Lecture 4 Predictors of Intelligence Tests Flashcards
environmental and genetic influences on intelligence
Nature or nurture?
Galton (1869): ‘natural abilities’ might be inherited, in the same way as physical features in the animal world (Darwin)
But Galton saw the influence of environment too - made the distinction between ‘nature’ and ‘nurture’
Heritability studies
How much of a certain attribute is caused by inheritance of genes; how much is caused by environment?
Estimated genetic heritability: average estimate of the proportion of variance thought to be accounted for by genetic factors across a population
Can only estimate
Plomin (2004) 3 main ways to study heritability
Three main ways to study heritability (Plomin, 2004):
Family studies – parents, siblings, children
Twin studies: monozygotic (identical) vs dizygotic (fraternal)
Adoption studies
Heritability studies simplified
Three factors:
A: additive genetic effects
C: common environment
E: unique environment
Identical twins share genes and environment: A+C
Fraternal twins share half genes and environment: 1/2 A + C
Adopted siblings share environment but no genes
Early estimates of heritability and intelligence vs now
Early estimates of heritability of intelligence quite high (genes + environment)
69% (Eysenck, 1979)
74% (Herrnstein & Murray, 1994)
Genes and environment are likely to interact
effects of environment will differ due to genetic makeup (genes x environment)
Epigenetics
Heritability different for…
Heritability is different for verbal and non-verbal IQ:
80% verbal IQ, 70% non-verbal IQ (Posthuma et al., 2001)
And for different ages:
23% early childhood and 62% middle childhood (Davis et al., 2009)
And across cultures:
70 – 80% in the Netherlands (Posthuma et al., 2001)
49 – 57% in Vietnam (Lyons et al., 2009)
Twins and adoption studies, assortive mating
Twin and adoption studies may not be representative
Identical twins have more similar environments than same-sex fraternal twins (Kamin & Goldberger, 2002)
Adoption agencies favour affluent families
Assortive mating
Couples tend to have a lot in common… or contrast strongly – affecting genetic variance (Mackintosh, 1998)
APA Task Force
The APA Task Force estimated that the heritability of intelligence ranges from 40% to 80% (Neisser et al., 1996)
Mackintosh (1998) 30% to 75%
Chipeur et al. (1990) 50%
Summary (heritability)
The extent to which intelligence is inherited can be estimated from twin and adoption studies
Estimates range from around 40% to 80%
Evidence suggests that genes and environment interact – not simply additive
Bouchard and Segal (1985) environmental influences
Bouchard and Segal (1985) listed 21 environmental factors that may contribute towards intelligence
malnutrition, weight at birth, height, years in school, parental economic status, parental education and influence, average TV viewing, self confidence, criminality, emotional adaptation…
… but the list is potentially limitless
The APA Task Force environmental influences
The APA Task Force identified four major areas of environmental influence on intelligence
Biological variables – nutrition, antenatal care
Family – socioeconomic status, birth order
School and education
Culture
Biological factors
Nutrition – as discussed last week
Positive effects of:
breastfeeding (Oddy et al., 2004)
vitamins (Benton & Roberts, 1988)
But controversial – issues with controlling for other factors (socioeconomic status, maternal intelligence) and replication
More biological factors
Toxins and deficiencies can have a negative effect on IQ – leads to gains when corrected
Lead exposure
Environmental exposure to lead associated with lower IQ in children (Baghurst et al., 1992; Tong et al., 1996)
Iodine deficiency
Meta-analysis (Qian et al., 2005) showed 9-point gain when children in iodine-deficient areas given supplements
Biological factors 3
Prenatal factors - pregnant women advised to stop smoking and drinking
Fetal alcohol syndrome (FAS) associated with impaired cognition (Kodituwakku et al., 1995)
but link between low levels of drinking and intelligence less clear (Jacobson et al., 2004; Bailey et al., 2004)
Maternal smoking (20+ per day) associated with lower IQ age 18 / 19 (Mortensen et al., 2005)
Family environment
Shared and non-shared environments
Socioeconomic status
Family size and birth order
Shared and non-shared environments
-Shared environment – family home, parents, school, etc.
