This issue addresses a important need in practice - knowledge the pediatrician should have to guide parents in rearing their children. Do you ever wonder if the advice you give is correct? If you are like me, you had relatively little training but give guidance freely based on personal experience and what has been in the literature. Unfortunately, the scientific basis for understanding what parenting does and how it works is relatively sparse. To put parenting guidance on a scientific basis, we need more theoretical constructs like Sameroff's transactional model of regulation. This is described more than once in this issue, and the paper by Regalado and Halfon (pp 3137) goes in to some detail, with an example of how it works. This model describes relationships as dynamic, influenced by continuous give and take interactions between individuals (in this issue parents and children or between siblings) and the environment. Regalado and Halfon point out that this "transactional model of regulation provides a useful framework for conceptualizing parenting problems and types of interventions." But there is a dearth of other principles like this.
The following is a search of the essence of parenting collected from sociologie and anthropologie reviews in Science, with a sprinkling of the author's ideas. The latter are primarily directed at tying the concepts together and to parenting. They are usually identified by parentheses.
THE SUCCESS OF HUMANS IS BASED ON THE TRANSMISSION OF KNOWLEDGE
Humans have had unprecedented success in dominating the environment during the 100,000 or 200,000 years that modem mankind has been here. Humans are closely related to chimpanzees. Genetically, we are 99% identical, and a visitor from outer space would classify humans as a third species of chimpanzee.1 O'Neill, reporting on the meeting "What is Life?" held in Dublin, Ireland, in 1993, traced the origin of our success by looking at what sets us apart from chimpanzees and other species. Our success was felt to be based on our ability to create or invent; the first signs of this are evidence of tool-making about 50,000 years ago. But human inventiveness required the development of language.1 Without language, each individual would have to start from scratch without a knowledge base to use for discoveries and inventions. So it is more accurate to state that human's success depends on our unique ability to transmit knowledge through language. As a consequence, every human can potentially leam from the experience of all others, alive now or at any time in the past. Put in another way, the invention of the modem jet airplane began with the discovery of fire and metallurgy.
How does this transmission occur? Cavalli-Sforza and colleagues indicate that it happens on three levels: vertical from parent to offspring (ie, by parenting), horizontally between members of the same generation (siblings as well as all other contemporary individuals), and obliquely between generations and by unrelated individuals. However, the family is at the foundation of this: information was first transmitted from parents and older siblings to children both in the evolution of humans and in the development of each child throughout time. But, as humans evolved, they increased the horizontal and oblique transmission of knowledge in ways that increased efficiency and were not dependent on the family. This included written language, schools and universities, the printing press for books and newspapers, and more recently, radio, television, and the worldwide web. Because the products of these are additive, knowledge has grown at an exponential rate.
Actually, more than just knowledge is transmitted to individuals by horizontal, vertical, and oblique methods. Knowledge implies the information is true, correct, and understandable. In fact, incorrect information, biases, prejudices - all concepts of culture are transmitted in these ways. So the transmission of information is actually the ransmission of culture.
THEORIES OF CULTURAL TRANSMISSION AND EVOLUTION, AND THEIR EFFECTS ON PARENTING
In 1971, CavaHi-Sforza noted a detailed analogy between biological evolution through DNA and cultural evolution by the transmission of ideas and concepts through language.2 His ideas and the ideas of others reviewed in recent literature2'4) have led to a theory of cultural evolution. At the core of this is the concept that the knowledge and concepts that are transmitted by language are subject to selection according to their value to the individual, family, and culture. The force that drives biological evolution is selection: those individuals who inherit the most advantageous genetic traits are most likely to pass these to the next generation.
Selection also drives cultural evolution, but on multiple levels. Like biological evolution, the individual who inherits the most advantageous cultural "traits" is more likely to survive and produce offspring who will "inherit" similar knowledge and concepts by vertical/parenting and other means of transmission. At another level, selection occurs because individuals intentionally choose those concepts and adopt those inventions that they believe will be most advantageous to themselves and their families. Man alone, as far as we know, understands his own mortality and will make conscious decisions to avoid pain and prolong his an his progeny's lives and well-being. On a third level, cultural concepts and cultures themselves compete with each other. The culture with the mosteffective knowledge and beliefs will survive and grow (ie, be selected) because its individuals thrive and because individuals of less-effective cultures may choose or be converted through conquest to join a culture judged more effective by selection. Simon points out that transmission of information between individuals within a culture enhances the creativeness, the productivity, and the cooperation of individuals within the culture and thus contributes to its success and survival.4
Cultural evolution can be compared with genetic evolution through six evolutionary forces:3 (1) By cultural mutations, which includes intended innovations for a purpose as well as random errors that are imitated into culture. Genetic mutation is only by copy error, which is assumed to be random. (2) Cultural transmission involves social learning from parents and non-parents whereas genes are inherited only from parents. (3) Cultural migration includes the movement of people with their ideas or the movement of information without people, whereas biological migration includes only the movement of people with their genes. (4) Cultural drift is due to sampling fluctuations in finite populations, common to cultural and biological evolution. (5) Cultural selection is unique to cultural evolution because it involves decision making by individuals. (6) Natural selection, or consequences of Darwinian fitness is common to both types of evolution.
