|A book by|
William H. Calvin
UNIVERSITY OF WASHINGTON
SEATTLE, WASHINGTON 98195-1800 USA
The Throwing Madonna|
Essays on the Brain
Copyright 1983, 1991 by William H. Calvin.
You may download this for personal reading but may not redistribute or archive without permission (exception: teachers should feel free to print out a chapter and photocopy it for students).
Scanned, OCR'ed, and webbed -- but NOT proofread (14 Jan 97)
The Ratchets of Social Evolution
One needs to look near at hand in order to study men, but to study man one must look from afar.
JEAN JACQUES ROUSSEAU(1712-1778),
Essai sur l'origine des lingues.
Because the guides of human nature must be examined with a complicated arrangement of mirrors, they are a deceptive subject, always the philosopher's deadfall. The only way forward is to study human nature as part of the natural sciences, in an attempt to integrate the natural sciences with the social sciences and the humanities.
I can conceive of no ideological or formalistic shortcut. Neurobiology cannot be learned at the feet of a guru.
E. O. WILSON, On Human Nature
How is it that we can put ourselves in someone else's shoes, try to feel what he or she is feeling, experience things from another's perspective? Is empathy another one of those items on the human uniqueness list, something that we have but other animals lack? Or is empathy part of a continuous evolutionary development, a quality for which we can seek antecedents?
And so one must ask: What is the evolutionary advantage of being able to judge someone else's feelings? If you have ever lived in the Middle East and shopped each day in the souk, you will have seen such skills exercised and polished. There is nothing to prepare you for dealing with an American used-car salesperson like bargaining in the souk; you might even save the cost of a middle Eastern vacation if you practice enough. But upon your return, don't pick an expatriate Israeli, Turk, or Arab used-car salesperson with whom to haggle. For, like the shopkeepers in the souk, they are likely to be especially skillful at judging (from your eyes, from the expression on your face, from changes in your posture) whether hardening their price is likely to settle the deal or just send you off to try another dealer instead.
The oriental bazaar is just an organized version of the kind of bartering that has been going on for a long time, which is in turn built on top of the social give-and-take of primate societies. In the souk, the advantages to each party can be expressed in the artificial currency of shekels and diners; in primate society, judging another's emotions is useful for establishing dominance hierarchies which, in turn, often determine access to food, mates, or the best nesting spot. A monkey that is skillful at judging when he can get away with something without a fight, or good at judging another's willingness to fight or acquiesce, is more likely to perpetuate his genes. Similarly, a mother who is good at judging an infant's needs from subtle clues is more likely to have an infant that survives to perpetuate her genes.
So it comes as no surprise that the human brain has a specialized region of cerebral cortex for recognizing the emotions expressed on another person's face. It had been suspected that the right side of the brain had something to do with this, as right-brain stroke victims sometimes lack an ability to tell if a family member is happy or sad. Now a group of researchers at the University of Washington in Seattle have shown exactly where it is. They studied epileptic patients undergoing brain surgery with local anesthetics. The awake patients looked at a movie screen in the operating room, on which the pictures of faces were projected. The faces were those of actors and actresses acting out a particular emotion: fear, disgust, happiness, anger, sadness, surprise, and neutral. The patient's job was to name the emotion expressed, simply matching the face to a list of the seven emotions which appeared on the following slide. A simple job, which patients do quite well over and over as the slides flash by. But on some slides, their right brain is electrically stimulated while they are viewing the emotional face.
In general, nothing happens. The electricity is faint enough that the patient cannot tell whether it is on or off. Only when one small area of the exposed right brain is stimulated does something happen: the patient makes a mistake, such as calling the happy face "disgust" or the fear face "anger." The area which seems to specialize in emotional faces is located above the right ear, in the rear part of the middle temporal gyrus; typically, it has nothing to do with their epilepsy and seems to be part of the normal brain. When the applied electricity confuses the neural circuits there, the emotional judgment is faulty--even though perception and memory tasks are not affected. And, when a stroke or tumor damages that piece of cerebral cortex, the defect in judgment for emotional faces may persist for some time.
It is not yet clear whether this is truly an "emotional" specialization or just an area for advanced spatial-pattern recognition tasks which detects other features besides emotion; indeed, the emotional faces specialization was not even what the researchers were primarily researching--the slides were merely being used to distract the patients during a memory task for tilted lines. But when they started analyzing the right-hemisphere maps that resulted, the emotional faces task was always disrupted at the same patch of middle temporal gyrus in one patient after another. There was no consistent bias to the errors-- patients didn't tend to report happy faces as sad, or vice versa-- but just judged wrong when stimulated.
