In the most extreme scenario favoring genetics, if intelligence differences among people were due 100% to genetic factors inherited as a random mix of genes from a mother and a father, it would still be the case that some genes and their expression can be modified by environmental factors. The interaction of genetic expression and non-genetic variables is called epigenetics, discussed later in this section. In the 100% genetic scenario, it also would be the case that a person may well be liberated from a bad, suboptimal, or constraining environment by winning the genetic lottery for high intelligence (i.e., the random mix of genes from both parents including the salient ones for intelligence). The all (or mostly) gene scenario would also lead to a practical suggestion for maximizing your child’s intelligence: find the smartest mate you can (simple but perhaps not so easy). In the 100% genetic scenario, a person who loses the genetic lottery and has low intelligence would be constrained in some important aspects of life success even if they had the best supportive or enriched environment money could buy.
In the other extreme scenario, if differences in intelligence involved no genetic mechanisms, each person’s intelligence would be determined by the influences of their environment, especially during childhood when brain development is maximal and the child’s ability to choose favorable environments is minimal. The all (or mostly) environment theory easily leads to a Behaviorist or “Blank Slate” view that anyone could develop high intelligence, or any other psychological attribute, if only the right environmental ingredients were available (Watson, 1930). This view is quite popular despite the general demise of classical Behaviorism. Moreover, the “Blank Slate” view of human potential has limited empirical support for most aspects of behavior (Pinker, 2002), and virtually no support at all for intelligence, as we will see later in this chapter.
The popular middle position holds that both genes (nature) and environment (nurture) explain differences in intelligence. The older simple version of this position was that genetic and environmental factors both contributed about equally. We now recognize that genes and environments interact in that gene expression may be sensitive to environmental variables. This is the essence of epigenetics, which is the study of how environments influence the ways genes function. It’s hard to apportion intelligence variance to just genes when complex interactions with environments are part of the mix. Epigenetics is a relatively new field, but there are already some promising indications of progress. A longitudinal study of Romanian orphans, for example, has identified risk for cognitive and psychiatric problems partially attributable to the extreme social deprivation in early life experience. DNA analyses indicate specific genetic alterations are related to the degree of deprivation (Drury et al., 2012). Animal research suggests that some changes in gene expression related to environmental factors may actually be heritable (Champagne & Curley, 2009). This is exciting research, but so far there is no direct connection to human intelligence, although there is considerable emerging research on the epigenetics of memory (Heyward & Sweatt, 2015). Environmental variables like exposure to language in early childhood (Kuhl, 2000, 2004) influence brain neurobiology and development, but how these factors may relate to intelligence has not yet been demonstrated. How many epigenetic factors may contribute to intelligence differences is not yet known, but the concept reinforces the assumption that any salient environmental variables work through biological mechanisms, genetic or not. For now, the weight of the evidence emphasizes the influence of genes on intelligence, with or without known epigenetic influences.
2.1 The Evolving View of Genetics
It may surprise you to know that the definition of a gene is not what it was just a few years ago (Silverman, 2004). Prior to the technology-driven Human Genome Project, genetic researchers expected to find about 100,000 genes because genes code proteins, which are the basic building blocks of life. Humans have at least 100,000 proteins. Each gene was thought to code one protein. However, the Human Genome Project initially reported only about 25,000 protein-coding genes and that number has been revised downward to perhaps less than 20,000 (Ezkurdia et al., 2014). This means that each gene can express itself in many different ways, and the mechanisms for controlling gene expression are largely unknown. Gene expression is just a way of saying that genes turn on and off over the life span. This results in a constantly changing mix of proteins influencing all aspects of biology in complex and dynamic interactions. What exactly are the switches or triggers that turn genes on and off and how might the switches interact with environmental factors? How do the myriads of gene protein products interact with each other in multistep, cascading sequences? Such issues are the focus of the nascent field of epigenetic research.
Historically, most researchers have assumed that intelligence, no matter how it was defined, develops in childhood and is strongly influenced by environmental factors, especially home life and social culture. In this view, whatever role genes might play is minimized, and some even argue for a zero contribution of genes. Although this view about the importance of early environment seems reasonable, and even flattering to proud parents, the evidence for strong environmental effects on intelligence, especially in early childhood, is surprisingly weak, as we will see. Epigenetics provides a concept for the continued consideration of theories about the importance of environmental factors for intelligence, but epigenetic research on intelligence is just beginning (Haggarty et al., 2010). Nonetheless, like climate change, the data that support a major genetic component to intelligence are compelling and the number of genetic deniers and minimizers is diminishing rapidly.
Generally, we don’t like to think about any constraints on potential life achievement so the idea that intelligence has a major genetic component is not readily embraced as a good thing. This is so especially in some academic social science circles where there is vested interest in the study of cultures. In fact, there is a decades-old concerted effort to undercut, deny, and impugn any and all genetic studies of intelligence (Gottfredson, 2005). A similar effort in the 1960s and 1970s regarding the “myth” of schizophrenia and other psychiatric disorders has all but disappeared. Much of the anti-genetic feeling originally arose as a moral response to eugenic movements in the nineteenth and early twentieth centuries, the aftermath of Nazi atrocities during the 1930s and 1940s, and most recently for our context here, to one specific paper published in 1969 by an educational psychologist at the University of California, Berkeley named Arthur Jensen. We will discuss this infamous paper shortly.
