a consistently high rate, so their total energy expenditure depends on their weight and on how much work they are doing. Carnivores, such as dogs and wolves, have smaller intestines than plant eaters, such as horses, cows, or antelope. In species that are adapted to eating more easily digested foods, such as sugar-rich fruits compared to fibrous leaves, guts are also relatively small: fruit-eating chimpanzees or spider monkeys have smaller guts than the leaf-eating gorillas or howler monkeys. Those reduced guts use less total energy than larger guts and therefore give a species with a high-quality diet some spare calories to allocate elsewhere in the body.
The discovery that gut size varies substantially gave Aiello and Wheeler the opening they were looking for. Relative to their body weight, primates with smaller guts proved to have larger brains—just the kind of trade-off that had been expected. Aiello and Wheeler estimated the number of calories a species is able to save by having a small gut, and showed that the number nicely matched the extra cost of the species’ larger brains. The anthropologists concluded that primates that spend less energy fueling their intestines can afford to power more brain tissue. Big brains are made possible by a reduction in expensive tissue. The idea became known as the expensive tissue hypothesis.
Some species other than primates show a similar pattern, capitalizing on small guts to evolve particularly large brains. An elephant-nosed mormyrid fish from South America has a relatively tiny gut and is able to use an astonishing 60 percent of its energy budget to power its exceptionally large brain. Other animals follow the principle of an energy trade-off but gain muscle instead of brains. Birds that have small amounts of intestinal tissue tend to use their spare energy to grow bigger wing muscles, presumably because for a bird, better flight can be even more important than a bigger brain. Different kinds of trade-offs have also been proposed. Species with relatively low muscle mass have been found to have relatively large brains. The general lesson is that bigger brains must be paid for somehow. How animals with small guts make use of their energy savings depends on what matters to them. In primates the tendency to use energy saved by smaller guts for added brain tissue is particularly strong, presumably because most primates live in groups, where extra social intelligence has big payoffs.
The expensive tissue hypothesis predicted that major rises in human brain size would be associated with increases in diet quality. Aiello and Wheeler identified two such rises. The first brain-size expansion was around two million years ago from australopithecines to
Homo erectus
. In line with the Man-the-Hunter scenario, the scientists credited this rise in brain size to the increased eating of meat. Second was a little more than half a million years ago, when
Homo erectus
became
Homo heidelbergensis
. They attributed this rise to the only other obvious candidate for an improvement in dietary quality: cooking.
I believe that Aiello and Wheeler were right in their principles. But they were wrong in their specifics because they assumed there was only a single increase in brain size from australopithecines to
Homo erectus.
In actuality, that phase of our evolution occurred in two steps: first, the appearance of the habilines, and second, the appearance of
Homo erectus
. Meat eating and cooking account respectively for these two transitions, and therefore for their accompanying increases in brain size.
The expensive tissue hypothesis provides an explanation not only for the substantial increases in brain size that occurred around the time of human origins, but also for the many other rises in brain size before and after two million years ago. Consider first our last common ancestor with chimpanzees, which lived around five million to seven million years ago. We can reconstruct this pre-australopithecine ape as living in rain forest and resembling a chimpanzee. Closely related to gorillas as well as chimpanzees, these ancestors likely had brains comparable in volume to those found in great apes living today, and therefore had larger brains than are found in living monkeys. The apes’ big brains compared to those of monkeys are nicely explained by the expensive tissue hypothesis, because great apes have high-quality diets for their body weights. They eat relatively less fiber and fewer toxins than monkeys.