Is this a right-brain speciality, or is there a symmetric representation in the left brain of his emotional faces area? That's hard to say, as researchers are always busy studying language whenever they get a chance to study a patient's left-brain functions. In the one patient whose left brain was studied with the emotional labels test, stimulation had no effect on such judgments (although it produced naming errors and affected verbal short-term memory when language tasks were tested at the same middle temporal gyrus sites). Certainly, the right brain is "dominant" for judging facial emotions. If we look at another person, it is the emotion expressed on the right side of their face (in our left visual field, which goes first to our right brain) which we register more easily than the emotion expressed on the other side of their face. Such experiments are done with so-called chimeric faces, where parts of two different pictures of the same person are pasted together so that one side is happy and the other sad. If the slide is flashed on the screen only briefly, we will tend to see it as uniformly happy or uniformly sad, depending upon which one was on our left side.
Are there such specialized neural circuits in monkeys, or cats, or frogs? In particular, just how much of this neural specialization do we share with the great apes? All but 1 percent of our nonrepeated DNA sequences are identical to those of chimpanzees and gorillas, suggesting that we may have inherited a great deal in common from our common ancestor. Accounts of chimpanzee behavior in the wild suggest that facial emotions are used extensively for communication; indeed, misreading an angry expression for a happy one would be fraught with implications for fitness. So it would not be surprising to find a similar area in the brains of the great apes--but no one knows yet. This is one aspect of neurobiology (another being language) where the human research is ahead of the animal research.
Judging another's emotions is quite important if you are thinking of cooperating. The major approach to the cooperation problem comes from game theory--not its gambling applications, but the attempts to use it to account for social behaviors. At first glance, evolution seems to be all about perpetuating your genes through your own survival and that of your descendants. So how can we account for self-sacrifice, as when someone saves another person's life at the cost of his or her own? Or monastic traditions (altruism is not just human--consider the bees that spend all their lives supporting their queen, who is the only one to reproduce). Most explanations for this (e.g., kin selection theories) note the genes in common between queen and drone, the tendency for self-sacrifice to happen mostly for close relatives-- one's genes live on through their survival. How else could the denial of personal advantages evolve? But a broader way of looking at this is by asking how cooperation evolved, examining cooperation between neighbors as well as altruism between relatives, all of which suggests an important role for neural specializations which observe another person's face.
Cooperation in this sense occurs when you forego immediate advantages--where, for example, you yield the opportunity for food to allow another to eat. There can be long-term advantages to such cooperation: reducing the fights over food, with their spoilage of some of the edibles, the energy expended in threatening one another, the time wasted in maneuvering which could be used in searching for other food. But it works only if the other animal reciprocates next time. There are many advanced cooperative ventures (such as those discussed earlier as an alternative to chimp nut cracking in treetops); here, however, we are concerned with simple alternation and sharing, the first rung on the cooperation ladder.
The problem is often posed in simplified form as the Prisoner's Dilemma: two prisoners sharing a cell who can either fight each day over the single plate of food placed in their cell, or who can alternate or share the food. If they both cooperate from the beginning, fine. But if either defaults by taking advantage of the other, then both lose because of the continuing fights. And they are likely to get locked into noncooperation, eternally suspicious of each other. No question that cooperation is better--but it often appears to be unstable, not likely to survive. This simplified problem in socioeconomic theory has been extensively studied by Anatol Rapoport at the University of Vienna. Indeed, over 2000 papers have been written over the years by researchers on this dilemma and its philosophical implications!
Now if you are smart enough to figure out in advance that cooperation is better, and strike a deal with the other person, then there is no problem--but how did cooperation ever evolve in simpler animals if each animal is only looking out for itself (or secondarily for known relatives)? How could genes for cooperation gain a foothold if the rest of the population, the competitors for food, didn't have them and always "looked out for Number One"? In a world of individual success and failure, how were the seeds of cooperation nurtured?