-Non-shared environment – friends, teachers, personal relationship with parents
Non-shared environment sometimes underestimated, but can affect intelligence (Baker & Daniels, 1990; Plomin & Daniels, 1987)
Shared environment
Within family factors: Interaction with shared genetics to affect behaviour (Harris, 1995)
Reiss (1997) Model of genetic transmission – Passive, Child-effects, Parent-effects
Non-shared environment
May be more important than within-family factors (Harris, 1995)
Peer groups and social groups outside of the home
-e.g., reading books may be encouraged or discouraged
-gender stereotypes
-role within the group (leader / follower)
What is Socioeconomic status (SES)
SES is based on income, parental education level, parental occupation and status in the community
Related to a number of factors that might influence intelligence:
number of children, employment opportunities, health, environment, …
Socioeconomic status
Socioeconomic status is significantly correlated with IQ scores (r = 0.3 to 0.4) (Jensen, 1993a, Rushton & Ankney, 1996)
British National Child Development Study (NCDS)
-17,000 babies born in 1958
-10-point IQ difference between Class 1 (professional occupations and Class V (unskilled occupations)
-Even when controlling for financial hardship, birth weight, size of family, overcrowding, type of accommodation and residence area
SES Adoption studies
Improving socioeconomic status can improve intelligence
Adoption studies show IQ gains of 12 – 16 points when children are moved from low to high SES families (Wahlsten, 1997)
Family size and birth order
Family size
-Children from bigger families have a lower IQ (elmont & Marolla, 1973)
Birth order
-First-born children are more intelligent, then second-born, etc. (Belmont & Marolla, 1973)
-This probably accounts for the family-size effect (Rodgers et al., 2000)
-likely to be due to other factors associated with family size (SES, parental IQ, resource dilution, social factors)
Education
Meta-analysis (Ceci, 1990; 1991):
Raise in IQ of 2.7 points per year of schooling
Reductions over school holidays
More important than age for verbal tests, makes a smaller contribution to non-verbal tests (Cahan & Cohen, 1989)
But also linked to SES – difficult to separate effects
High SES = better preparation for school, better schools, better parental education, etc.
Culture
‘Culture’ refers to people’s individual values and the values of their society
Lay concepts of intelligence are different in different cultures
Neisser et al. (1996) suggest that culture can affect not only intelligence, but also the type of intelligence that might develop
Summary (environment)
There are numerous potential environmental influences on intelligence
The main areas that have been considered are biological factors, family environment, education and culture
But it’s important to remember that genetic and environmental factors will interact
Dark History (Eugenics)
Galton’s Eugenics Movement (1883): The science of improving a population by controlled breeding to increase the occurrence of desirable heritable characteristics
-Positive eugenics: Encouraging reproduction in those with desirable traits
-Negative eugenics: Discouraging or eliminating reproduction in those perceived to have poor hereditary traits
Dark History (Sterilization)
1922: The US Model Eugenical Sterilization Law
the feeble-minded (low intelligence), insane, criminal, epileptic, blind, deformed, dependent
42,000 people were sterilised between 1941 and 1943 (JAMA, 1945)
Hitler’s Sterilisation law in Germany in 1933: 20,000 people were sterilized for being ‘feeble minded’ between 1933 and 1939
In 1948, the United Nations affirmed that all people should be free to marry and to found a family
The Bell Curve controversy
Hernstein and Murray (1994): ‘The Bell Curve: Intelligence and Class Structure in American Life’
Herrnstein and Murray (1994) noted that:
-Low IQ is associated with negative outcomes (education, employment, health, crime, etc.)
