Genetic influences also account for a major portion of the continuity in behavior or personality over long periods of time.10 This finding has a very important implication. The fact that some personality characteristics show up early and persist through childhood and adulthood does not mean that environmental effects on personality must occur at an early age. The evidence suggests that these early-appearing and stable characteristics are due to inborn predispositions rather than to early experiences. This means that environmental influences on personality might occur at a later stage of development than many theorists have assumed.
The problem is not how to account for the continuities and consistencies in behavior: it’s to explain why there isn’t more continuity and consistency, in view of the .45 heritability of personality characteristics.
One answer is that genes can contribute, not only to stability, but also to change. Though the genome itself remains the same throughout life, some genes are turned on early while others (like the ones that determine breast size in women or baldness in men) don’t begin to function until later in development.
Moreover, behavior in different contexts may be influenced by somewhat different suites of genes. The researchers who studied activity level found that genetic influences accounted not only for all the similarity in behavior between laboratory and school, but also for some of the differences. There were genetic influences that were specific to one setting or another, perhaps because the activity being measured was of a different sort—fidgeting in the lab, say, versus participation in active games on the school playground.11
But genetic influences account for only a minor portion of the differences in behavior from one setting to another. Mischel assumed that most of the differences result from learning and I agree with him on this. As he put it, “Individuals discriminate sharply between situations.”12 Those children who behaved in an obnoxious fashion with their parents but not with their peers might have learned that their parents would put up with it but their peers would not. Those children who behaved in an obnoxious fashion with their peers but not with their parents might have been big and strong enough to bully the other kids on the playground but not big and strong enough to bully their parents. In the previous chapter I described studies that showed that firstborns are more aggressive than laterborns at home but not at school. Children discriminate sharply between situations.
But what about babies? Ah, now we’re getting to the nitty-gritty.
Generalization or discrimination? It’s something you would have to decide if you were designing a baby. Since no two situations are ever exactly alike, learning would be impossible without some generalization. And repeating a previously learned behavior in a new (but similar) situation would presumably be easier than starting all over from scratch. On the other hand, repeating the previously learned behavior could be hazardous, if it turned out to be inappropriate in the new (though similar) situation. Thus, the designer of the baby has to balance the risks of overgeneralization against the inefficiencies of undergeneralization.
But undergeneralization is a serious disadvantage only if storage capacity is at a premium. If the baby’s mind were unlimited in its storage capacity—in its ability to acquire and retain new patterns of behavior, properly labeled as to where and when they should be trotted out—then undergeneralization would appear to be the safer policy.
The evidence indicates that this is in fact the way the baby’s mind was designed. A baby is far more likely to err by failing to generalize than by generalizing too freely.13 Evolution provided the young of our species with a brain so large it can barely make it through the birth canal, and then shrugged her shoulders when the question came up about the inefficiencies of undergeneralization.
As demonstrated by the experiments of the developmentalist Carolyn Rovee-Collier and her colleagues, a young baby’s initial inclination is to discriminate, rather than to generalize. Rovee-Collier begins her experiments by teaching a baby a simple trick. She ties one end of a ribbon to the baby’s ankle, the other to a mobile suspended over the crib. When the baby kicks his foot, the mobile jiggles. Most babies are delighted to discover they can make the mobile jiggle; they show their enthusiasm by greatly increasing their rate of kicking the beribboned foot. Moreover, they will remember this trick several days later: if they’re shown the mobile again, they will again kick the appropriate foot. But they will do this only if the experimental setup hasn’t been changed. If the blue doodads hanging from the mobile have been replaced with red doodads, or if the crib has been rolled from the bedroom to the den, the baby will gaze cluelessly at the mobile, as though he had never seen such a thing in his life.14
Unexpected failures to generalize have been demonstrated in older babies, too. When babies first start to crawl at eight or nine months of age, they will attempt to crawl down slopes that are much too steep for them to negotiate. The developmentalist Karen Adolph trained babies by letting them crawl on an inclined plane—a carpet-covered walkway—that could be adjusted to different angles. When the walkway was too steep, the babies would tumble down it, landing safely but ignominiously on a cushioned surface at the bottom. After repeated experiences with the experimental setup, the babies learned which slopes they could crawl down without falling and which were too steep.
Then these babies learned to walk. Incredibly, there was no transfer at all from crawling to walking. In the first weeks of walking, the babies attempted to toddle down the same steep slopes that had caused them to plunge headlong as crawlers, with the same results.15
The studies of Rovee-Collier and Adolph have to do with motor learning. Their findings tell us something about the learning mechanism that enables babies to acquire physical skills. The mechanism evidently comes with a factory default setting: a bias against generalization. And yet we know perfectly well that generalization occurs. “Obviously,” as Mischel observed, “a person does not have to learn everything afresh in each new or slightly different situation.”16 You walk into a dark hotel room and automatically grope for the light switch. You don’t have to learn afresh, in every new room you enter, how to turn on the lights. The toddler who learns to drink from a cup at home doesn’t have to relearn the skill in the day-care center.