Here is a crucial point to keep in mind as we introduce genetic studies: throughout this book, whenever we talk about the effects of any variable or factor on intelligence, we are actually referring to the effects on intelligence differences (variance) among people.
As the term implies, behavioral genetics generally refers to the study of behavioral traits and are of two basic kinds: quantitative, and molecular. The former, with roots in Mendel’s experiments with peas, deals with establishing whether a genetic component (a genotype) may exist or not for a behavior or characteristic (the phenotype) and if so, how much variance can be accounted for by genes. Quantitative genetics includes modeling a mode of gene transmission (e.g., dominant or recessive). Twin and adoption studies are primary methods of quantitative genetics and we will review key studies and some surprising findings that support a strong role for genes and a minimal role for environmental variables in explaining intelligence differences among individuals. Molecular genetics is a newer field and uses various DNA technologies and methods to identify genes that are related to variation in specific traits and, in the case of intelligence, how those genes work to influence brain development and brain function. This ambition is as complex as any goal in any scientific field. Molecular genetic findings related to intelligence so far are quite tentative with little replication of specific genes identified as possibly related to intelligence. Nonetheless, there is progress and the findings reviewed later in this chapter are somewhere between intriguing and amazing.
The early enthusiasm for molecular genetic techniques began about 20 years ago with the optimistic promise of imminent discovery of a few genes that accounted for considerable variance in intelligence. This has not happened and more than a few critics of the genetic view emote a bit of glee over the failure to identify specific intelligence genes so far. Early indications, however, suggested that the hunt for intelligence genes actually was a hunt for “generalist” genes, each of which influenced multiple cognitive abilities that underlie intelligence. Kovas and Plomin (2006) summarized this view simply: “genetic input into brain structure and function is general not specific” (p. 198). Two key concepts are: one gene can affect many dissimilar traits (pleiotropy), and many genes can affect one trait (polygenicity).
Although the concept of generalist genes is controversial, a broad consensus has emerged that intelligence is heritable and polygenetic. For example, one study based on 3,511 unrelated adults concluded that there are many intelligence genes that all together may account for 40%–50% of variance in general intelligence (Davies et al., 2011) although no one gene yet accounts for more than a tiny portion of variance. Additional research supports pleiotropy for diverse cognitive abilities (Trzaskowski et al., 2013a). Researchers investigating schizophrenia, autism, obesity, and many other gene-rich targets, even height, find similar polygenetic and pleiotropic results. At this stage, the heritability of human intelligence is well-established, and there are even emerging data in chimpanzees (Hopkins et al., 2014). There are interpretation issues regarding some aspects of the genetic data that are still unresolved (Nisbett et al., 2012; Shonkoff et al., 2000) and may remain so until specific genes for intelligence are identified and confirmed. Recent efforts to minimize the importance of genetic influences on intelligence in favor of environmental influences (Nisbett, 2009) do not stand scrutiny (Lee, 2010). Fortunately, there are even newer findings about intelligence that may signal real progress toward discovering specific genes and their effects. Before reviewing recent noteworthy studies in both quantitative and molecular genetics, let’s start with some history to put the genetic story of intelligence into context by reviewing a surprising failure and an alleged fraud.
2.2 Early Failures to Boost IQ
The failure hit the fan in 1969 without warning. In the early 1960s, President Lyndon Johnson committed the USA to a war on poverty. One aspect of this admirable federal effort was aimed at a major concern that had been observed for decades. Poor children, especially from minority groups, tended to score lower on cognitive tests, including IQ tests. At the time the consensus among most educators, psychologists, and policy makers was that any cognitive gaps revealed by tests, especially for intelligence, were due mostly or entirely to educational disadvantages and therefore could be eliminated if poor children got the same early education opportunities that middle- and upper-class families routinely provided. Such opportunities were virtually unavailable to the poor, especially prior to the 1954 Supreme Court decision striking down race-based separate but equal approaches to education. The solution for eliminating any cognitive gaps seemed obvious and the idea of compensatory education resulted in the federally funded Head Start Program. Prior to Head Start, several different compensatory education demonstration projects had been implemented on a limited basis. Some of these projects were reporting encouraging and even dramatic positive results at reducing cognitive gaps and increasing IQ scores. These efforts were the basis for the optimistic view that Head Start would be a great success at eliminating the gaps.
