8.4

Genes, Environment, and the Development of Intelligence No issue in psychology has produced more acrimonious debate than the issue of how heredity and environment influence intelligence. Even people who recognize that intelligence, like all human qualities, is constructed through the continuous interaction of genes and environment often forget this fact and take extreme positions that are based more on wishes, emotions, and ideologies than on logic and evidence. Bronfenbrenner’s (1993) bioecological model of development provides a useful starting point for thinking about genetic and environmental influences on intelligence. This model, which is discussed in more depth in Chapter 9, envisions children’s lives as embedded within a series of increasingly encompassing environments. The child, with a unique set of qualities, including genetic endowment and personal experiences, is at the centre. Surrounding the child is the immediate environment, especially the people and institutions with which the child interacts directly: family, school, classmates, teachers, neighbours, and so on. Surrounding the immediate environment are more distant, and less tangible, forces that also influence development: cultural attitudes, the social and economic system, mass media, the government, and so on. We now examine how qualities of the child, the child’s immediate environment, and the broader society influence the development of intelligence. In the movieMy Fair Lady, Eliza Doolittle found it easier to don the clothing of an upper-class lady than to adopt the haughty reserve viewed as appropriate by that class at that time. This scene of opening day at the Ascot Racecourse, as well as the movie as a whole, makes the argument that differences that might be attributed to nature are actually the product of nurture. Qualities of the Child Children contribute greatly to their own intellectual development through their genetic endowment, the reactions they elicit from other people, and their choice of environments. Genetic Contributions to Intelligence As noted in Chapter 3, the genome substantially influences intelligence. This genetic influence varies with age (Figure 8.5): it is moderate in early childhood and becomes large by adolescence and adulthood (Plomin & Deary, 2015). Reflecting the same trend, IQ scores of adopted children and their biological parents become increasingly correlated as the children develop, even without contact between them, but the scores of adopted children and their adoptive parents become less correlated over the course of development (Plomin et al., 1997). FIGURE 8.5 Changes with age in factors influencing intelligence As children grow into adults, the influence of genetics on individual differences in intelligence increases, whereas the influence of shared aspects of family environment decreases. (Data from McGue et al., 1993) One reason for this increasing genetic influence is that some genetic processes do not exert their effects until late childhood, adolescence, or adulthood. For example, some types of synchronization of activities of distant brain areas are not evident until adolescence or early adulthood, and the extent of such synchronization, which influences a variety of cognitive activities, reflects genetic influences (Uhlhaas et al., 2010). Another reason is that children’s increasing independence with age allows them greater freedom to choose environments that are compatible with their own genetically based preferences but not necessarily with those of the parents who are raising them (McAdams & Olson, 2010). Advances in genetics have inspired research aimed at identifying genes that explain individual differences in intelligence. These efforts have led to identification of a large number of genes that are associated with intellectual disability (Inlow & Restifo, 2004) and also to identification of a large number of other genes that are consistently related to normal variation in intelligence (Trzaskowski et al., 2014). However, all known correlations between individual alleles of genes and IQ are very small. These findings suggest that genetic influences on intelligence reflect small contributions from many different genes, as well as complex interactions amongst them, rather than reflecting one or a small number of “intelligence genes” (Chabris et al., 2015; Mukherjee, 2016; Nisbett et al., 2012). Simply put, there is no “intelligence gene” or even a small group of “intelligence genes.” Genotype–Environment Relations As noted in Chapter 3, the environments children encounter are influenced by the children’s and their parents’ genotypes. Sandra Scarr (1992) proposed that gene–environment relations involve three types of processes: passive, evocative, and active. Passive effects of the genotype arise when children are raised by their biological parents. These effects occur not because of anything the children do but because of the overlap between their parents’ genes and their own. Thus, children whose genotypes predispose them to enjoy reading are likely to grow up in homes with plentiful access to reading matter because their parents also like to read. The passive effects of the genotype help explain why some correlations between biological parents’ and their children’s IQ scores are higher when the children live with their biological parents than when they live with adoptive parents. Evocative effects of the genotype emerge through