So the question is not whether generalization occurs, but when and how it occurs. When do babies regard two situations as “same” and when do they regard them as “different”? How do their experiences teach them that two situations they initially regarded as different are in fact equivalent?
Rovee-Collier’s experiments are again informative. To teach babies to generalize the foot-kicking trick from one mobile to another or from one setting to another, she discovered, all you have to do is to vary the training: train them with several different mobiles or in several different locations. When the training setup is varied in this way, the baby learns that the rule “kick your foot to make the mobile jiggle” holds true under a variety of conditions. Babies learn this readily but they don’t come into the world already knowing it, even though it seems obvious to us that a trick that made a mobile jiggle in the bedroom should also make it jiggle in the den.
Interesting, but it still doesn’t solve the problem. If varying the training setup teaches babies to generalize the foot-kicking trick, how come they don’t learn that on their own, given that no two situations are ever exactly alike? Even if the same mobile is hanging overhead and the crib hasn’t been moved, other things are bound to have changed. Maybe the first time the baby saw the mobile, sun was streaming through the window; the second time it was a rainy day. Why does a baby spontaneously generalize the foot-kicking trick from a brightly lit bedroom to a dimly lit bedroom, but not from the bedroom to the den?
Two situations are never exactly alike in every detail, and that is why, as Steven Pinker has pointed out, there must be an “innate similarity space.”17 Babies must be more tuned in to certain kinds of differences than to others. The innate similarity space determines whether an infant will initi
ally regard a new situation as the same as a previous one or as different. Thus, to know how the baby will cut up the continuum of experience into different “situations,” we have to know something about the baby’s mind. As the psychologist Peter Gray put it, “You can’t predict to what degree an individual will generalize from one stimulus to the next unless you understand something about the individual’s mental concepts.”18 If the two stimuli evoke different mental concepts, generalization may not occur even if the stimuli are very similar in size, shape, and color. If the two stimuli evoke the same mental concept—which might mean only that the individual can’t tell them apart—the individual is likely to respond in the same way to both. Babies can learn from experience that the same mental concept applies to two situations that initially occupied different positions in the innate similarity space.
The research on infant attachment provides some useful information about the baby’s innate similarity space. I alluded to attachment theory earlier, when I made fun of the idea that your life is going to be blighted if you and your mother didn’t happen to hit it off during the first twelve months of your life. Now I will take a more serious look at it.
According to the attachment theorists, a baby develops a mental concept—they call it a “working model”—of his relationship with his mother. The way the researchers assess the quality of this relationship is by putting a year-old infant and his mother into a unfamiliar laboratory room, having the mother leave the baby alone for a short time (most babies get upset at this point), and then watching what happens when the mother returns. According to attachment theory, a baby who has developed a “secure attachment” will greet his mother joyfully and will allow her to soothe away his tears. One whose attachment is insecure may ignore his mother, or keep on crying, or alternate between clinging to her and pushing her away. Attachment researchers have shown that a baby is more likely to behave in a securely attached fashion with his mother if she has taken good care of him, during the preceding months, by being responsive to his needs.19
There is nothing wrong with the attachment theorists’ explanation of the baby’s behavior; I think the baby really does form a mental model of his relationship with his mother and develop expectations about how she will behave. On the basis of his past experiences with her, he may regard her as someone he can rely upon to make him feel better when he is upset; alternatively, he may recognize that she is not useful under such circumstances. But—here’s where I part company with the attachment theorists—his life is not going to be blighted if he comes to the latter conclusion, because his mother is not going to be the only person in his life and he doesn’t expect people to be all alike. Babies who have received nothing but kindness from everyone they encountered during their first twelve months will nonetheless cry out in fear if a stranger tries to pick them up.
The fundamental attribution error shows that adults expect individuals to be consistent—more consistent than they really are. The work of the attachment theorists shows that babies have the same bias. The laboratory setting is brand new to a baby who participates in one of these studies—he’s never been in this room before—and yet he expects his mother to behave the way she has behaved in more familiar settings. She may be wearing different clothing or a different hairstyle but she’s still his mother, and the baby expects, on the basis of his prior experiences with her, that she either will or will not be helpful. But he doesn’t generalize these expectations to other people, even if they’re the same sex, age, and color as his mother. He doesn’t reach up his arms to the stranger.