-IQ is largely inherited
-Interventions aimed at improving IQ were not successful (e.g., Head Start)
-There are race differences in IQ (highest for Asian-Americans, then white Americans; lower for black Americans)
The Bell Curve: Themes
Hernstein and Murray (1994) discuss four main themes in the book:
-The cognitive elite
-Socioeconomic status and IQ
-The relationship between intelligence and race
-Implications for social policy
The cognitive elite
Described the emergence of a ‘cognitive elite’ in America, separated according to intelligence
-Found that college admissions were based on intelligence more than social class or wealth…
-… and that intelligence was the most powerful predictor of job success and workplace productivity
Predicted that intelligence would become the basis of the American class system
Socioeconomic status and IQ
Herrnstein and Murray (1994) investigated several contextual factors linked with IQ:
-poverty
-schooling
-unemployment
-family life
-welfare dependency
-crime
IQ predicting economic and social welfare
Thought that IQ was a strong precursor to poverty – more so than socioeconomic conditions
-e.g., IQ predicts likelihood of people completing high school
Concluded that intelligence is more important than socioeconomic status in predicting eventual economic and social welfare
Race and IQ
Compared IQ test results by ethnicity
Asian Americans scored highest (5 points above white Americans)
White Americans scored 15 points higher than black Americans (102 vs. 87)
Average immigrant IQ = 95
Race and IQ 2
Focused on the fact that IQ is largely heritable
Suggested that social factors were reducing average US intelligence
-parents with low IQ have more children
-increased immigration
Concerned that lower national IQ would increase social problems
Implications for social policy
One conclusion might be that more needs to be done to increase the intelligence of certain groups…
But Herrnstein and Murray thought:
-attempts to increase IQ have often failed (nutrition, Head Start)
-as IQ is inherited, IQ cannot be effectively raised through environmental changes
-affirmative action programmes have ‘failed’ and lead to resentment
Herrnstein and Murray proposed policy
Money for education should be shifted away from supporting disadvantaged children (predominantly African American) and towards those with high intelligence (predominantly white and Asian American)
Society will be improved through raising the intelligence of the nation and valuing the gifted
Criticisms of the Bell Curve
The scientific community suggested that the book was oversimplified and had flawed analysis
The American Psychological Association (APA) – Task force to investigate (Neisser et al., 1996)
-Conclusion: no definite evidence that race differences were genetic, not cultural
Criticisms of the Bell Curve focus on three main points:
the underlying assumptions (Gould, 1995)
the statistics and evidence base
The underlying assumptions (The Bell Curve)
Stephen Gould (1995)
Six key premises underlie their conclusions:
1. There is a general factor of cognitive ability (intelligence) on which people differ
- IQ tests measure intelligence accurately
- IQ scores reflect what most people mean by the word intelligence
- IQ remains stable across the lifetime
- IQ tests are not biased against any social or ethnic groups
- Heritability of intelligence is between 40% and 80%
Criticisms of their assumptions (Bell curve)
There is a general factor of cognitive ability (intelligence) on which people differ
-Maybe not, ‘g’ is not the only theory, also multifactor / multiple intelligences / cognitive abilities
IQ tests measure intelligence accurately
-depends on theory of intelligence. Also, tests can measure verbal, non-verbal, cognitive abilities, etc.
IQ scores reflect what most people mean by the word intelligence
-cultural differences in concepts of intelligence
More criticisms of their assumptions (bell curve)
IQ remains stable across the lifetime
-IQ scores are stable over time, and educational schemes have not lead to sustained benefits, BUT: benefits of schooling, adoption and nutrition
IQ tests are not biased against any social or ethnic groups
-APA task force (1996) agreed with this, but theory underlying IQ tests may be culturally biased
Heritability of intelligence is between 40% and 80%
-A big range!
