Why do We Make Gestures (Even when No One Can See Them)?

The gesture doesn't work all the time

Why do we gesture? An obvious answer is that we gesture to communicate. After having just taken down his opponent in a two-legged flying tackle, the soccer player puts on his most innocent face while making a perfect sphere with his hands. This gesture conveys the following thought: “Ref, I was playing the ball! Sure, my opponent may be lying there writhing in pain and will soon be carried off on a stretcher but that’s beside the point. I do not deserve a red card.”

But we also gesture when our conversation partner cannot see us. Years ago I saw a madwoman walking in the Atlanta airport. See seemed to be talking to no one in particular while gesticulating vehemently. For a moment I was worried she might pull out a machine gun and mow us all down. But when she got closer I noticed she was speaking into a little microphone that was connected to her mobile phone (a novelty at the time). Evidently, the person that was on the receiving end of her tirade could not see her maniacal gestures.

So why do we gesture when no one can see our hands? According to one explanation, such gestures are merely habitual. We’re used to making gestures, so we keep on making them anyway even when nobody can see them. It’s a bit like a child on a tricycle. He lifts his legs but the pedals keep rotating. There is motion but it is not functional. The problem with this explanation is that it implies that we expend a lot of energy on a useless activity.

An alternative explanation proposes that gesturing is functional. It helps us retrieve words from our mental lexicon. The speaker says “At the fair we went into a uh…” He falls silent and makes a circular motion with his hand. He then looks relieved and finishes the sentence with “Ferris wheel.” The idea here is that the motoric information that drives our gestures is somehow connected with our word representations in our mental lexicon. The latter get activated when the gesture is made. Though plausible, a problem with this explanation is that is does not specify why the gesture is needed in the first place. If the motor program that drives the gesture is already present in the brain, then why loop out of the brain to make the gesture?

In a paper coming out today in Frontiers in Cognitive Science, a group of us—spearheaded by graduate student Wim Pouw—ventures an answer to this question.* People make noncommunicative gestures to reduce memory load and to externally manipulate information. We need to keep concepts active across stretches of time while performing a task, for instance solving a problem or conversing over the telephone. Rather than relying on memory to keep this information active, we outsource this task to the hands. They provide proprioceptive and visuospatial information that sustains access to a concept over time and allow us to perform operations on it (for instance manual rotation).

Support for this proposal comes from several sources. One is a classic paper by David Kirsh and Paul Maglio. Kirsh and Maglio observed that expert Tetris players often rotate objects on the screen before inserting them into their slots. They could have used mental rotation but instead prefer to rely on what Kirsh and Maglio call epistemic actions, operations on the environment to support thought.

Another line of support for our proposal comes from research on abacus users. People who learned arithmetic on an abacus make bead-moving finger gestures during mental arithmetic, when there is no abacus available. The better they are at mental arithmetic, the fewer gestures they make. This is consistent with the notion that noncommunicative gestures are epistemic actions that serve to reduce memory load. When you’re better at a task, performing it requires fewer memory resources, so you need to rely less on gestures.

So the next time you see people make gestures to no one in particular, you know that they’re just giving their memory a little hand. And if you want to know more about this topic, just read our paper.

* Our proposal was inspired by work by Richard Wesp and colleagues and by Andy Clark (see paper for references).