children influencing other people’s behaviour. For example, parents will read more stories to a child who shows interest in the stories than to one who is uninterested. Active effects of the genotype involve children’s choosing environments that they enjoy. High school students who love reading will read a great deal, regardless of whether they were read to when young. The evocative and active effects of the genotype help explain how children’s IQ scores become more closely related over time to those of their biological parents, even if the children are adopted and never see their biological parents. Influence of the Immediate Environment The influence of nurture on the development of intelligence begins with a child’s immediate environment of family and schools. Children influence their own development: these children’s positive reactions to their father’s reading ensure that he will want to read to them in the future. Family Influences If asked to identify the most important environmental influence on their intelligence, most people probably would say, “My family.” Testing the influence of the family environment on children’s intelligence, however, requires some means of assessing that environment. How can something as complex and multifaceted as a family environment be measured, especially when it differs for different children in the same family (e.g., if parents favour one child over another, the family environment can be very different for the two children)? Bradley and Caldwell (1979) tackled this problem by devising a measure known as the Home Observation for Measurement of the Environment (HOME). The HOME samples various aspects of children’s homelife, including organization and safety of living space; intellectual stimulation offered by parents; whether children have books of their own; amount of parent–child interaction; parents’ emotional support of the child; and so on. Table 8.1 shows the items and subscales used in the original HOME, which was designed to assess the family environments of children from birth to age 3 years. Subsequent versions of the HOME have been developed for application with preschoolers, school-age children, and adolescents (Totsika & Sylva, 2004). TABLE 8.1 Sample Items and Subscales on the HOME (Infant Version) Emotional and Verbal Responsivity of Mother Mother spontaneously vocalizes to child at least twice during visit (excluding scolding). Mother responds to child’s vocalizations with a verbal response. Mother tells child the name of some object during visit or says name of person or object in a “teaching” style. Avoidance of Restriction and Punishment Mother does not shout at child during visit. Mother does not express overt annoyance with or hostility towards child. Mother does not interfere with child’s actions or restrict child’s movements more than three times during visit. Organization of Physical and Temporal Environment When mother is away, care is provided by one of three regular substitutes. Child is taken regularly to doctor’s office or clinic. Child has a special place in which to keep toys and “treasures.” Provision of Appropriate Play Materials Child has push or pull toy. Child has stroller or walker, kiddie car, scooter, or tricycle. Child has learning equipment appropriate to age—cuddly toy or role-playing toys. Maternal Involvement with Child Mother tends to keep child within visual range and to look at the child often. Mother “talks” to child while doing her work. Mother structures child’s play periods. Opportunities for Variety of Daily Stimulation Mother reads stories at least three times weekly. Child eats at least one meal per day with mother and father. Child has three or more books of their own. Information from Bradley and Caldwell (1984). At all ages, children’s IQ scores are positively correlated with scores on the HOME. A large multisite study, which included a group of children from Hamilton, Ontario, found that HOME scores of 12-month-olds correlated positively with the IQ scores of the same children at 3 years of age (Bradley et al., 1989). Once children go to school, scores on this measure of family environment also correlate positively with their reading- and math-achievement test scores (Bradley et al., 2001). The relation is usually stable over time; HOME scores of 2-year-olds correlate positively with IQ scores and school achievement of the same children at age 11 years (Olson, Bates, & Kaskie, 1992). When HOME scores are relatively stable over time, IQ scores also tend to be stable; when HOME scores change, IQ scores also tend to change in the same direction (Totsika & Sylva, 2004). In addition, an abbreviated and adapted form of the HOME has proven useful for assessing home environments in countries very different than the United States and Canada, including Bangladesh, Brazil, India, Nepal, Pakistan, Peru, South Africa, and Tanzania (Jones et al., 2017). Thus, assessing varied aspects of a child’s family environment allows prediction of the child’s IQ. Stimulating home environments, especially those in which adults and children undertake challenging tasks together, are associated with high IQ scores and high achievement in school. Given this evidence, it is tempting to conclude that better-quality home environments cause children to have higher IQ scores. Whether that is actually the case, however, is unknown. The uncertainty reflects two factors. First, the intellectual environment