Bigger numbers, not better brains, smarten human cultures

Tool innovations take off as populations grow and share diverse ideas

CULTURE CLUB  While part of a team, computer-game players saw examples of how to draw an arrowhead, left, and a fishing net, right, before trying to copy or improve on the originals. Larger groups in this game developed more effective designs, suggesting that big populations drive cultural evolution.

ISEM/CNRS

Language, computers and other novelties of human cultures are primarily the products of living in large groups, a new study suggests.

Technological advances and the accumulation of other know-how get a jump start as populations expand, say evolutionary biologist Maxime Derex of the University of Montpellier 2 in France and his colleagues. Their laboratory experiments, reported November 13 in Nature, indicate that improvements in tool design occur more frequently as group size grows. Such advances spread rapidly as group members copy whatever works best.

The importance of brute numbers for social learning helps to explain why the human species, which originated around 200,000 years ago, displayed a rapid burst of cave painting and other complex cultural practices around 45,000 years ago, a time of population expansion, Derex proposes.

And because Neandertals generally lived in smaller groups than ancient people did, greater numbers rather than intellectual superiority gave Stone Age people the upper hand in tool-making, proposes anthropologist Joseph Henrich of the University of British Columbia in Vancouver, who was not involved in the new study.

“It’s better to be social than smart,” Henrich says.

Still, the transmission of knowledge in cultural groups is messy and difficult to capture in lab experiments, writes zoologist Peter Richerson of the University of California, Davis in an accompanying commentary in Nature. Long-term field studies represent one way to confirm and extend findings from computer games such as the one Derex employed, Richerson says.

In the new investigation, 366 male college students watched a video demonstrating how to draw an arrowhead or a fishing net. Then they spent 15 trials of a computer game drawing one of the tools. Volunteers could switch from one tool to another on any trial. The computer calculated how much food each tool design would yield in the game’s virtual world.

Participants were randomly assigned to groups of two, four, eight or 16. After each trial, players viewed the food scores of others in their group and spent 40 seconds studying the step-by-step procedure to build any one of their teammates’ designs.

The two largest groups did best at generating at least one accurately constructed arrowhead and fishing net in the experiment’s last three trials. Tool quality also increased with group size.

Players usually started out making arrowheads that were slightly worse and nets that were much inferior to the ones in the video. By the end of the game, only eight- and 16-member groups made arrowheads that were better than the video example and nets on a par with the demonstration net. Larger groups were more successful because participants in these groups had an increased probability of seeing and copying occasional changes in net construction that improved the fishing net’s ability to nab fish, the researchers say.

But the benefits of increasing group size may have a limit, at least in this lab game. Performance in the two largest groups differed little, perhaps because extra information in groups of 16 was as distracting as it was helpful, Richerson suggests. 

Bruce Bower has written about the behavioral sciences for Science News since 1984. He writes about psychology, anthropology, archaeology and mental health issues.