How does parenting fit into cultural evolution? These two are tied together in dual ways. First, cultural evolution is dependent on parenting for vertical transmission from parent to child. Parents also influence horizontal transmission between siblings and try to choose other horizontal or oblique sources for their children. Second, parenting is shaped cultural evolution. The culture may determine parenting practices through religious beliefs and through political, economic, and legal customs relating to marriage and the family. In other words, we inherit parenting behaviors from our parents and our culture. These parenting customs have been shaped by thousands of years and probably consist of behaviors that we understand, think we understand, don't understand, and/or don't even realize we are carrying out. Differences in customs for rearing of boys and girls may serve as an example of this complexity.
Is there scientific data supporting that vertical transmission of culture occurs from parent to child? This concept remains largely untested, but CavalliSforza and colleagues studied quantitative aspects of vertical transmission of six cultural "traits," and found strong evidence of parent to child transmission for religious and political qualities.2
THE INFLUENCES OF BIOLOGICAL EVOLUTION ON PARENTING
A comparison of the physical and physiologic characteristics of humans compared with nonhuman primates and other mammals provides some insight into how men and women genetically evolved to parent. There are two primary strategies for reproductive success in nature: the shotgun or fast-track approach ("r selection") produces a large number of offspring who will then have little or no parental protection and a high mortality rate while young. At the other end of the spectrum is the rifle approach (K selection) which produces a small number of offspring but requires the investment of parenting to protect and develop the small numbers of young. In general, species using K selection compare with those using r selection in being larger and having larger brains for their size and a longer adult life span.5·6 Individual species utilize varying degrees of both strategies, but primates have developed a higher form of the K approach. Among all species, humans have developed the highest form of the K strategy. By this view, we evolved to maximize parenting (and thus transmission of knowledge and culture). Our physiologic adaptations to this include a relatively long gestational period, thought to be important to permit maximal brain growth before birth,7 a long period of infancy and sub-adult life to extend time for parenting (and the intrafamily transmission of culture) before reproductive age is reached, and a longer total life (perhaps to allow time to transmit culture and allow it to be tested by selection). One can trace the development of these adaptations through rising orders of primates and they progress as expected.5
As noted previously, humans and chimpanzees share all but 1 % of their DNA, and chimpanzees are also most closely related to us in terms of life stages for parenting. The gestation period of the chimpanzee is 34 weeks, infancy and sub-adult stages are longer in the chimpanzee than in other primates, but humans require about 5 years more than chimpanzees to reach reproductive age and the total life expectancy of humans is about 30 years longer. The chimpanzee infant does not begin to walk until 6 months, rarely goes more than a few yards from his mother until over 9 months, and will not survive if the mother dies before he is at least 5 years old. Females reach sexual maturity at about age 10. Chimpanzees (and other nonhuman primates) are physiologically able to reproduce until death. Woman is unique in having a life span that extends for 30 or more years beyond the time she can have children. (This may allow time to rear the last child and perhaps help parent grandchildren and other relatives before death).
Humans and some primates have developed in another way that improves K selection strategy.5 Parental investment in rearing (including the transmission of culture) can be increased qualitatively and quantitatively by having two parents instead of one. Monogamous mating (humans qualify for this - the term is not absolute) is only found among primates in which the male is directly involved in parenting and thus seems to be associated with keeping the father around to help raise the young. Human sexuality is unique in that sex can occur throughout adult life, and is not bound to an estrous cycle. In fact, it can continue long after menopause. The physical distinctions between men and women are also more pronounced than those between males and females of nonhuman primates. These differences may facilitate the bonding between parents and the continued parenting investment of the father.5 (They also parallel the social forces that cultures have evolved to keep parents together. And having two parents from different backgrounds instead of one is especially important for the vertical transmission of culture in humankind.)
THE THEORY OF SOCIOBIOLOGY: ALL ORGANISMS STRIVE TO PERPETUATE THEIR OWN DNA
The concept of sociobiology was popularized during the 1970s. It states that an individual's genetic fitness is not solely measured by his survival and reproduction of offspring but also by enhancing the fitness of other relatives who share family genes. The theory was developed because prior models of evolution could not explain what appeared to be "altruistic" behaviors that benefited non-offspring genetic relatives but had no benefit for the altruist, or even put it at increased risk.8·9 For example, young birds often help parents care for younger siblings in the nest. And Florida scrub jays will help protect younger siblings from predators by calls of alarm or even by confronting the predator, but only for the nest of their parents, a parent and a stepparent, or a sibling. In primate society, older siblings often care for younger siblings while the more experienced but unrelated members of the group do not help. (This helps explain why human parenting with the transmission of culture often extends beyond our own children to biologically related children.)