that parents establish in the home is almost certainly influenced by their genetic makeup. Second, almost all studies using the HOME have focused on families in which children live with their biological parents. These two considerations imply that parents’ genes may influence both the intellectual quality of the home environment and children’s IQ scores; thus, the home intellectual environment as such may not cause children to have higher or lower IQ. Consistent with this possibility, in the few studies in which the HOME has been used to study adoptive families, the correlations between it and children’s IQ scores are lower than in studies of children living with their biological parents (Plomin et al., 1997). Thus, although scores on the HOME correlate with children’s IQ scores, whether causal relations exist between the two remains uncertain. Shared and non-shared family environments The phrase “family intellectual environment” is often taken to mean characteristics that are the same for all children within the family: how much the parents value education, the number of books in the house, the frequency of intellectual discussions around the dinner table, and so on. As discussed in Chapter 3, however, each child within a given family also encounters unique, non-shared environments. In any family, only one child can be the firstborn and receive the intense, undivided parental attention early in life that being born first tends to bring (for better and worse). Similarly, a child whose interests or personality characteristics mirror those of one or both parents may receive more positive attention than other children in the family. If homes that are extremely lacking in intellectual stimulation are excluded from consideration, such within-family variations in children’s environment may have a greater impact on the development of intelligence than do between-family variations (Petrill et al., 2004). In addition, the influence of the non-shared environment increases with age, and the influence of the shared environment decreases with age, as children become increasingly able to choose their own friends and activities (Plomin & Daniels, 2011; Segal et al., 2007). With age, children increasingly shape their own environments in ways that reflect their personalities and tastes. The relative influence of shared environments and genetics varies with family income. Amongst children and adolescents from low-income families in the United States, the shared environment accounts for more of the variance in IQ scores and academic achievement than genetics does. In contrast, amongst children and adolescents from middle- and high-income families in the United States, the relative influence of shared environment and genetics is reversed (Nisbett et al., 2012; Turkheimer et al., 2003). These differing patterns are found as early as age 2 years (Tucker-Drob et al., 2011). Interestingly, the differing relations between shared environments and genetics amongst high- and low-income families have not been found in Australia, Germany, Great Britain, the Netherlands, or Sweden (Tucker-Drob & Bates, 2016). This may be because those countries ensure access to quality education regardless of family income, thus making the children’s intellectual development less dependent on the family in which they grow up. Influences of Schooling Attending school makes children smarter. Evidence for this conclusion came from a study that examined IQ scores of older and younger Israeli children within grades 4, 5, and 6 (Cahan & Cohen, 1989). As indicated by the gradual upwards trends in the graphs in Figure 8.6, older children within each grade did somewhat better than younger children within that grade on each part of the test. However, the jumps in the graphs between grades indicate that children who were only slightly older, but who had a year more schooling, did much better than the slightly younger children in the grade below them. For example, on the verbal-oddities subtest (which involves indicating which word in a set does not belong with the others), the results show a small gap between 123- and 124-month-old 4th-graders but a large gap between both of them and 125-month-old 5th-graders. A meta-analysis of studies on the relation between IQ and total number of years of formal education indicated that an extra year attending school increases IQ scores by 1 to 5 points (Ritchie & Tucker-Drobb, 2018). FIGURE 8.6 Relations of age and grade to performance on two parts of an IQ test The jumps between grade levels indicate that schooling exerts an effect on intelligence test performance beyond that of the child’s age. (Data from Cahan & Cohen, 1989) Another type of evidence indicating that going to school makes children smarter is that average IQ and achievement test scores rise during the school year but not during summer vacation (Huttenlocher, Levine, & Vevea, 1998). The details of the pattern are especially telling. Children from families of low SES and those from families of high SES make comparable gains in school achievement during the school year. However, over the summer, the achievement test scores of low-SES children tend to stay constant or drop, whereas the scores of high-SES children tend to rise (Alexander, Entwisle, & Olson, 2007). The likely explanation is that during the academic year, schools provide children of all backgrounds with relatively stimulating intellectual environments, but