Hack97.Look Where I m Looking | Mind Hacks. Tips and Tools for Using Your Brain

Hack 97. Look Where I'm Looking

We are innately programmed to follow other people's eye gaze to see what they are looking at. It's so deeply ingrained that even cartoon eyes can interfere with our mental processing of direction.

Eyes are special. They're part of a two-way sense. Wherever I look, you can tell what I'm looking at. You can tell if I'm paying attention to you or not, as well as hazarding a good guess as to what I'm really thinking about. Following gaze isn't a learned behavior. As far as the brain's concerned, gaze direction is a first-class citizen of the real world, as important as location. In the case of location, the Simon Effect [Hack #56] demonstrates that we have a tendency to react to a prompt in the same direction as that stimulus. This hack shows that we interpret gaze direction in much the same way as location: a cartoon pair of eyes looking in one direction has the same effect.

10.6.1. In Action

A team at the University of Padua in Italy constructed an experiment to see the effect of gaze.1 They drew a pair of cartoon eyesjust two ovals with a colored oval (the iris) within each, as shown in Figure 10-5. The irises were colored either blue or green, and the cartoon could be looking either straight ahead or to one of the sides.

Figure 10-5. Cartoon eyes similar to the ones used in the experiment: show this page to someone and watch what her eyes dosee if you can catch her just flicking off to the right as the cartoon eyes trigger her automatic gaze-following routine

People taking part in the experiment had to report the color of the irises, hitting a button on the left for blue and on the right for green. The apparent gaze direction wasn't important at all. Despite that, it was faster to hit the button for green on the right when the eyes were looking the same way (to the right) and slower when they were looking the other way. The same held true for blue and the eyes looking left.

Thinking this might be nothing to do with the ovals looking like eyes, to investigate further, the team put together another task. Instead of ovals, the cartoon "eyes" were squares, with square "irises" in each, and looked much less like eyes (as shown in Figure 10-6). And sure enough, the significant reaction time difference (between gaze pointing in the same direction as the response key and in the opposite direction) went away.

Figure 10-6. Square "eyes" similar to the ones used in the experiment: we don't follow the gaze of robot eyes

It's possible this is why the X Windows toy "xeyes," which puts a pair of eyes on your computer desktop that follow your mouse cursor around, is so uncannily handy for avoiding losing your pointer.²

R.D.

10.6.2. How It Works

Essentially, this experiment shares a mechanism with the Simon Effect [Hack #56] . Given that the brain translates gaze direction into location, the same effect gets triggered: if attention has already been directed to the left because of the stimulus, it takes a very short time more to make a response on the right.

It makes sense that we treat gaze with such respect. If someone's looking at us, it usually means that some kind of interaction, for good or ill, is in motion. And if there are a few people looking at the same place, they've probably spotted something you should know about. Gaze is physically so tiny, but figures large in our social world. Just think how it feels to make eye contact with a stranger (sometimes good, sometimes embarrassing) or how difficult it is to have a conversation with someone looking elsewhere. If you doubt this, try repeatedly glancing over somebody's shoulder while you're talking with him, and see how long it takes him to crossly ask you what you're up to or glance over his shoulder to see what you're looking at.

I've a suspicion that this is why arrows work so well as symbols. Given we can look at cartoons of eyes and still follow gaze, how reduced can that cartoon get and still work? OK, it doesn't work with square eyes, but the salient features of eyes are that they're long and pupils go into the corner and pick up an arrowhead shape when they're looking off to the side. Just like an arrow, in factwhich leaves me wondering whether arrows (in print, in signage, and everywhere) are very good learned symbols or whether they tap into something deeper.

M.W.

This deep gaze perception is also sometimes referred to as shared, or joint, attention. The use of two people paying attention to the same object is most obvious for infants. When infants are learning, they need to be able to make associations between objects, words, actions, and so on. If there were no way to point or direct an infant's attention, it'd be next to impossible to teach her anything. At about 12-18 months, children follow eye gaze: they can observe your eyes, tell what you're looking at, and look over there instead. It's this automatic mechanism of joint attention that is used in making a shared understanding of the world.3

Pointing performs much the same function as gaze, indicating where to pay attention. Of course, this doesn't work for dogs, as is well-known by anyone who's tried pointing as a way to get a dog to look somewhereonly to have the animal stare at the end of the pointing finger. The only way to teach association for dogs, who don't understand such abstract symbols as pointing, is to wait until they're paying attention to whatever you want to teach them about anyway and then do whatever you were going to do. It could be argued that our real defining characteristic, as humans, is the ability to understand the symbol "that." Without it, we wouldn't be able to talk about objects that aren't present, and wouldn't be able to learn from other people's mistakes.

(Some dog lovers may disagree, but then this is what makes them dog lovers, isn't it?)

M.W.

Shared attention is one thing that infants use as a springboard to develop an understanding of other minds, not just the world of objects. Children with autismindividuals with an impaired understanding of other people's intentionsdon't follow gaze automatically as other children do.

10.6.3. In Real Life

The effortless, rapid, almost unconscious encoding of gaze direction makes it an ideal social signal, which, of course, leaves it open for abuse. Spot how many advertisements have a large face with eyes directed exactly at what they want you to read.

10.6.4. End Notes

Zorzi, M., Mapelli, D., Rusconi, E., & Umlità, C. (2003). Automatic spatial coding of perceived gaze direction is revealed by the Simon Effect. Psychonomic Bulletin & Review, 10(2), 423-429.
Lowcoders have written a Mac OS X clone of xeyes, called iEyes (http://www.lowcoders.net).
Driver, J., Davis, G., Ricciardelli, P., Kidd, P., Maxwell, E., & Baron-Cohen, S. (1999). Gaze perception triggers reflexive visuospatial orienting. Visual Cognition, 6(5), 509-540.