Fruit, honey and other carbohydrate-rich foods may have helped supply the energy needed as human brains grew larger over millions of years, according to a new evolutionary model.
The study challenges versions of human evolution that focus mainly on meat by arguing that glucose-rich plant foods also may have played an important role. But the research does not prove that sugar caused the human brain to expand or establish exactly what early human ancestors ate.
Published in Science, the study modeled the glucose needs of humans and their ancestors across roughly 4 million years. Researchers combined evidence from metabolism, physiology, fossils, archaeology, genetics, primate diets and food composition to estimate how growing brains and reproductive demands could have been fueled at different stages of evolution.
The model suggests that early human ancestors may have relied heavily on fruit and other naturally sweet foods before cooking made starch-rich roots and tubers easier to digest. Honey also may have provided a concentrated source of energy when it was available.
“People often think of carbohydrates as breads and cereals, but for most of human evolution carbohydrates came from fruits, honey and underground plant foods,” said Jennie Brand-Miller, a professor of human nutrition at the University of Sydney and the study’s lead author.
The human brain is unusually large relative to body size and requires a substantial amount of energy. Glucose is its primary fuel under ordinary conditions, although the body can also make glucose from other compounds and use more ketones when carbohydrate intake is low.
That means early humans did not necessarily have to obtain every gram of glucose directly from carbohydrate. The researchers’ argument is instead that carbohydrate-rich foods may have offered an efficient way to help meet rising energy and glucose demands.
The model examined what the researchers called obligatory glucose demand, including the glucose used by the brain, red blood cells, kidneys and reproductive tissues. It also considered the additional demands of pregnancy, fetal development and milk production.
“When we think about the foods that supported human evolution, we need to consider not only brain growth but also the nutritional requirements of pregnancy and raising children,” said David Raubenheimer, a professor of nutritional ecology at the University of Sydney and a study co-author.
Those reproductive demands may have made readily available carbohydrate especially valuable. Still, the study did not examine food cravings or show that modern pregnancy cravings for fruit reflect an ancient need for glucose.
The researchers divided human evolution into six stages and estimated how different proportions of plant and animal foods might have met the body’s needs. They concluded that fruit was likely especially important for earlier hominins whose diets predated widespread cooking.
As fire and food processing became more common, underground plant foods such as wild yams and rhizomes may have supplied more digestible starch. Cooking changes the structure of starch and generally makes it easier for digestive enzymes to break down.
The release accompanying the study states that raw starch does not release glucose, but that is too absolute. Uncooked starches vary in digestibility, and some can be broken down. Cooking generally makes much more of their energy available.
The research also does not diminish the possible role of animal foods. Meat and other animal foods could have supplied concentrated energy, protein, iron, vitamin B12 and other nutrients that supported growth and development.
The more complete picture may be that human evolution depended on dietary flexibility rather than one decisive food. Fruit, honey, cooked starches and animal foods could all have contributed at different times and in different environments.
The role of honey is particularly difficult to determine. Honey provides concentrated carbohydrate, and many modern foraging communities value it highly. But it leaves little direct archaeological evidence, making it difficult to know how often early humans found or ate it.
The model therefore shows that honey and fruit could plausibly have helped meet energy needs. It cannot determine how much either food contributed to brain growth.
“Our desire for sweetness is instinctive,” Brand-Miller said.
A preference for sweet foods may have helped human ancestors identify ripe fruit and other valuable energy sources when food was less abundant and less predictable. That does not mean modern diets high in added sugar are beneficial.
Sugars in whole fruit come packaged with water, fiber, vitamins, minerals and other plant compounds. Modern sweets and sugary drinks can deliver much larger amounts of sugar with fewer accompanying nutrients and are available throughout the year.
The study also cannot determine how much carbohydrate people should eat today or whether a particular modern diet is best. It does not compare low-carbohydrate and high-carbohydrate diets, measure disease risk or test the effects of fruit, honey or grains in living participants.
Its contribution is historical and theoretical: carbohydrate-rich foods may deserve a larger place in explanations of how humans met the energy demands of larger brains, pregnancy and childhood.
“Understanding the role of carbohydrates alongside meat, fat and protein provides a more complete picture of the human evolutionary journey,” Brand-Miller said. “Our study suggests sugars and starches may have helped shape the development of the human brain and many of the traits that make us human.”
The findings offer a useful counterpoint to the idea that one food group alone drove human evolution. They do not turn evolutionary history into a modern prescription or show that more sugar is better.
The study was funded by the Powerful Plants UKRI Horizon Europe Guarantee Fund. Brand-Miller is a co-author of books about carbohydrates and the glycemic index and oversees a glycemic index testing service at the University of Sydney. She consults for the China National Research Institute of Food and Fermentation Industries and the Novo Foundation, serves on the scientific advisory board of ZOE Global and was president of the Glycemic Index Foundation from 2002 to 2024.