One answer to this depends upon being able to recognize another animal as an individual--and being able to remember if he took advantage of your cooperation during your last encounter with him. If he did, you don't cooperate with him this time. Then you forget about it, not continuing to hold a grudge. This, you may remember from childhood, is called "tit for tat," and it is a simple yet powerful strategy which appears to have a good chance of evolving. Provided, of course, that the rewards of cooperation sufficiently exceed the usual yield from mutual noncooperation. And provided that there is a good chance of encountering the other individual again. Rapoport has applied it to the Prisoner's Dilemma, where there is an excellent chance of the two individuals having to deal with each other again in the future (in contrast, for example, to the chance of two strangers meeting on the sidewalk of a big city ever encountering each other again).
Provided that you cooperate on the first move, it solves the Prisoner's Dilemma: both prisoners will win through cooperation and will be unlikely to get locked into mutual noncooperation. But that doesn't necessarily say anything about evolution. Could "tit for tat" have evolved in a large group of interacting individuals? Is it stable once started, or could it be wiped out by enough cheaters? Robert Axelrod and William D. Hamilton have recently shown that Rapoport's "tit for tat" strategy sheds a great deal of light upon how cooperation could have evolved by Darwinian selection.
First of all, how could it have gotten started? It probably would have required a small group of individuals, likely to encounter one another again, some of whom had the "cooperativity gene." An isolated tribe with many relatives (and probably a lot of inbreeding so that even a recessive cooperativity gene would sometimes be expressed) would have been an ideal setup, but it really doesn't demand kinship in the manner of most altruism theories, just an isolated cluster of interacting individuals. It surely required a plentiful setting, such as the tropics, where making a living was easy, so that those who waited didn't starve. Later, when life inevitably got harder, the cooperation established in the bountiful years might have helped make the species more efficient in using scarce resources.
Could this trait have survived if the "tit for tat" individuals then joined with a larger group lacking the trait--the (probably apocryphal) trusting farm youth moving to the big city? Yes, say Axelrod and Hamilton, because the cooperating individual would lose only once to a me-firster before switching to noncooperation--but the advantages of cooperation would be manifest whenever he did encounter another cooperator. This lack of backsliding means, as they note, that "the gear wheels of social evolution have a ratchet."
The main requirement for the "tit for tat" strategy to survive is that an individual must not be able to get away with violations without others being able to retaliate effectively. In general, this means that you must be able to identify the other individual with sufficient certainty. And be capable of remembering him long enough to retaliate on the next encounter, whenever that is. And have a sufficient probability of indeed encountering the other individual again.
Obviously, a brain good at recognizing faces would be a better brain to benefit from the virtues of cooperation. There are areas of the human right brain that are used (though not exclusively) for remembering faces. Facial recognition is disrupted by electrical stimulation of the right brain and by many right-brain strokes; this involves areas of the right temporal lobe involved in memory for spatial arrangements (but not the emotional faces specialization noted earlier). For example, at the same site where short-term memory for faces is disrupted (typically the superior temporal gyrus and around the back end of the right Sylvian fissure into the parietal lobe a short distance), memory for the angle at which a line was tilted is also disrupted. If the patient has trouble remembering the line's angle of tilt, he or she will also likely have trouble remembering a picture of a face after an eight-second distraction task. So improved memory for spatial arrangements could be quite helpful to a cooperation strategy since it would aid facial recognition. Or vice versa: A cortical specialization shaped by the cooperation-linked success of remembering faces might also come in handy for remembering the leaning trees which mark the route home.
A good long-term memory is another prerequisite for cooperation success under "tit for tat." We tend to take memory for granted without inquiring into the evolutionary pressures that might shape memory capabilities. A simple type of memory of obvious utility involves food aversion: avoiding a food which made you sick once before. The Garcia phenomenon occurs when an animal (something as simple as a snail will do) is fed a novel food; then, several hours later, a drug is injected that makes the animal ill. The animal, presumably thinking that the novel food made it sick, will avoid that novel food upon the next presentation, even if it is weeks later. It takes only this one exposure to produce a long-lasting memory, and so food aversion is a favorite example of one-trial learning studied by psychologists. For omnivores like humans, always sticking new things in our mouth to try them out, such a memory mechanism for taste and smell would be under strong evolutionary selection.
Many food dislikes in humans can be related to such a fortuitous pairing of a food with an unrelated illness causing nausea, and such food aversions can last a lifetime (a corollary is: Don't eat your favorite food when you are catching the flu--you might spoil its taste forever!). But if a food is going to make you sick, it will usually do so within a few hours after you eat it (one-month delays, as for the infectious hepatitis mentioned in Chapter 7, are unusual). So you need to remember what you ate, no matter what, for as long as it takes to react to it. For the snail, this memory starts fading after several hours, unless permanently embedded by the delayed aversive reaction to the food.