Researchers have found that the babies of mothers who are suffering from postpartum depression behave in a solemn, subdued fashion when they’re with their mothers. When they’re with other familiar caregivers, however, these babies behave normally: they are much livelier. The subdued behavior, according to the researchers, is “specific to their interactions with their depressed mothers.”20
Whether behavior is generalized from one situation to another depends on whether the situations are regarded as the same or as different. For humans, probably the most salient cue to “same” or “different” is the cast of characters. A change in dramatis personae means that this must be a different play, with a different script.
Babies, as I said in chapter 1, are born with a tremendous interest in learning to tell people apart and in collecting information about them; the people-information acquisition device is ready to kick into action on the day it is delivered. A baby whose mother is depressed doesn’t expect everyone to be depressed; one whose mother loves him doesn’t expect everyone to love him. Babies can be securely attached to one caregiver and insecurely attached to another.21
Just as other developmentalists find weak correlations between a child’s shyness or obnoxiousness in one setting and another, attachment researchers find weak correlations between the security of a toddler’s attachment to his mother and his later success in other social arenas—friendships with peers, for instance.22 My interpretation of all these correlations is the same: they result, not from generalization, but from genetic influences on the measured outcomes. The methods used by attachment researchers provide no controls (or inadequate controls) for genetic differences among their subjects. I take the correlations they find as evidence that characteristics the child is born with, including physical appearance, will help or hinder him in a variety of social situations. The cute face or ready smile that made his mother dote on him23 will also stand him in good stead with other people, including peers.
Early experiences may have an effect on later behavior in another way—a way that doesn’t fit the description of learning plus generalization. Early experiences, or the absence thereof, may affect the neurophysiological development of the brain. Since every kind of learning must have neurophysiological effects on the brain (strengthening synapses or whatever), you may be wondering how I could make a distinction between these two kinds of processes. Perhaps the best way to explain it is to give you some examples.
The first has to do with vision. Neuroscientists David Hubel and Torsten Wiesel won a Nobel Prize for their research, which showed that it is possible to produce permanent changes in the visual system of a cat by restricting its vision in various ways during a “critical period” in the early weeks of its life. Patterned stimulation to both eyes is required for normal development of the visual system. If a kitten is permitted to see out of only one eye during the critical period, vision in the occluded eye will be permanently impaired. The problem is not in the eye—it’s in the brain. The brain area originally designated to serve the occluded eye has evidently been reassigned to other duties.24
A second example has to do with language. Babies who aren’t exposed to a language—spoken or signed—during the first few years of life never become highly skilled communicators. Generally the way this happens is that a baby is born deaf and the parents either aren’t aware of it or don’t know what to do about it, so that the child’s introduction to sign language (or to spoken language, if the hearing deficit is correctable) comes late. Deaf children who aren’t exposed to sign language until school age never become proficient signers. The most eloquent users of sign language are the deaf offspring of deaf parents, who started learning it in infancy. In those rare and tragic cases where children have no opportunity to learn a language before they reach puberty, they never get the hang of it at all.25
My final example is attachment. There is some evidence that human babies who fail to form an attachment during the first two or three years of life may be permanently impaired in their ability to form close relationships. Under ordinary circumstances, all human babies become attached; that’s why attachment theorists speak of secure versus insecure attachments, rather than attached versus unattached babies. Even babies who are abused by their mothers become attached to them, though there is a greater chance that the attachment will be insecure.26 An unattached child is likely to be found only in the worst kind of institutional setting, where ther
e is frequent turnover in caregivers. A recent study of some children who had spent their first two or three years in Romanian orphanages found that some—but not all—showed persistent abnormalities in social behavior. At age six these children tended to be indiscriminately friendly to all adults, even strangers, and they failed to look to their adoptive parents for reassurance at times when most children would do so. But, as I said, not all the orphanage children showed these symptoms. The researchers weren’t sure why some children appeared to have recovered fully from their early deprivation. Perhaps they were genetically more resilient, or perhaps they had managed to find someone to become attached to in the orphanage. There is evidence that attachments to other children can serve, at least to some extent, as a substitute for an attachment to a caregiver.27
The same is true for monkeys. As I mentioned in chapter 3, baby rhesus monkeys that have been separated from their mothers develop into reasonably normal adults if they are raised with peers. However, researchers have found subtle differences between mother-reared and peer-reared monkeys.28 Such differences do not surprise me; peer-rearing does not fall within the normal range of environments for a baby monkey. Because an infant mammal could never survive in the wild without a mother, its brain “expects” a mother to be present. The infant mammal’s brain relies on the presence of a mother, the same way it relies on patterned visual stimulation, to fine-tune its development after birth.
No Two Alike Page 17