when school is not in session, fewer children from low-SES families have experiences that increase their academic achievement. Influence of Society Intellectual development is influenced not only by characteristics of children and their local environments but also by broader characteristics of the societies within which they develop. One reflection of societal influences is that in many countries throughout the world, average IQ scores consistently increased over the 80-year period from 1930 to 2010. This phenomenon has been labelled the Flynn effect, in honour of James Flynn, the researcher who discovered the trend (Flynn, 1987, 2009). For example, in the United States, the gains have been roughly 10 points (Flynn & Weiss, 2007). Given that the gene pool has not changed appreciably over this period, the increase in IQ scores must be due to changes in society. The sources of the Flynn effect remain controversial. Some researchers argue that the key factors are improvements in the lives of low-income families, such as improved nutrition (Lynn, 2009) and health (Eppig, Fincher, & Thornhill, 2010). These researchers point to evidence that in some countries, the increase in IQ scores has been greatest amongst those in the lower part of the IQ score and income distributions, who previously had received poor nutrition and health care. For example, amongst Danes born from 1942 to 1980, there was no change in the scores of people in the top 10% of the IQ distribution, but there was a large change amongst those in the bottom 10% (Geary, 2005). IQ score changes in some other countries, including Norway and Spain, show a similar pattern. However, gains in other countries, including Britain, France, and the United States, have been comparable throughout the range of IQ scores and family incomes (Nisbett et al., 2012). An alternative explanation for the rise in IQ scores is increased societal emphasis on abstract problem solving and reasoning (Flynn, 2009). Supporting this interpretation, scores on tests of fluid intelligence, which reflect abstract problem solving and reasoning, have increased roughly twice as much as scores on tests of crystallized intelligence, which measure previously acquired knowledge (Nisbett et al., 2012; Pietschnig & Voracek, 2015). Video: Education in Middle Childhood Relatively new technologies, such as video games, may have contributed to these gains in fluid intelligence. Playing video games has been found to improve a variety of cognitive processes: executive functioning (e.g., Flynn & Richert, 2018), selective attention (e.g., Glass, Maddox, & Love, 2013), problem solving (Blumberg & Randall, 2013), and even reading (Franceschini et al., 2017) and math (Deater-Deckard et al., 2014). Some studies have not found such positive effects of playing video games (e.g., Powers et al., 2013; Unsworth et al., 2015), but the literature as a whole seems to support the hypothesis that playing video games contributes to increases in fluid intelligence (Blumberg et al., 2019). Recently, however, findings on the Flynn effect took an unexpected turn. A study of almost 400,000 U.S. adults showed a reverse Flynn effect, with IQ scores declining from 2006 to 2018; the decreases were consistent across age groups, education levels, and gender (Dworak, Revelle, & Condon, 2023). Similar reverse Flynn effects have been found in recent years in Finland, France, and Norway (Bratsberg & Rogeberg, 2018; Dutton, van der Linden, & Lynn, 2016). Even with these recent decreases, current IQ test scores in Finland, France, and the United States remain considerably higher than those from 1930 to 1990; however, the reasons for both the prolonged increases and the recent decreases remain the subject of vigourous debate. Although many conclusions about intelligence remain controversial, one conclusion that sparks little controversy is that poverty hinders intellectual development. In the following sections, we consider how poverty affects children’s development in different societies and how it contributes to differences in IQ scores and school achievement. We will also consider risk factors associated with poverty that adversely affect intelligence, as well as programs that can enhance the intellectual development of children from families with low incomes.Effects of Poverty The negative effects of poverty on children’s IQ scores are indisputable. Even after taking into account the mother’s education, whether both parents live with the child, and a variety of characteristics of the child, the adequacy of family income for meeting family needs is related to children’s IQ scores (Duncan & Murnane, 2014). Further, the more years children spend in poverty, the lower their scores tend to be (Korenman, Miller, & Sjaastad, 1995). Poverty can exert negative effects on intellectual development in numerous ways. Chronic inadequate diet early in life can disrupt brain development; missing meals on a given day (e.g., achievement test day) can impair intellectual functioning on that day; reduced access to health services can result in more absences from school; conflicts between adults in the household can produce emotional turmoil that interferes with learning; insufficient intellectual stimulation can lead to a lack of background knowledge needed to understand new material; and so on. Poverty does not guarantee that these adverse influences will affect any individual child—some children from