It's a Jungle in There: How Competition and Cooperation in the Brain Shape the Mind (24 page)

In one experiment that supported ideomotor theory, people wrote a single letter of the alphabet as quickly as they could. In one condition, the letter to be written was designated visually. In a second condition, the letter to be written was designated auditorily. In a third condition, the same participants
said
the letter after seeing it. In a fourth condition, the same participants said the letter after
hearing
it. In general, the time to start producing the letter was less when the letter was going to be produced in the same modality as the modality signaling its production. The interaction between the modality of the stimulus and the modality of the response—quicker responding with speech for a heard stimulus and quicker responding with writing for a seen stimulus—can be explained with ideomotor theory: Perceptual inputs that resemble perceptions produced by actions trigger those actions more quickly than inputs that don’t resemble those actions’ perceptual outcomes.
⁴⁹
Relatedly, actions that yield simple perceptual consequences turn out to be easier to perform than actions that yield complex perceptual consequences.
⁵⁰

Another source of support for ideomotor theory pertains to mirror neurons. These are neurons that fire when you observe actions you can carry out. For example, if you’re a monkey and a neuron in your brain fires when you eat a banana, if that same neuron fires when you see someone else eat a banana, that neuron can be considered a mirror neuron. Mirror neurons have not yet been directly observed in humans, but many researchers are convinced they exist in people.
⁵¹

The existence of mirror neurons has led to interesting speculations about their role in cognition. According to some researchers, mirror neurons might provide the neural substrate for empathy. The apparent indifference to the humanity of others by some has been ascribed to the possible absence or malfunction of mirror neurons in their brains.
⁵²

Regardless of whether you think mirror neurons underlie empathy, the existence of such cells makes a lot of sense in the jungle scenario. It’s
congenial with the jungle hypothesis that a special niche is occupied by neurons that fire in the way that mirror neurons do, being activated both by the perception and by the production of particular actions. The reason this expectation is congenial with the jungle view is that neurons that contribute to the production of actions also benefit from the perceptions those actions provide, as suggested in
Chapter 2
. To the extent that rich opportunities exist for neurons with these properties, it makes sense that mirror neurons should grace the brain. I would venture to say, based on the arguments just given, that the presence of mirror neurons will turn out to be more the rule than the exception in the animal kingdom.

Another expectation one might have about mirror neurons is they have the status they do at least partly by virtue of experience. If the production of actions benefits from the perception of those same actions, and vice versa, then the more often actions co-occur with their associated perceptual consequences, the greater the expected strength of the connection between them.

Several studies have yielded data consistent with this hypothesis. In the most famous one, dancers with different kinds of expertise—dancers of ballet and dancers of the Brazilian dance, capoeira—watched videos of ballet dancers or capoeira dancers. The ballet dancers’ brains showed more activity when they watched ballet dancers than when they watched capoeira dancers, and the capoeira dancers’ brains showed more activity when they watched capoeira dancers than when they watched ballet dancers. This outcome fits with the hypothesis that ballet dancers have mirror neurons for ballet but not for capoeira, whereas capoeira dancers have mirror neurons for capoeira but not for ballet. The brain areas where the differential responses were recorded were consistent with the known anatomy of mirror neurons in monkeys.
⁵³

Personal experience also points to the plausibility of mirror neurons. I’ll report some of my own personal anecdotes that are suggestive of mirror neurons in my own brain. None of the anecdotes proves that my brain has mirror neurons, but I find it useful to reflect on the stories to keep alive the possibility that they do.

One occurred while I was giving a lecture. During the lecture, while holding a microphone in one hand, I held an empty water glass in the other. A gentleman in the first row kindly offered to fill the glass for me. He indicated, through a gesture he made, that he would gladly pour water from a large pitcher into the glass I was holding with one hand while I held the mike with the other hand and continued to speak. While I watched him pour the water into the glass, I stopped talking. I couldn’t talk! There was no way I could! That’s how great my concentration was, though, of course, I didn’t need to concentrate on the pouring because I wasn’t pouring. All I had to do was stand still, or perhaps not even that, for if I moved, the generous gentleman would
have compensated for my inadvertent moves and kept the water flowing into the glass. The truth was that I was also pouring the water, if only vicariously. My mirror neurons, if I have them, were triggered and their activity spilled over into my behavior.

A second example came from an experience I had with my wife, Judy, in connection with filling out a long, complicated form for a financial transaction. Each of us had to sign and date the form on a number of pages. I had done all my signing. Then I brought the form to Judy and stood by her as I watched her leaf through the pages, searching for the places to sign. As she prepared to turn one page after another, I found myself licking my right index finger, getting ready to turn the next page as well, albeit in thin air and with no contact with the page itself.

A third example concerned my walking pace as I saw a car approach an intersection as another car approached on a perpendicular street. High bushes prevented the two drivers from seeing each other, except at the last moment. The car I was especially concerned about was traveling at a high speed. In my mind’s eye I saw the fast-approaching car about to crash into the other car. What did I do? I slowed my walking! In hindsight, this made no sense. Slowing my walk wouldn’t slow the fast-moving car, but slowing is what I did, apparently reflecting my desire for the driver of the fast-moving car to slow down. Here again, I vicariously carried out an action, quite possibly because of mirror neurons.

