ENDURING ISSUES OF INTELLIGENCE

ENDURING ISSUES
Two enduring issues that have bedevilled research in intelligence are the genetics of IQ, and the relationship between race, genes and intelligence.

THE GENETICS OF IQ
Before considering whether intelligence ‘runs in families’ and, more specifically, how we can tell whether there is a genetic contribution to differences in IQ, it might be helpful to look at a few basic terms and methods from quantitative behavioural genetics – the discipline that aims to answer these questions. Our genotype is the genetic complement, coded in DNA, that we inherit from our parents. (genotype our genetic complement, coded in DNA, that we inherit from our parents) No two people have identical genotypes except identical twins. (phenotype the expression of our genes in behavioural traits that we can measure) The expression of those genes in behavioural traits that we can measure is called our phenotype. Phenotypes can vary because of genotypic differences and/or because the environment affects how our genes are expressed. IQ test scores are phenotypic measures, and intelligence is one of the most frequently researched traits in behavioural genetics simply because IQ represents one of the most re liable and important psychological measures. Genetic contributions to IQ differences can be estimated by comparing the similarity of IQ in individuals of different degrees of genetic relatedness while also assessing environmental similarities and differences. Heritability is a statistic that represents the proportion of phenotypic variance that is due to genetic differences – that is, the extent to which differences in measured intelligence are due to genetic differences. The maximum possible heritability is 1.0 (100 per cent of the difference is inherited) and the minimum is 0 (none of the difference is due to genetic differences).

The influence of environment
The influence of the environment on phenotypes comes in two main forms. There are differences between families (levels of income, parental rearing style, number of books in the home, etc.) which make children raised in a particular home more similar to each other than to children reared in a different home. This source of differences is often called the effect of the shared environment. The second kind of environmental influence is differences within the same family (in birth-order, children’s friends, school teachers, etc.). These effects make children in the same family different from each other and are referred to as non-shared environment effects. We can measure the influence of the common, or shared, environment by comparing individuals who are reared together or apart. The extent to which pairs of individuals are more similar when they are brought up in the same home is a measure of the importance of the common or shared environment. For example, if the home environment makes a difference it should incr ease the similarity of, for example, identical twins when they are reared together compared with when they are reared apart (i.e. when they are adopted into different homes). Similarly, the extent to which siblings who are reared together in the same home but who are genetically unrelated (because one or both is adopted) are similar to each other gives an estimate of the influence of shared environment. The effect of non-shared environmental variance can be detected in a number of ways. The most obvious is to measure the extent to which identical twins reared together (i.e. with both genetic and shared environmental variance in common) are different from each other due to the non-shared environmental influences they may experience when growing up (e.g. at school, or from peers). Do we inherit our IQ? Studies on the influence of genetic differences on intelligence are in broad agreement. Intelligence, as measured by IQ tests, has a substantial heritability. Estimates of heritability vary between 80 per cent (B uchard et al., 1990) and 50 per cent (Plomin 1990). So even the more conservative estimates argue that genetic differences are far from trivial – they are at least as important as environmental differences, and maybe more so. The Bouchard et al. (1990) study is particularly important because these researchers measured a number of variables that can potentially confound twin studies (such as the length of time the twins had been in contact with each other) and attempted to determine their influence on the estimate of heritability. It turns out that these effects are minor, contributing at most 3 per cent to the estimate of 70–80 per cent heritability in their study. The many studies from the Colorado Adoption Project (see Plomin, 1990) estimate the heritability of intelligence at about 50 per cent. They suggest that the shared environment is more influential early in development than in later life. For example, the correlation between adopted children and their biologically unrelated siblings (who are usually reared from birth in the same family) averages around 0.2– 0.3 before their teenage years.

The importance of life events
Over the whole lifespan it seems that the most important environmental differences are those that are non-shared and unique to the individual concerned (that is, they are not shared by members of the same family). So rather than the major socio-economic variables (which represent a large part of the shared, or common, environmental variance) being the principal environmental contributor to difference in intelligence, it is unique life events that make up the major environmental contribution.


[Sir Cyril Burt (1883–1971) encouraged new methodological rigour in data analysis through his use of factor analysis of complex data sets. He also contributed significantly to the development of intelligence testing methods, schools for children with intellectual disabilities, child guidance clinics, and the 11+ testing system in the UK in which all 11-year-olds were assessed for intellectual potential to provide optimal educational opportunities. However, it is probably for his analysis of twin IQs that he is best known. Burt compared twins raised together with those adopted out and concluded that intelligence is largely hereditary. In the latter part of his career, Burt was charged with falsifying data in his groundbreaking twin studies, but his findings have been supported by more recent research.]


In a review of adoption and twin studies, Scarr (1992) estimated that the contribution of the shared environment to differences in IQ is approximately zero by adulthood. This is consistent with the finding that the heritability of g increases throughout our lives (McGue et al., 1993), beginning at about 20–30 per cent in early childhood and increasing to about 50 per cent after adolescence (Bishop et al., 2003; Spinath et al., 2003). This may be explained by the increasing influence of the biological underpinning of intelligence across the lifespan, as the effect of the shared environment decreases. All this means that, irrespective of our shared environment, most of us find ways ultimately to realize our genetic potential, depending on the effects of our idiosyncratic life events (i.e. nonshared environment).


Finding the IQ gene(s)
Most recently great excitement has surrounded the methodology of quantitative trait loci (QTL), which attempts to associate particular genes with specific behaviours. Researchers compare the DNA of a tightly defined group of individuals considered ‘high’ on some trait with the DNA of control individuals who, ideally, only differ by being ‘low’ on the same trait. In so doing, they hope to find genes that contribute to difference between the two groups. This method has been successful at finding genes that appear to be associated with discrete pathological conditions, such as reading disorder (Cardon et al., 1994) and autism (Bailey et al., 1995). But the general consensus is that intelligence must be polygenic, which means that many genes contribute in an additive or dose-related fashion to IQ differences. If this is right, current QTL methods have very little chance of discovering the individual genes that each contributes only a relatively small proportion to the overall genetic effect. Even so, some research ers claim to have discovered a gene that is over-represented in individuals with a very high g (Chorney et al., 1998). While exciting, this methodology is new, and its results should be treated with caution.

Almost everyone now accepts that there are genetic influences on IQ differences, but the most important recent discoveries concern environmental rather than genetic influences, particularly the finding that it is the non-shared environment that has a lasting effect on individual intellectual differences. The challenge is to move on from the heritability issue to theories of how genetic predispositions may interact and correlate with environmental circumstances to produce the patterns of IQ differences that we find in our society (see Scarr, 1992).