-‘Race’ is socially defined, largely based on skin colour, and may not reflect underlying genetic differences
In summary of the criticisms of their assumptions
None of the assumptions are entirely wrong, but there is enough debate to undermine the conclusions
Criticisms of statistics & evidence base
Kamin (1995): socioeconomic status, IQ, economic and social welfare are all correlated - not causation
Kamin (1995) questions the validity of some of the measures used in the analysis…
-e.g., socioeconomic status was self-reported
… and some of the underlying research studies
-Review paper (Lynn, 1991) – genetics underlying intelligence differences between ethic groups
-Included studies where IQ was confounded by language or cultural differences
-Other studies were overlooked or reported as ‘inconsistencies’
Summary (race differences)
There are differences in IQ between different racial groups
The Bell Curve proposed that these were genetic and therefore unchangeable
The Bell Curve has been criticised on the grounds of its underlying assumptions, statistics, and evidence base
Considerable care needs to be taken in this debate to avoid links to eugenics
Sex differences in general intelligence
Generally no (or very small) differences found in general intelligence (g)
Girls scored slightly higher than boys on the original Stanford-Binet test (Terman, 1916)
Systematic review of 120 studies found no sex differences on Raven’s Matrices (Court, 1983)
review showed no sex differences on Wechsler tests (Anderson, 2004)
Some small sex differences
But: One meta-analysis did show sex differences (Lynn & Irwing, 2004)
-only took data from large studies (n > 500)
-57 studies, 30 countries, more than 80,000 people
Found small sex differences (d = 0.16 to 0.33)
-No sex differences up to the age of 15
-Age 15 – 19, men scored 2 points higher
-Male undergrads scored 3 – 5 points higher than female
-Adult men scored 5 points higher than women
Sex differences in specific intelligences
Men score better on measures of spatial ability than women (Maccoby & Jacklin, 1974)
Women tend to score better than men on measures of verbal ability
Meta-analyses of sex differences in specific abilities (Feingold, 1988; Hedges & Nowell, 1995; Hyde & Linn, 1988; Linn & Petersen, 1985; Voyer et al., 1995)
Men score higher:
Spatial perception (S-M)
Mental rotation (M-L)
Mechanical reasoning (adolescents, L)
Women score higher:
Verbal production (M)
Spelling, language & perceptual speed (S)
Biological explanations: brain size
IQ correlates with brain size (0.33, McDaniel, 2005)
Men have 10% larger brains than women
Lynn: Boys and girls mature at the same rate until 7, then girls mature faster until 14 or 15. Age 16+, boys develop larger average brain sizes
Spatial intelligence
Three biological explanations for sex differences in spatial ability have emerged:
Evolutionary
Brain functioning
Testosterone
Evolutionary
Jones et al. (2003) Good spatial abilities needed for:
-Foraging for food
-Covering a bigger range
-Competing for resources (fight)
Males who were good at those things: more likely to pass their genes on to the next generation
Brain functioning
Haier et al., (2005) Voxel-based morphometry (VBM)
IQ associated with different patterns of grey and white matter in men and women
Concluded that there was no single brain structure underlying intelligence
Testosterone
Testosterone is related to spatial ability
Route-learning: men tend to use cardinal directions and distance (Choi & Silverman, 2002)
Higher levels of testosterone are associated with faster responses and lower error rates on a mental rotation task (Hooven et al., 2004)
Stereotypes outside of education
Boys tend to play with toys that develop spatial ability (Levine et al., 2005)
Children aware of the sort of toys they ‘should’ play with (dolls / trucks) (Wood et al., 2002)
Spatial ability may develop more in boys because they are allowed more freedom to explore their neighbourhood (Entwistle et al., 1994)
Stereotypes within education
Subject choice may follow a pattern based on sex (Halpern & LaMay, 2000)
-Boys may choose maths and science
subjects that they excel at due to good
spatial abilities
-And girls may miss out on the opportunity
to develop spatial abilities because sciences
are stereotyped as ‘male’ subjects (Halpern
& LaMay, 2000)
More stereotypes within education
Self-fulfilling prophesies
-Tends to occur more within social groups than within the classroom (Jussim & Harber, 2005)
Stereotype threat (Steele, 1997)
-e.g., telling participants that a mathematics test shows sex differences leads to poorer performance by women (Spencer et al., 1999)
Sex differences and emotional intelligence
Women score significantly higher than men on Salovey and Mayer’s ability model of EI
-effect size 0.18 – 0.30 (medium) (Day & Carroll, 2004)
-effect size 0.30 (Schulte et al., 2004)
Emotional-Quotient Inventory (EQ-i; Bar-On, 1997)
-No significant difference between men and women on overall EI scores. Small effect sizes (<0.16) on subtests
Men and women score higher…
Men score higher:
Self-regard
Independence
Flexibility
Problem solving
Stress tolerance
Optimism
Women score higher:
Emotional self-awareness
Empathy (0.45)
Social responsibility
Interpersonal relationships
Summary (sex differences)
Men may score slightly higher on general IQ tests – but it’s a very small effect
Men tend to have better spatial ability
Women have sometimes been shown to have better verbal ability and emotional intelligence
Differences may be biological (brain size, evolution, brain functioning, testosterone)
Or they may be environmental (stereotypes, self-fulfilling prophesies, stereotype threat)
