What are we doing when we look away during a conversation?

In face to face conversation, we often look away from the person we're speaking with. Somewhat paradoxically, the closer people sit to their conversation companions, the less often they look at them.

But other factors influence how often we avert our gaze, too. When we are asked personal questions, or difficult questions, or possibly when we are trying to deceive, we look away more often. When we talk with someone via a remote video monitor, we look at them more often than when we engage in the same type of conversation face to face.

So what's the cause of this behavior? Do several different causes lead to looking away, or is the root cause the same for all of them? Perhaps we look away when we are feeling socially challenged. After all, difficult questions, or social intimacy, or the heightened social awareness involved in deceiving others could all lead to the same feeling of being put on the spot.

But another explanation is possible at least some of the time. We get a great deal of information by looking at faces, and this information places a significant load on our cognitive systems. Perhaps, when we're asked a difficult question and need to concentrate, looking away from a face helps us focus on the cognitive demands of the question.

So do we look away because we're self-conscious, or because it helps us concentrate? Gwyneth Doherty-Sneddon and Fiona Phelps devised an experiment to distinguish between these two possibilities. They asked 8-year olds four different types of questions: verbal, arithmetic, episodic memory, and autobiographical memory. Each question was rated by the childrens' teachers as easy, medium, or hard. Half the kids were tested face-to-face, by a questioner sitting across a table from them. The other half were tested via a real-time video link. Their questioner appeared on a video monitor. The children were videotaped as they gave oral responses to the questions. Later the tape was analyzed to determine how often they averted their gaze from the questioner.

Doherty-Sneddon and Phelps expected that the kids using the video monitor would avert their gaze less often. They were interested in a different aspect of the results: whether the kids in each condition would respond differently when more difficult questions were asked. Here are the results (click on the picture for a larger version):

The graphs show the proportion of the time the kids looked away as they responded to the questions. As expected, the kids in the face-to-face condition looked away more of the time than those in the video monitor condition. Otherwise, however, the results were essentially the same—as questions became harder, the children looked away more. Doherty-Sneddon and Phelps argue that if gaze aversion were solely due to self-consciousness, the kids in the face-to-face condition would avert their gaze proportionately longer as the questions became more difficult. Since this was not the case, the reason for looking away is probably simply to reduce the overall cognitive demand and focus on the question.

So it appears that there are at least two reasons we look away from others while we talk to them: because of our self-consciousness or embarrassment at the intimacy of the situation, and because averting our gaze enables us to focus on the ideas behind what we're saying. This is not to say there aren't additional reasons. As Doherty-Sneddon and Phelps point out, there are different expectations in different cultures for how much we should look at each other. However, their work does appear to demonstrate that there is more to gaze aversion than just social nicety.

Doherty-Sneddon, G., & Phelps, F.G. (2005). Gaze aversion: A response to cognitive or social difficulty? Memory & Cognition, 33(4), 727-733.