The Harvard Educational Review asked Arthur Jensen, a noted educational psychologist, to review the claims of these early compensatory efforts (Head Start had not yet been implemented long enough to be included in this review). Jensen’s article (1969) was entitled, “How much can we boost IQ and scholastic achievement?” Here is the opening sentence: “Compensatory education has been tried and it apparently has failed.” Jensen continued with over 100 pages of detailed analysis of intelligence research that revealed little if any lasting effect of the early compensatory efforts on either IQ scores or school achievement. That alone was bad enough in the political context of widespread enthusiasm for the nascent Head Start Program, but the article got worse when Jensen discussed genetics. He first reviewed studies of environmental effects on intelligence. He concluded that the empirical evidence for any major environmental effects on intelligence in general, and especially for the g-factor, was actually quite weak. He then argued that one reason for this would be that variance in intelligence, especially the g-factor, was mostly genetic. He summarized genetic studies that appeared to validate this view. In 1969, this conclusion was a bit of a stretch given the paucity of both environmental and genetic studies with large samples and solid research designs. However, the article, already offensive to the majority view that intelligence derived mostly from environment, went even further with a controversial suggestion, and controversial is an understatement. Because IQ scores appeared to be impervious to compensatory efforts and because genes played an important role, Jensen asserted the hypothesis that the average intelligence differences found for some racial groups compared to whites (he focused on black/white differences) might have a genetic component. And with the publication of that hypothesis, research on intelligence all but ended for more than a generation.
The negative response to Jensen’s review article was ferocious. The most vicious responses were directed to the inflammatory inference that blacks were intellectually inferior because of their genetic makeup and to the general idea that genes played a major role in intelligence and the environment did not. Jensen’s concluding paragraphs about the importance of adjusting teaching methods to match the learning capabilities of individual students to maximize school achievement for all children received virtually no attention (see Section 6.6). In any case, critics have spent decades attacking Jensen personally and his arguments. As mentioned briefly in the last chapter, another book published in 1973, IQ in the Meritocracy (Herrnstein, 1973) created a similar firestorm regarding the role of genetics in intelligence. Given the racial inferences and the hot emotional atmosphere, few researchers or their students opted to focus their careers on any questions at all about intelligence. Getting federal research support for researching intelligence became virtually impossible. Almost overnight, intelligence research became radioactive.
Head Start pushed ahead and similar compensatory research efforts included increasingly intensive interventions. In the 1970s and 1980s, Jensen’s critics attacked the validity of IQ tests and scores, the existence of the g-factor, quantitative genetics in principle, and even the integrity and motivation of individual researchers. One simple argument was that any average group differences in intelligence test scores were most likely due to test bias and had no meaning. The bias hypothesis, as noted in the previous chapter, has been studied extensively and has little empirical support. As far as test scores being without real meaning, there is extensive evidence, as noted in Chapter 1, that scores predict many aspects of life (Deary et al., 2010; Gottfredson, 1997b). Moreover, in the next two chapters, neuroimaging shows that intelligence test scores are correlated to a variety of structural and functional measures of the brain; findings that would be impossible if the test scores were meaningless. Some critics challenged whether the g-factor was merely a statistical artifact, a view not supported by many sophisticated psychometric studies (Jensen, 1998; Johnson et al., 2008b). Other critics went beyond debate about data and attacked Jensen and some behavior genetic researchers with ad hominem charges of explicit racism. Jensen was once asked directly if he was a racist. His answer was, “I’ve thought about this a lot and I have come to the conclusion that it’s irrelevant” (Arden, 2003, p. 549). I knew Jensen for many years and I understand his point was that his interpretation of data, even if it was motivated by unconscious racism, was testable and falsifiable by objective scientific methods. He was confident that future research could potentially refute any of his hypotheses. He wa
s, as far as most observers could perceive, unflappable in the face of personal attacks because he was completely driven by data. In my view, he would not have been disappointed at all if new data showed him to be wrong.
The point in summarizing this incendiary period in the history of intelligence research is to help explain the origin of the negative valence that intelligence research still carries to some extent today. The modern neuroscience studies that are the focus of this book have helped the field move beyond these old destructive controversies. While the basis of average group differences on psychometric tests of intelligence and other cognitive abilities is still unsettled, the major role of genetics for explaining intelligence differences among individuals is firmly established, as detailed in the next section. It is also the case that the weight of evidence from modern studies of intensive compensatory education, now rebranded as early childhood education, still fails to find lasting effects on IQ scores and even short-lived increases are not clearly related to the g-factor (te Nijenhuis et al., 2014). Contrary to Jensen’s review, newer, more intensive studies indicate that some important aspects of academic achievement, like graduation rates, apparently do improve (Barnett & Hustedt, 2005; Campbell et al., 2001; Ramey & Ramey, 2004). There also are some quantitatively sophisticated estimated projections that IQ scores for disadvantaged children potentially could increase dramatically given the right program components at early ages (Duncan & Sojourner, 2013), although no such gains have been realized let alone tested for durability. It is my view that there are many good reasons to support early childhood education that do not depend on whether IQ scores change or not due to genetics or other reasons. Including IQ in the discussion about early education probably doesn’t help make the case. More about the neuroscience potential for increasing intelligence will be detailed in Chapter 5.
The Neuroscience of Intelligence Page 7