impoverished families receive adequate diets and health services, do not encounter unusual amounts of emotional turmoil, receive considerable intellectual stimulation, and so on—but poverty increases the likelihood of adverse influences. Consistent with this analysis, in all countries that have been studied, children from higher-income homes tend to score higher on IQ and achievement tests than do children from lower-income homes (Ganzach et al., 2013). Large mean differences between children from less and more affluent backgrounds are already present in measures of reading and math knowledge when children enter kindergarten (see Figure 8.7; Larson et al., 2015), and the differences grow larger over the course of schooling (Goodman, Gregg, & Washbrook, 2011). FIGURE 8.7 Reading and math scores upon entering kindergarten, by SES Even at the beginning of kindergarten, children from families with lower incomes and education lag far behind peers from more affluent and educated backgrounds. (Data from Larson et al., 2015) The larger the gap in incomes amongst families in a country, the larger the difference in intellectual achievement of children from high-income and low-income homes. Thus, the gap in intellectual achievement in the United States and other countries with large income differences is larger than the gap in intellectual achievement amongst children in countries in which the income gap is smaller, such as the Scandinavian countries and, to an extent, Canada, Germany, and Great Britain. As shown in Figure 8.8, children from affluent U.S. families score, on average, about the same on mathematics-achievement tests as children from affluent families in some comparison countries with more equal incomes, including Canada and Japan. In contrast, children from low-income U.S. families score far below children from low-income families in those same comparison countries. FIGURE 8.8 Relation in three countries between fathers’ occupational status and children’s math achievement As research by Robbie Case at the University of Toronto has demonstrated, Canadian and Japanese children whose fathers hold low-status jobs perform, on average, far better on math-achievement tests than do children whose fathers hold comparable jobs in the United States. In contrast, Canadian children whose fathers have high-status jobs perform, on average, as well as children whose fathers hold comparable jobs in the United States and almost as well as children from similar backgrounds in Japan. (Data from Case, Griffin, & Kelly, 1999) The key difference is that low-income U.S. families are much poorer, relative to others in their society, than their counterparts in many other high-income countries. Thus, as shown in Figure 8.9, in 2021, 15.1% of U.S. children lived in families with less than 50% of the median family income; as can be seen in this figure, the percentage of families this poor relative to others in their country was far lower in a number of other countries (Statista, 2022). FIGURE 8.9 Child poverty rates by country Canada has a moderate rate of child poverty, relative to most other high-income countries. (Data from Statista, 2022) Risk Factors and Intellectual Development In the popular media, reports on how to help all children reach their intellectual potential often focus on a single factor, such as the need to eliminate poverty or the need to eliminate racism or the need to preserve two-parent families or the need for high-quality day care or the need for universal preschool education. However, no single factor, nor even any small group of factors, is the key. Instead, many factors contribute to weak intellectual development. To capture the impact of these multiple influences, Arnold Sameroff and his colleagues developed an environmental risk scale (Sameroff et al., 1993) based on a number of features of the environment that put children at risk for low IQ scores (e.g., head of household being unemployed or working in a low-status occupation, mother who did not complete high school, no father or stepfather in the home, a large number of recent stressful life events, maternal mental health problems, and negative mother–child interactions). Each child’s risk score is a simple count of the number of major risks facing the child. Sameroff and his colleagues measured the IQ scores and environmental risks of more than 100 children when they were 4-year-olds and again when they were 13-year-olds. They found that the more risks in a child’s environment, the lower the child’s IQ score tended to be. As shown in Figure 8.10, the effect was large. The average IQ score of children whose environments did not include any of the risk factors was around 115; the average score of children whose environments included six or more risks was around 85. The sheer number of risks in the child’s environment was a better predictor of the child’s IQ score than the presence of any particular risk. Subsequent studies demonstrated similarly strong relations between number of risk factors and school grades (Gutman, Sameroff, & Cole, 2003). FIGURE 8.10 Risk factors and IQ scoreFor both younger and older children, the more risk factors in the environment, the lower the average IQ score. (Data from Sameroff et al., 1993) The Sameroff (1993) study also provided an interesting perspective on why children’s IQ scores are highly stable. It is not just that children’s genes remain constant; over time, their