I was a member of my high school debate team, and I did fairly well, but my partner, Glenn, always got better marks from the judges. Most often, they praised his hand gestures, which were proclaimed to be “expressive” and “informative.” One year our topic was arms control, and our opponents were arguing that “NATO standardization” could help reduce U.S. arms sales. Glenn didn’t understand their argument, so during our precious three minutes of preparation time, I explained it to him. Then he stood up and delivered his rebuttal, using the most graceful hand gestures imaginable. The judges unanimously said on their ballots that they thought Glenn understood the arguments better than any of the debaters on either side.

Could those hand gestures really be the key to Glenn’s high marks from the judges? There’s plenty of research suggesting that informative hand gestures make speech easier to understand. But additional research also suggests that the person doing the gesturing can learn more effectively than someone who doesn’t gesture while learning. Maybe during the course of his speech, Glenn actually did come to understand the argument better than anyone in the room.

A team led by Susan Wagner Cook took 84 third- and fourth-graders and tested them to see if they could do a simple type of math problem, like this:

4 + 3 + 6 = ___ + 6

Selecting only the children who didn’t yet understand how to solve this sort of problem, they worked one-on-one with the kids, modeling one of three methods to solve it: either using gestures, using speech and gestures, or using speech alone. Then everyone was taught the same method of solving using both speech and gestures:

For example, after putting the correct answer in the blank for the problem 4 + 9 + 3 = 4 + ___, the instructor said, “I want to make one side (while sweeping the left hand under the left side of the equation) equal to the other side (while sweeping the right hand under the right side); so four plus nine plus three equals sixteen, and four plus twelve equals sixteen; one side (gestures under the left side) is equal to the other side (gestures under the right side).

This was explained for six different problems, after each of which the child was asked to solve a problem of his or her own, mimicking the behavior they had seen at the outset (gestures alone, speech and gestures, or speech alone). The children were then given a post-test on a new set of problems. Finally, everyone was given a follow-up four weeks later. Here are the results:

These graphs can be a little confusing at first, but they are very revealing. On the post-test given immediately after they learned to solve the problem, everyone’s scores were roughly the same: Some students solved all the problems, and some solved none, with only a few in the middle of the scale. But on the follow-up four weeks later, big differences emerged. Kids who had solved their practice problems using either gesture alone or gesture and speech solved about the same number of problems in the follow-up as they had in the post-test. But many of the kids who had solved problems correctly in the speech-only group didn’t fare as well on the follow-up; only a few scored well on both the post-test and the follow-up.

Cook’s team says this difference must be due to the students’ own gesturing. Since all the students had been taught using both speech and gestures, their own gestures must have been responsible for the learning.

How does gesture do this? At this point, we can only speculate. Perhaps gestures are relatively easy, freeing cognitive resources for other tasks. Maybe gestures help us encode thoughts into memory. Or maybe they help build links to the external environment, which in turn makes things easier to synthesize and recall.

COOK, S., MITCHELL, Z., & GOLDINMEADOW, S. (2008). Gesturing makes learning last Cognition, 106 (2), 1047-1058 DOI: 10.1016/j.cognition.2007.04.010