It is also easy to see the utility of memories that last a year--the deer remembering its migration path between summer and winter pastures. Animals lacking the two-hour food memory capability, or the one-year migration path memory mechanism, might not survive very well. But what is the comparable utility of a memory that lasts a week? What strong evolutionary selection pressures might first shape such a capability? One candidate is the individual recognition needed for cooperation strategies (and dominance hierarchies) to work. One telltale sign of "tit for tat" involvement would be if facial memories for cheaters were much better than for other individuals--if, like the food taste which preceded the nausea, the cheater's face were burned into the memory circuits. At least until after the next encounter.
Elaborations on the basic "tit for tat" strategy can be even more beneficial: Suppose that you can estimate, from the emotions expressed on the other person's face, whether to initially cooperate with them or not? Such an ability, presumably aided by the right middle temporal gyrus emotional faces recognition area, might cut your losses before they started.
But other variants on "tit for tat" may present problems. For example, too extended a period of noncooperation following a violation could, just as in the Prisoner's Dilemma, lock the individuals into perpetual noncooperation (forgiveness has an evolutionary virtue, no less). Axelrod conducted a contest of strategies submitted by sixty-two game theorists from around the world to solve the cooperation problem for which the Prisoner's Dilemma is an elegant metaphor; a computer simulated 200 successive encounters using each strategy proposed--and kept score. Many of these strategies were much more elaborate than "tit for tat," utilizing techniques dear to theorists such as Markov processes and Bayesian inference.
Yet the childish "tit for tat" strategy emerged clearly triumphant: it was not only evolutionarily stable, but was the most robust strategy tested. Thus the search for improvements on "tit for tat" initially focuses upon improving the chances within its basic framework: improving individual recognition, improving long-term memory, improving first-encounter prediction of poor risk individuals, improving chances of future encounters (by, for example, clannishness--preferring a local community in which most interactions take place with known individuals), and keeping strategies flexible (something that human frontal lobes do: people with extensive damage there tend to go get locked into one particular strategy when approaching a problem and might never break their noncooperation strategy to try cooperation again, just as in the original Prisoner's Dilemma).
If this all sounds like a possible foundation for how social life developed--well, it may be just that. Attaining such an understanding is the strong motivation behind those 2000 papers on the Prisoner's Dilemma and all those game theory papers on "hawks and doves." Now there are some ways both for looking at the human brain's specializations and for appreciating the "social ratchets" that furthered them. The next decade should be a very interesting one for the interaction between neurobiology and social theory.
As Anne Roe and George Gaylord Simpson said in their book Behavior and Evolution: "It should by now be obvious that there is, indeed, a general theory of behavior and that the theory is evolution, to just the same extent and in almost exactly the same ways that evolution is the general theory of morphology." What may shape a wing may also shape behavior.
The Throwing Madonna:
Essays on the Brain (McGraw-Hill 1983, Bantam 1991) is a group of 17 essays: The Throwing Madonna; The Lovable Cat: Mimicry Strikes Again; Woman the Toolmaker? Did Throwing Stones Lead to Bigger Brains? The Ratchets of Social Evolution; The Computer as Metaphor in Neurobiology; Last Year in Jerusalem; Computing Without Nerve Impulses; Aplysia, the Hare of the Ocean; Left Brain, Right Brain: Science or the New Phrenology? What to Do About Tic Douloureux; The Woodrow Wilson Story; Thinking Clearly About Schizophrenia; Of Cancer Pain, Magic Bullets, and Humor; Linguistics and the Brain's Buffer; Probing Language Cortex: The Second Wave; and The Creation Myth, Updated: A Scenario for Humankind. Note that my throwing theory for language origins (last 3 essays) has nothing to do with the title essay: THE THROWING MADONNA is a parody (involving maternal heartbeat sounds!) on the typically-male theories of handedness.
Many libraries have it (try the OCLC on-line listing, which cryptically shows the libraries that own a copy), and used bookstores may have either the 1983 or the 1991 edition.
- Powell's Books in Portland lists used copies in their web database.
Email || Home Page || The Calvin Bookshelf || Table of Contents || End notes for this chapter || Continue to read next chapter