environment tends to remain fairly constant as well. The study revealed that there was just as much stability in the number of risk factors in children’s environments at ages 4 and 13 years as in their IQ scores over that period. Although Sameroff and his colleagues described their measure as a “risk index,” it is as much a measure of the quality of a child’s environment as of its potential for harm. High IQ scores are associated with favourable environments as much as low scores are associated with adverse ones. This is true for children from low-income families as well as for children in general. Low-income parents who are responsive to their children and provide them with safe play areas and varied learning materials have children with higher IQ scores (Bradley et al., 1994). Thus, high-quality parenting can help offset the risks imposed by poverty. Programs for Helping Children from Low-Income Families Beginning in the early 1960s, it became clear that children from economically disadvantaged families were at risk for negative outcomes. Child development research contributed to this consensus by demonstrating that children’s environments significantly affected their cognitive growth (Dennis & Najarian, 1957; Hunt, 1961). As a consequence, over the next decade, many intervention programs were initiated to enhance the intellectual development of preschoolers from families with low incomes. In a comprehensive analysis of 11 of the most prominent early-intervention programs—all of which focused on 2- to 5-year-old African American children from low-income families—Lazar and colleagues (1982) found a consistent pattern. Participation in the programs, most of which lasted a year or two, increased children’s IQ scores substantially when the test was given at the end of the program. However, over the next 2 or 3 years, the gains decreased, and by the fourth year after the end of the programs, no differences were apparent between the IQ scores of participants and those of nonparticipants from the same neighbourhoods and backgrounds. Similar effects have been found subsequently for math and reading achievement—initial gains fade all too quickly (Bailey, 2019; Bailey et al., 2020). One reason is that children in the programs forget what they learned earlier; a bigger reason is that children who were not in the programs catch up by learning the skills and concepts that children in the programs learned earlier (Kang et al., 2019). Fortunately, other effects of these experimental programs aimed at helping preschoolers from lower-income backgrounds are more enduring. Program participants are more likely to later graduate from high school than preschoolers from low-income backgrounds who do not participate, and they are less likely to be assigned to special-education classes, held back in school, or be arrested by age 18 (Reynolds, Temple, & Ou, 2010). This combination of findings may seem puzzling. If the intervention programs did not result in lasting increases in IQ or achievement test scores, why would they have led to fewer children being assigned to special-education classes or being held back in school? A likely reason is that the interventions had long-term effects on children’s motivation and conduct. These effects would help children do well enough in the classroom to be promoted with their classmates, which in turn might make them less likely to leave high school early and less likely to turn to criminal activity. In Box 8.1, we discuss one prevention program in Canada—Better Beginnings, Better Futures (BBBF)—and one in the United States—the Carolina Abecedarian Project—that have shown the possibility of producing enduring gains in both IQ scores and school achievement. BOX 8.1 APPLICATIONS Highly Successful Early Interventions: Better Beginnings, Better Futures (BBBF) and the Carolina Abecedarian Project The difficulty of producing enduring gains in IQ and achievement test scores of children from lower-income families led some early evaluators to conclude that intelligence is unalterable (Jensen, 1973; Westinghouse Learning Corporation, 1969). However, the same findings motivated other researchers to find out if interventions that started in infancy, continued for a number of years, and attempted to improve many aspects of children’s lives might produce enduring increases in IQ. Two such projects, the Better Beginnings, Better Futures (BBBF) project and the Carolina Abecedarian Project, clearly demonstrate that enduring changes can be achieved. BBBF began in 1991 as a 25-year longitudinal prevention project, in response to the results of the 1983 Ontario Child Health Study. This study indicated that 1 in 6 children had an emotional or behavioural problem and that children from socioeconomically disadvantaged homes were at higher risk for these problems (Peters et al., 2010). BBBF was based on a universal prevention approach. That is, rather than identifying particular children and families who were at risk for adverse outcomes, the project focused on communities in Ontario that were defined as “at risk.” These communities were considered economically disadvantaged: they had high numbers of families living below the low-income line, high levels of unemployment, high numbers of children on social assistance, and high numbers of children living in subsidized housing. In addition, factors existed that were considered indicators for poor child development: infant