Sociobiology helps explain why males of monogamous species will help care for their own young but make efforts to ensure that their mate is not carrying the progeny of another male. Mountain bluebirds are monogamous. Experiments have shown that when a male mountain bluebird was placed near a female while her mate was gathering food during the breeding season, the returning male attacked both his mate and the other male. In one case, the returning male drove his mate from the nest and took another mate, which is virtually unheard of among these birds.8
Is there evidence that sociobiology influences human family behavior? Daley and Wilson have summarized evidence for sociobiology of "evolutionary social psychology" in reference to family violence and homicide.10 As sociobiology would predict, family members who are not blood relatives are much more likely to kill each other than those who are genetically related. In 1972, 19% of all homicide victims in Detroit were married to the perpetrator whereas only 6% were blood relatives. Similarly, 10% of 1980 Miami homicide victims were married to the perpetrator whereas only 1.8% were blood relatives. And male sexual proprietariness or jealousy was the dominant issue in marital violence, consistent with the sociobiology model. An interview of North American spouse killers indicated that, in almost all cases, the husband's concern with his wife's fidelity or her intention to leave him initiated his violence regardless of whether it was the husband or wife who was ultimately killed by the other.
Daley and Wilson also studied child abuse, the most severe form of parenting failure, to test these theories. Although the great majority of step-parents provide good care for stepchildren, rates of child abuse are over-represented among stepparents compared with biological parents.10 For example, a study in Great Britain found almost equal numbers of children were killed by stepfathers and genetic fathers, even though fewer than 1% of all children lived with a stepfather at that time. A child living with one or more substitute parents in the United States in 1976 was almost 100 times more likely to be fatally abused than the same age child living with genetic parents. A 1974 to 1983 Canadian study found similar results. Thus the results were consistent with predictions of evolutionary social psychology or sociobiology. Richard Dawkins has popularized an extreme view.11 He contents that organisms serve their genes rather than the other way around. "We are machines for propagating DNA... it is every living object's sole reason for living." His "selfish gene" theory contends that human consciousness is just an adaptation that genes use to advance their own interests.
INSIGHT INTO PARENTING DERIVED FROM TWIN STUDIES
We are all the products of nature (genetic inheritance) and nurture (parenting and other environmental influences). If it were possible to dissect the effects of parenting from those of genetics and sources of nurture other than parenting, we would know much more about what parenting contributes to the development of children. The attempt to do this is "in progress" using studies of twins.12'14
The influence of genetics on the inheritance of behavior can be explored by comparing the incidence of a given behavior in identical twins who have been reared together and identical twins who have been reared apart. Although useful, this approach is limited because it is difficult to find identical twins reared apart from birth. However, insight into the genetic contribution can be obtained by comparing a behavior in identical and fraternal twins. Because fraternal twins share half of their genetic material while identical twins share all of this, the genetic influence on a behavior is estimated by doubling the difference in the correlation for a given "trait" among identical and fraternal twins. Nearly all behavioral disorders, personality characteristics, and measures of intelligence are to some degree genetically inherited according to data from both of these approaches. The genetic heritability of vocational interests, scholastic achievement, and personality, measured along several parameters, range from 40% to 50% in various twin studies.12'14 Although genetics appears to have a somewhat lower contribution towards intelligence during childhood, by adolescence, the dominant effect on intelligence seems to come from genetics. The genetic impact on intelligence rises to as high as 80% by adult life.13
By definition, whatever variance remains after genetics is subtracted is due to nurture or measurement error. The last can be determined by test -retest studies and is relatively small in larger studies. Twin studies divide nurture into shared environment and non-shared environment. The shared environment relates to the nurture that is shared by twins or other siblings and thus provides insight into the effects of parenting. Non-shared is everything else.
The shared environment component can be determined as illustrated in the following two hypothetical examples: If the correlation for a trait is 75% for identical twins living in their biological home and 50% for fraternal twins also living with biological parents, 50% of the variance can be ascribed to genetics (75% - 50% X 2); 25% is not shared and 25% is shared, excluding sampling error. In contrast, if identical twins correlated 50% and non-identical twins correlated 25% for a given behavioral trait, 50% of the variance can be explained by genetics, as before, but all of the remaining 50% is due to non-shared environment. This type of analysis leads to the unexpected observation that most environmental influence for intelligence, personality, and psychopathology is nonshared, whereas the shared environment, presumably created primarily by parenting, accounts for less than 10% of variance for these.