Chimpanzees have a cranial capacity of around 350 to 400 cubic centimeters (21.6 to 24.4 cubic inches). Australopithecines, with the same body weight as chimpanzees or even slightly less, had substantially larger cranial capacities, about 450 cubic centimeters (27.5 cubic inches). Following Aiello and Wheeler’s hypothesis, australopithecine diets should therefore have been higher in quality than the diets of living chimpanzees. This seems likely. During seasons of plenty, australopithecines would have eaten much the same diet as chimpanzees or baboons do when living in the kinds of woodland that australopithecines occupied—fruits, occasional honey, soft seeds, and other choice plant items. It was when fruits were scarce that australopithecines must have eaten better than their chimpanzee-like ancestors. Present-day chimpanzees that are short of fruit turn to items specific to their rain-forest homes, eating foliage such as the stems of giant herbs and the soft young leaves of forest trees. In their drier woodlands australopithecines would have found few such items. The most likely alternatives were starch-filled roots and other underground or underwater storage tissues of herbaceous plants. These would have been ideal.
Carbohydrates are stored abundantly in corms, rhizomes, or tubers of many savanna plants and are highly concentrated sources of energy-rich starch in the dry season. These food reserves are so well hidden that few animals can find them, but chimpanzees do dig for tubers occasionally, sometimes with sticks, and australopithecines would have been at least as skillful and well-adapted: their chewing teeth are famously massive and somewhat piglike, suited to crushing roots and corms. An important location for australopithecine food sources likely would have been the edges of rivers and lakes, where sedges, water lilies, and cattails grow well and provide a natural supermarket of starchy foods for hunter-gatherers today.
The underground energy-storage organs of plants have a quality anticipated by the expensive tissue hypothesis: they have less indigestible fiber from plant cell walls than foliage, making them easier to digest and therefore a food of higher value. A dietary change from foliage to higher quality roots is thus a plausible explanation for the first increase in brain size, from forest apes to australopithecines five million to seven million years ago.
During the second sharp increase, brain volume rose by about one-third, from the roughly 450 cubic centimeters (27 cubic inches) of australopithecines to 612 cubic centimeters (37 cubic inches) in habilines (based on measurements of five skulls). The body weights of australopithecines and habilines were about the same, so this was a substantial gain in relative brain size. Given the archaeological evidence, the big dietary change at this time was more meat eating, so meat should have made this brain growth possible. To account for such a large increase in brain size, it seems likely that habilines processed their meat. Apes and humans are disadvantaged: their teeth cannot cut meat easily, their mouths are relatively small, and as William Beaumont noticed in the case of Alexis St. Martin, their stomachs do not process hunks of raw meat efficiently.
Chimpanzees also show that eating unprocessed meat is difficult with ape jaws. They chew their animal prey intensely, but small bits of undigested meat sometimes appear in their feces. Perhaps because of this hard work and inefficiency, chimpanzees sometimes decline the opportunity to eat meat despite their usual enormous enthusiasm for it. After chewing meat for an hour or two, a chimpanzee can abandon a carcass and relax or eat fruit instead. Chimpanzees of the Kanyawara community in Kibale National Park, Uganda, occasionally forgo meat-eating opportunities without chewing muscle at all. I once saw Johnny, an avid chimpanzee hunter of red colobus monkeys, do this even though he appeared hungry for animal protein. He first killed an infant red colobus monkey, brought it to the ground, ate its intestines, then left the carcass lying unseen by other chimpanzees. He immediately returned to the trees, rapidly killed another infant, and repeated his prior actions: he again brought his prey to the ground, ate the intestines, and left the rest to rot. His preference for the softer parts was typical. When chimpanzees kill a prey animal, they normally eat such parts as the guts, liver, or brain first. They can swallow those quickly. But when eating muscle, chimpanzees are forced to chew it slowly, taking as much as an hour to chew one-third of a kilogram (three-quarters of a pound). They can get as many calories per hour by chewing fruits as they can by chewing meat. The habilines would have faced the same challenge. If they had relied on unprocessed meat for as much as half their calories and had eaten their meat as slowly as chimpanzees, with certain cuts of meat they would have had to spend several hours a day chewing it. The digestive costs likewise would have been high, since the gut would have been busy digesting for many hours.