mortality, low birth weight, lone-parent families, adolescent pregnancy, low maternal educational achievement, high rates of leaving high school early, and immigration (see Peters et al., 2010, for further details). The three Ontario sites chosen for participation in BBBF were Cornwall, Highfield (in the greater Toronto area), and Sudbury. Following selection of the BBBF sites, two comparison sites with similar characteristics were selected: Ottawa–Vanier and Etobicoke (in Toronto). All children between the ages of 4 and 8 who were living in one of the selected neighbourhoods could participate in the program, as could the children’s families. Drawing on the ecological model of development advanced by Bronfenbrenner (detailed in Chapter 9), the project targeted child development in a variety of areas, including physical, social, and cognitive development, as well as mental health. Furthermore, in keeping with the ecological model, families, schools, and neighbourhoods were targeted for change. The specific programming delivered at each site varied; it was developed within the specific communities, with extensive consultation from community members. The accompanying table provides a list of programs that were offered at all three sites for a total of 4 years. Better Beginnings, Better Futures Programs Common to the Three Sites Program Type Program Description Child-focused In-class and in-school programs Childcare enhancements Before- and after-school activities School “breakfast club” Recreation programs Parent-focused Home visitors Parent support groups Parenting workshops One-on-one support Childcare for parent relief Family- and community-focused Community leadership development Special community events and celebrations Safety initiatives in neighbourhoods Community field trips Adult education Family camps Outreach to families Information from Peters and colleagues (2010). The impact of programming was measured at grades 3, 6, and 9. For the purposes of this chapter, we will focus on school functioning and academic achievement, which were assessed at grades 6 and 9. At grade 6, children from the BBBF sites had higher math achievement and fewer school suspensions than did those from the comparison sites. At grade 9, children in the BBBF programs required fewer special-education services, were less likely to repeat grades, and were rated by teachers as being better prepared to learn and having greater potential to go further in school (Peters et al., 2010). A follow-up of children at grade 12 indicated that the positive effects continued, with children from BBBF sites requiring fewer special-education services (Peters et al., 2010). As illustrated in these graphs, at grade 9, children who had taken part in BBBF were less likely to use special-education services (left) and to display hyperactive/inattentive behaviours in class (right) than were children from comparably disadvantaged backgrounds from the comparison sites. The benefits of participating in BBBF remained evident 6 years after the end of the program. (Data from Roche, Petrunka, & Peters, 2008) Not only did the BBBF program yield long-term benefits to the children enrolled, this program yielded economic benefits as well. A recent analysis showed that those children who were enrolled in BBBF programs used fewer government programs from ages 4 to 18, leading to a savings of $2.50 for every $1.00 invested by the government (Peters et al., 2016). A similarly successful early intervention program in the United States was the Carolina Abecedarian Project, which operated at one North Carolina site between 1972 and 1985. In the Abecedarian (pronounced “a-bee-cee-dar-ee-un”) program, children who were at high risk for poor outcomes due to low family income, the absence of a father in the home, low maternal IQ score and education, and other factors began attending a special day-care centre by the time they were 6-month-olds and continued to do so through the age of 5 years. Children aged 3 years and younger underwent a program that emphasized general social, cognitive, and motor development; for 4- and 5-year-olds, the program also provided systematic instruction in math, science, reading, and music. At all ages, the program emphasized language development and ensured extensive verbal communication between teachers and children. Program personnel also worked with the children’s mothers outside the day-care centre to improve their understanding of child development. Families of children in the experimental program were provided with nutritional supplements and access to high-quality health care. Families of children in a comparison group received similar nutritional and health benefits, but the children did not attend the day-care centre. This well-planned, multifaceted program proved to have lasting positive effects on the IQ scores and achievement levels of children in the experimental group. At the age of 21 years, 15 years after the program had ended, these children had mean IQ scores 5 points higher than the children in the control group: 90 versus 85 (Campbell et al., 2001). Participants’ achievement test scores in math and reading were also higher. As with less encompassing intervention programs, fewer participants were ever held back in school or placed in special-education classes. At age 30, a higher percentage of children in the experimental group than in the control group had graduated from university: 23% versus 