The observations that genetics accounts for about half and shared family environment accounts for little or none of the remainder holds for a broad array of twin studies and traits like extroversion, neuroticism, a sense of well-being, tolerance, and a "good impression."14 The conclusion is that environmental influences make children in the same family as different from one another as are children in different families.14 (Perhaps this has been selected in evolution to emphasize diversity for ideas and inventions in culture.) The source of the non-shared variance that overpowers shared influence is uncertain. Non-shared environment would include that created by the culture, outside of the family; differential treatment of twins or siblings by a given mother and father; and treatment of siblings by each other. Siblings interact in various supportive, competitive, and combative ways, and twin data suggest a significant part of nonshared environment may come from relationships with brothers and sisters.14
One would expect that differences in the personalities of children and parents, arising from genetic or non-shared environment, could lead to differential treatment by parents within a family. Although data that this happens are not convincing, the concept that we may create our own environment through our genetic constitutions is growing.'2,13 Someroff's model of transactional regulation would suggest that genetically acquired personality differences would eventually create unique non-shared environments, both within and outside of the family.
MEANWHILE, BACK AT THE OFFICE...
So what does this mean to the practice of pediatrics? To be honest, most of the above discussion is based on theory or assumptions rather than direct measurements, and much remains to be validated by scientific means and some is fairly controversial. Thus, making ironclad conclusions about parenting based on these is not recommended. However, it is possible to derive some suggestions for parenting guidance from these theories and observations. In fact, most of these are similar to the recommendations in the articles of this issue, although perhaps derived in a different way.
First, these theories suggest that parenting is about the education, protection, and socialization of children or training them to live within their culture. Although a parent might not be able to substantially change a child's inherent personality or intelligence, the pediatrician may help parents teach their children how to live in society given that each child's personality and intelligence may be different. During the first few years of life, learning how to live in society may be about getting along with one's peers, siblings, and parents in spite of constitutional uniqueness. Thus pediatricians may make important contributions when parents solicit advice about specific socialization or learning problems. Techniques like timeout and behavioral modification can be used as parental techniques for socializing children (see Drabman and Jarvie15 for a "How to" description). Also, pediatricians might not be able to help parents change the personalities or intelligence levels of their children (but we need more data).
Second, children within the same family are naturally very different. Variations in personality and intelligence are to be expected and may primarily come from genetics and non-shared environment rather than being entirely the "fault" of the parents.
Third, each pediatrician's philosophy and principles abut rearing children will be derived from his or her individual nature and nurture; so each of us is at risk for projecting our own biases, based on personal experience, on our families.
Fourth, parenting is complex and not well understood because it is shaped by biological and cultural evolution. This makes it difficult for the pediatrician to give scientifically validated guidance and may also be why it is relatively hard to change parenting styles by advice in the office, especially if the advice is unsolicited. It may also be why gentle subtle advice works better than criticizing parenting.
These theories also suggest that continued support from both the biological mother and biological father, especially when these two do not live together, can be advantageous to a child. The pediatrician may choose to explain how paternity testing is done to the mother who requests this because she is seeking such support, or to the father who wants this information because he is considering not supporting a child he worries may not be his.
We need much more scientific and outcomesbased information on how to guide parenting.
1. O'Ne'll L, Murhphy M, Gallagher RB. What are we? Where did we crane from? Where are we going! Science. 1994;263:181-183.
2. Cavail-Storza LL, Feldman MW, Chen KH, Dornbusch SM. Theory and observation cultural transmission. Science. 1982;2 18:19-27.
3. CloningerCR. The dynamics of social learning. Science. 198 1;21 3:858-859.
4. Simon H. A mechanism of social selection and successful altruism. Science. 1990:250:1665-1668.
5. Lovejoy CO. The origin of man. Science. 1981:211:341-349.
6. Lewin R. Living in the fast track makes for small brains. Science. 1988:242:513-514.
7. Levvin R, How did humans evolve big brains? Science. 197 5:2 16:840-84 ?d. Kolaia GB. Sociobiolo^y I!); Models of social behavior. Science. 1975:187:50-51.
9. Schalter GB. The evolution of sociality. Science. 190:2601-264.
10. Daily M, Wilson M. Evolutionary social psychology and familv homicide. Science. 1988:242:519-524.
11. Easierbraok G. Of genes and meanmglessness. Science. 1997;277:892.
12. Bouchard TJ. Genes, environment, and personality. Science. 1994:264:1700-1701.
13. Plomin R, Owen MJ, McGuffin P. The genetic basis of complex human behaviors. Science. 1994:264:1 733- 1739.
14. Plomin R, Daniels D. Why are children in the same family so different from one another? Behavioral and Brain Sciences. 1987:10:1-16.
15. Drabman RS, Jarvie G. Counseling parents of children with behavior problems: The use of extinction and time out techniques. Pediatrics. 1977:59:78-85.