Study confirms social brain theory
The social brain theory - that animals in large social groups have bigger brains - has now been supported by a computer model.
For animals in smaller social groups, the cost of having a large brain outweighs the benefits.
Scientists used a simulation modelling technique to confirm that large social groups are only possible through sophisticated communication.
The study is published in Proceedings of the Royal Society B.
The human brain is a very costly organ which consumes a lot of energy. Animals that live in small social groups could therefore be at a disadvantage if they had large brains taking up processing power that could better be used elsewhere.
A team at Oxford University has now looked at the cognitive demands of making social decisions using a method called agent-based modelling, which models simplified representations of reality.
As expected, they found that more complex social decisions take up more 'brain' power.
The cognitive complexity of language evolved as social groups became larger and more complex, said lead author of the study Tamas David-Barrett from the University of Oxford.
He explained that a group of five is an ideal number to coordinate an event such as a hunt, but as the group size increases, the coordination involved would become increasingly complex.
"This paper shows what has been hypothesized since the late 1980s - that to live in a large group you need to have a large brain, and to live in a really large group like humans do, you need language.
"It was postulated this might be the case, but nobody has ever shown how it would happen. This paper provides a formal mathematical mechanism behind the theory."
The computational method used could also have other applications, like how inequality can emerge in a community, added Dr David-Barrett.
"We've introduced a new way of approaching how to model the way separate agents - ranging from dots on a screen to humans - get to some kind of communal or social action."
Ryota Kanai from the University of Sussex, who was not involved in the research, said it was an important study as it "refines existing theories by showing how the ability of information processing might interact with group size".
But he added that though the model supports the social brain hypothesis, it cannot alone provide an absolute proof for any theory on evolution.
"Unlike other types of science, it is difficult to repeat evolution in the laboratory. The model is based on abstraction of social interactions, which is great for the purpose of understanding the general relationships between group size and the ability to process social information.
"But it may not fully capture the complexity of real social interactions. For example, there might be more complex types of communication not considered in the model.
"Nevertheless, the study has made an important first step by offering a formal description of the social brain," Dr Kanai told BBC News.