6% (Campbell et al., 2012). A replication of the program demonstrated that the lower the mother’s educational level, the greater the difference the program made (Ramey & Ramey, 2004). What lessons can be drawn from these early-intervention projects? One important lesson is the benefit of starting interventions early and continuing them for substantial periods. A version of the Abecedarian program that ended at age 3 did not produce long-term effects on intelligence, nor did a program that provided educational support from kindergarten through grade 2 (Burchinal et al., 1997; Ramey et al., 2000). A more recent study demonstrated that 2 years, as opposed to just 1 year, of enhanced preschool programming led to greater improvements in the early literacy and numeracy skills of economically disadvantaged children (Domitrovich et al., 2010). A second crucial lesson is that the gains produced by successful early-intervention programs are likely due at least as much to improvements in children’s self-control and perseverance as to changes in children’s IQ scores (Heckman, 2011; Knudsen et al., 2006). Probably the most important lesson is the most basic: it is possible to design interventions that have substantial, lasting, positive effects on children’s intellectual development, regardless of income level. Project Head Start In response to the same political consensus of the 1960s that led to small-scale early-intervention programs, the U.S. government initiated a large-scale intervention program: Project Head Start. In the past 50 years, this program has provided a wide range of services to more than 37,000,000 children (National Head Start Association, 2023). At present, Head Start serves about 800,000 preschoolers per year, most of them 3- or 4-year-olds. The population served is racially and ethnically diverse. Of the families served in 2020–2021, 37% self-identified as Hispanic or Latino origin, 27% as Black or African American, and 24% as White non-Hispanic non-Latino origin; roughly one-third of families reported that a language other than English was the primary language spoken at home (Head Start/ECLKC, 2022). Almost all children in Head Start are from families with incomes below the poverty line; most are lone-parent families. In the program, children are provided with medical and dental care, nutritious meals, and a safe environment. Many parents of participating children work as caregivers at the Head Start centres, serve on policy councils that help plan each centre’s directions, and receive help with their own vocational and emotional needs. Consistent with the findings of the smaller experimental intervention programs that have been aimed at 3- and 4-year-olds, participation in Head Start produces higher IQ and achievement test scores at the end of the program and briefly thereafter. The strongest evidence for this conclusion comes from the Head Start Impact Study (U.S. Department of Health and Human Services [DHHS], 2010), an especially well-done study that included 5000 3- and 4-year-olds from low-income families who were on waiting lists to participate in a Head Start program. Half the children were randomly assigned to participate in Head Start; the other half followed another path of their parents’ choosing. The children comprised a nationally representative sample of the low-income population, and the Head Start centres in which the children enrolled were representative in terms of their quality. The results were very similar to those of the experimental preschool programs described in the previous section. Children who participated in Head Start showed better pre-reading and pre-writing skills at the end of a year in the program (U.S. DHHS, 2006), but their intellectual outcomes did not differ from those of nonparticipants by the end of grade 1 (U.S. DHHS, 2010) or grade 3 (Puma et al., 2012). On the other hand, participation in Head Start produced a number of positive effects that did endure: greater likelihood of graduating from high school and enrolling in university, better health and social skills, and lower frequency of later being held back in school, using drugs, and delinquency (Schanzenbach & Bauer, 2016). These important gains have contributed to the enduring popularity of Head Start. In Canada, the federal government launched a version of the Head Start program called the Aboriginal Head Start Program in Urban and Northern Communities (AHSUNC), now named First Nations Head Start. Initially, AHSUNC was provided to off-reserve Indigenous families in urban communities and the northern region, but it was expanded in 1998 to on-reserve communities (Health Canada, 1998). The program provides funding for early intervention for children under the age of 6 and their families, with programs designed and delivered by First Nations communities. AHSUNC focuses on six aspects: education, health promotion, culture and language, nutrition, social support, and parental/family involvement (Nguyen, 2011). The program reaches between 4600 and 4800 children annually (Public Health Agency of Canada, 2017). An evaluation study showed that AHSUNC has effectively increased First Nations children’s school readiness, including better language, motor, and academic skills. Furthermore, children and families report that exposure to Indigenous culture and language programming offered by sites yields long-term benefits (Public Health Agency of Canada, 2012b, 2017)