A final example concerns an experience I had at a dinner with some colleagues. One person there, a normally loquacious cognitive psychologist, had laryngitis and couldn’t speak. Gradually, the two people to one side of her got into a conversation that left her out, and the two people to the other side of her (one of whom was me) got into a conversation that also left her out. We had urged our hoarse friend to write down what she wanted to say, but after a couple of attempts at this, her motivation waned. I noticed this and thought of mirror neurons, so I suggested we write down messages to her, which of course was unnecessary because all of us had perfectly healthy voices. Yet the ploy worked. When our friend was handed a note, she smiled and wrote back. This went on for the rest of the evening. The method might not have worked if it weren’t for mirror neurons.

None of these examples proves that there are mirror neurons in the human brain. Still, they are consistent with the hypothesis that such neurons exist. That they do is, at the very least, expectable from the perspective that neurons opportunistically form bonds with other neurons if such liaisons are useful.

The final class of phenomena I’d like to consider in this chapter on action concerns the parallel, interactive nature of action control. Consider typewriting. For typewriting to work, your fingers must land on the keys in the right order. The sentence I just typed, for example, required that I hit the shift key first, the “f” key second, the “o” key third, and so on. If you think about how this behavior is managed, you might suppose that a typist’s fingers start moving toward their respective keys in the same order as the keys are pressed. The first finger to move might be the one that hits the first key, the second finger to move might be the one that hits the second key, and so on. This expectation is reasonable, except that it turns out to be wrong. As shown in movies of skilled typists, their fingers start heading toward their respective targets as soon as they can.
⁵⁴
Each finger seems to be driven by a little demon that’s intent on moving its finger to that finger’s next target as soon as possible. Saying this another way, each target tries to attract its associated finger as soon as it can, wooing it like a Greek-myth siren.

What’s to prevent the fingers from colliding? Why don’t the fingers just launch
en masse
to wherever they must go? The answer, according to a model of the serial ordering of behavior that I find particularly compelling, is that behaviors that are supposed to occur late are inhibited more than behaviors that are supposed to occur early (
Figure 10
). According to the model, it’s the degree of inhibition among behaviors that defines their serial order.
⁵⁵
In the model, the last element is inhibited by the element before it, by the element before that, by the element before that, and so on. The penultimate element is inhibited by all the elements preceding it but not by the one to come afterward. Similarly, the element before the penultimate element is inhibited only by the elements preceding it but not by the two last elements.
⁵⁶

This model is elegant because its principal assumption is that serial order is embodied in patterns of inhibition. It doesn’t say that response elements are linked to serial position tags, such as “position number 1,” “position number 2,” and so on. Claiming that there are position tags begs the question of how serial positions are represented. So does saying that keystrokes are defined by their positions in a hierarchical control structure, such as a tree whose bottom nodes can be read from left to right. That symbolism might be helpful for teaching, but it leaves unanswered the question of how the behaviors are actually controlled.
⁵⁷

Building on this inhibitory model of behavioral control, David Rumelhart and Donald Norman, working together at the time at the University of California (San Diego), modeled the timing of keystrokes in typing.
⁵⁸
The details of their model are less important than its general flavor, which, again, is “jungle-like.” Each key beckons its associated finger or, said another way, each finger tries to reach its attracting key as soon as it can, being held up only by inhibition from other responses and by physical interference.

FIGURE 10.
Ordering of events with an inhibitory network.

After Rumelhart and Norman introduced their model, a great many studies showed that motor output is not a single, unitary activity but instead reflects a kind of conglomerate of pulls and pushes, all occurring simultaneously. So, for example, when people reach for a target, their hands can be heavily influenced by the sudden appearance of another target.
⁵⁹
Similar results have been obtained for eye movements and for speaking.
⁶⁰
Tug-of-war is the rule rather than the exception when it comes to action control. Effectors are drawn in parallel to targets in the external environment.
⁶¹

All of the foregoing suggests that the way we move comports with the idea that it’s a jungle in there. Warring factions and, not to be forgotten, mutually supportive factions yield the patterns of activity that characterize our voluntary and involuntary actions. Theories of action control that proffer headmasters or headmistresses are unnecessary. Pushes and pulls, operating at the neural level, can account for the dynamics of action control, or that, at least, has been what I have hypothesized here.

As we approach the end of this chapter, I ask you to turn your mind back to the Center for International Research (ZiF), where this chapter began. At the conference I attended there back in June 2010, one of the speakers reviewed the research of a leading scientist in the field of perception and action. That earlier pioneer was Erich von Holst, the director of a Max Planck Institute in Germany. von Holst showed that the signaling of perceptual centers by motor centers, as discussed here in connection with blink suppression, saccadic suppression, and tickling, held in animals as well as humans.
⁶²
He also showed that limbs interact so that, in humans as well as in fish, if a limb starts wiggling more quickly than it did before, other limbs wiggle more quickly as well. The nature of the interactions is remarkably similar over species, attesting to the deep-seated nature of these effects.
⁶³

In the lecture at the ZiF conference, the speaker summarizing von Holst’s contributions, Michael Turvey, a distinguished researcher at the University of Connecticut and Haskins Laboratories (New Haven, Connecticut), reminded the audience that von Holst summarized all of his work with one pithy statement: