It's a Jungle in There: How Competition and Cooperation in the Brain Shape the Mind (9 page)

                                
Pay Attention!

To read these words in a meaningful way, you can shunt your attention back and forth while going from top to bottom. The students who conducted this study reasoned that if their listening subjects did this, they would not behave as Broadbent predicted, which would be to say “Dear 2 Sally 7 are 3” or “8 Aunt 5 how 9 you.” Instead, the students hypothesized, their subjects would say, “Dear Aunt Sally, how are you?” or “8 2 5 7 9 3.” Many participants did just this, suggesting that they tracked what made
sense
, not what entered one ear or the other.
⁴

If this result vitiates the claim of Broadbent, does it fit with
Darwin
’s theory? Can the result be explained by appealing to an inner jungle?

If you attend to what makes sense rather than to what strikes one eardrum or the other, the signals coming through your ears must be analyzed sufficiently to let you know what the signals mean. This implies that cognitive demons for analyzing meaning aren’t kept out by other demons blocking input to one of your ears.

Donald Broadbent was a professor at Oxford University (hardly a fly-by-night academy), so it’s a bit disquieting to think that an Oxford don could be dinged by anyone, let alone a couple of undergrads.
⁵
Even Oxford profs can be wrong, however, though being wrong isn’t quite so shameful as you might think. In science, being wrong but being enduringly stimulating is better than being right and quickly forgotten. One mark of the impact of a scientific theory is how long people dwell on it, even if only, in the end, to disprove it. Aristotle’s theory of motion was accepted for centuries before Newton supplanted it with something better. Newton’s theory of motion held for hundreds of years before it was overturned by Einstein. The fact that Aristotle was wrong and that Newton was also incorrect (at least beyond the range of everyday experience) doesn’t mean Aristotle and Newton are now derided. They’re both revered for the durability of their claims.

Should Broadbent’s model be rejected just because data from a particular experiment went against it? Maybe that experiment created an unusual situation, one that tapped into some unusual ability. Consider the following scenario, however.

Imagine that you’re at a crowded party, standing in the middle of a room, conversing with someone who strikes your fancy. What this person says captures your attention. You don’t have to try hard to focus hard on what he or she is saying. The other sounds in the room—the conversations, the laughter, the loud music—dissolve as you speak with this new acquaintance.

Then an odd thing happens. Out of the background, you hear your name. It isn’t yelled. It isn’t amplified through a megaphone. No one grabs your arm, tugs on you, and exclaims, “Ernesto, I’m talking to
you
!” Your name grabs you even though it’s mentioned quietly. Despite its low volume, it directs you to it. Though emanating from the background, the name attracts your attention the way a moth is drawn to a light in the darkness. The moth can’t help being pulled there. Likewise for your attention when your name sounds.

If your name pops out this way, you couldn’t have really shut out the auditory input containing it, could you? Instead, your name and, by implication, other sounds must have been subject to some sort of scrutiny. Cognitive imps in your head must have been listening all the time, and the cognitive imps linked to your name must have been pushier than others. When your name came in, those little demons must have jumped up and down feverishly, yelling “That’s me! That’s me! That’s me!” (metaphorically, of course).

The tendency to respond to your name even when it’s presented to an ostensibly unattended channel is called the
cocktail party
effect.
⁶
The term harks back to a time when people sipped cocktails at parties.

Is the cocktail party a reliable phenomenon, or does it only arise around martinis, whiskey sours, and Southern Comforts? It’s very reliable, as I show in a classroom demonstration in which I invite a student volunteer to come to the front of the class. I stand to one side of the student, read out loud from a book, and ask the student to repeat each word I say. This task is called
auditory shadowing
. With just a bit of practice—a minute at most—the student can perform the auditory shadowing task very well, provided the words he or she hears are familiar and the rate is comparable to his or her natural speech rate.
⁷

In the second part of the demonstration, I ask another student to stand on the shadower’s other side and read from another book. The shadower is now asked to repeat what the student reader reads and to pay no attention to what I say—something I figure the student has experience with, having sat through my lectures. In a matter of seconds, the shadower can do this. He or she can repeat what the other student says and is able to tune out what I say, or at least that’s what appears to happen. When I ask the shadower after the shadowing event what I read, he or she shrugs.

Then I demonstrate the cocktail party effect. In the midst of my reading, being careful not to raise my voice or do anything else to attract special attention, I utter the student’s name. If the student’s name is Ernesto, I say “Ernesto” as discretely as I can, as if the name were part of the text I’m reading.

When I say the student’s name, a remarkable thing often happens. The student immediately turns and looks at me, and says, “What?” or something similar. The effect is so dramatic that when the demonstration works, it gets gales of laughter from the students in the audience.
⁸

What does this demonstration reveal about the brain? It shows that it’s a jungle in there. In a jungle, there are dominance hierarchies. Similarly, when it comes to auditory perception, some representations (the dominant ones) need very little input to get highly activated whereas other representations (the less dominant ones) need more. Other representations are less dominant because their inputs fail to seize attention so reliably.

The way you can tell your name has special status is that you do something special when your name comes along. If you’re at a cocktail party, you stop listening to the person you’ve been listening to and look at the person who just mentioned you. If you’re standing in front of a class engaged in a shadowy demonstration with your cognitive psychology professor, you stop in your tracks and look at him or her when s/he refers to you. The fact that you stop what you’re doing and do something else suggests that attention permits action selection.

This view of attention as a vehicle for selecting actions was anticipated by the late-nineteenth-century philosopher and psychologist, William James, often called the dean of American psychology. In his encyclopedic work,
Principles of Psychology
, James wrote, “Everyone knows what attention is. It is the taking possession of the mind, in clear and vivid form, of one out of what seem several simultaneously possible objects or trains of thought. Focalization, concentration, of consciousness are of its essence. It implies withdrawal from some things in order to deal effectively with others.”
⁹
This definition does not restrict the effects of attention to the overt performance of selected acts but points, instead, to a more subtle, inner selectivity. Can this more subtle view be explored scientifically? To the extent that it highlights internal rather than only external experience, does this vitiate an action-centered view?

FIGURE 3.
The Rubin figure.

Consider an image that can be seen in one way or another, the so-called Rubin figure (
Figure 3
), named for the Danish psychologist, Edgar Rubin, who introduced it in 1915.
¹⁰
The Rubin figure can be seen either as a wine goblet or as two faces looking at each other. The most interesting feature of the figure is that you can see it in either of these ways, but not as both of them at once. The two perceptions flip back and forth. No matter how hard you try, you can’t see both of them simultaneously.

When I show the Rubin figure to students in my class, I ask them to raise their hands when they see a goblet and to lower their hands when they see the faces. For the students who are willing to participate in this exercise, their hands go up or down once every few seconds.
¹¹

A similar outcome is obtained with another reversible figure, the Necker cube (
Figure 4
), named after Louis Albert Necker, a Swiss crystallographer who first published an image of his famous cube in 1832.
¹²
The Necker cube is a two-dimensional rendering of a 3D wire-frame box. Because the 2D depiction is visually impoverished, lacking cues about the figure’s 3D orientation, it tends to be seen with one side close to the observer and the other side far from the observer. The two interpretations switch back and forth. The side that seems farther away suddenly jumps to the front, or vice versa. The flipping occurs again and again, as if of its own accord. Students who raise and lower their hands to show which side seems frontward change their hand positions at about the same rate as when they indicate their perceptual switches while viewing the Rubin figure.

FIGURE 4.
The Necker cube.

What accounts for this perceptual vacillation, and why am I speaking of it here in this discussion of attention? One reason is that perceptual instability suggests inner conflict. Two inner factions seem to duke it out. One wants one side of the Necker cube to be close; the other wants that side to be far. Each fights so hard for its point of view that your perception—your response to the image—gets jostled back and forth.

The other reason I’m speaking about the Necker cube is that the perceptual reversal of this stimulus also points to the importance of action selection. Seeing the Necker cube one way or the other is an action, albeit of an elementary kind. The action can take the form or lowering or raising your hand, or it can take the form of verbally reporting the position of a vertex of the cube: “Now it’s in front. Now it’s in back.” A reason for the vacillation may be that attention is not just for deciding how to interpret things but also for deciding what to do physically in relation to them. In the case of the Necker cube, it may be that you can see only one side of the cube as the front and one side as the back because if you were you to reach out and grab the cube in 3D, you’d be unable to grab one side in a near place and in a far place simultaneously. If you put your fingers around the back edge, your hand would be in the back, not in the front, and if you were to put your fingers around the front edge, your hand would be in the front, not in the back. Which edge is in the front or in the back is perceptually ambiguous in the 2D image, but the need to commit to a front reading or a back reading is
required
for the perception to serve action.
¹³

By this way of thinking, you don’t have to see attention as something mysterious. Attention exists, according to the view suggested here, because there are factions in the nervous system whose fates ultimately depend on the actions carried out.

An implication of this view is that any attentional bottlenecks that might exist in the nervous system should exist at the action side. A cognitive psychologist at the University of California (San Diego), Harold Pashler, has argued for this position.
¹⁴
He has described experiments that fit with the action-selection (or response-selection) hypothesis. Instructions to subjects in his experiments go something like this.

1. If a light appears on the left, press a button with your left hand.