Identical Twins, Different Families: Can Twin Studies Separate Nature from Nurture?
Few topics in A Level Psychology capture the nature-nurture debate as neatly as twin studies.
Take two ordinary siblings. If they have similar intelligence, personalities or interests, what caused the similarity?
They share some genes.
They probably grew up in the same home.
They may have attended similar schools, eaten similar food, been read similar books and been encouraged by the same parents.
Immediately, genetics and environment become tangled together.
Now consider twins.
And then consider something much rarer and scientifically fascinating:
identical twins who were separated when very young and raised in different families.
Suddenly we have something approaching a natural psychological experiment.
The twins have extremely similar genetic inheritance, but parts of their environments are different.
So if they grow into remarkably similar adults, does that demonstrate the power of genes?
And if they become very different, does that demonstrate the importance of environment?
As is so often the case in Psychology, the real answer is much more interesting than either of those simple conclusions.
First, Why Are Twins So Useful to Psychologists?
There are two main types of twins relevant to behavioural genetics.
Monozygotic twins
Monozygotic twins, usually called identical twins, develop when a single fertilised egg divides.
They therefore share virtually all their DNA sequence.
Dizygotic twins
Dizygotic twins, often called fraternal or non-identical twins, develop from two different eggs fertilised by two different sperm cells.
Genetically, they are approximately as similar as ordinary brothers and sisters, sharing on average about 50% of their segregating genetic variants.
That difference gives psychologists an extremely useful comparison.
Suppose we measure intelligence in large numbers of twins.
If identical twins are considerably more similar in intelligence than non-identical twins, one possible explanation is that genetic differences contribute to differences in intelligence.
That is the basic logic behind the classical twin study.
Correlation Is the Key
Psychologists usually do not expect two twins to receive exactly the same IQ score.
Instead, researchers look at correlations.
Imagine we tested hundreds of pairs of twins.
If Twin A scored highly and Twin B also tended to score highly, while lower-scoring Twin A partners tended to have lower-scoring Twin B partners, the correlation would be positive.
If identical twins produced a correlation of, say:
r = 0.80
while non-identical twins produced:
r = 0.45
the greater similarity between the identical twins would suggest a genetic contribution.
A simplified estimate sometimes introduced when explaining the classical twin method is:
Heritability = 2 x (rMZ - rDZ)
where:
rMZ = correlation between monozygotic twins
rDZ = correlation between dizygotic twins
However, real behavioural-genetic research uses much more sophisticated statistical modelling than this simple calculation.
And there is another extremely important warning.
Heritability is about variation within a population. It is not a percentage describing an individual person.
If a study estimates the heritability of intelligence at 60%, it does not mean that 60% of your intelligence was produced by your genes and 40% by your environment.
That interpretation is wrong.
Then Comes the Fascinating Case: Identical Twins Raised Apart
Identical twins raised together share two very important things:
extremely similar genes;
much of their childhood environment.
That creates a problem.
Perhaps they are similar because of their genes.
But perhaps they are similar because their parents treated them similarly.
Or perhaps both are involved.
This is why monozygotic twins reared apart are so valuable.
Imagine identical twins separated shortly after birth.
One grows up in London.
The other grows up in Edinburgh.
They have different parents, different schools, different friends, perhaps different family incomes and very different childhood experiences.
Thirty years later psychologists locate them and test them.
If their intelligence remains strongly correlated, shared upbringing becomes a much less convincing explanation.
Their genetic similarity becomes much more interesting.
The Famous Minnesota Twins Reared Apart Study
One of the best-known investigations was the Minnesota Study of Twins Reared Apart, associated particularly with Thomas Bouchard and colleagues.
Beginning in 1979, researchers studied more than 100 sets of twins or triplets who had been separated and raised apart, subjecting participants to extensive psychological and physiological testing.
Bouchard and colleagues reported that approximately 70% of the variance in IQ in their sample was associated with genetic variation. Identical twins raised apart also showed striking similarities across several other psychological characteristics.
This is powerful evidence for a genetic contribution to intelligence.
But notice the wording:
a genetic contribution.
It does not demonstrate that intelligence is fixed genetically.
The twins were not identical in intelligence.
That difference is important too.
Where identical genes are associated with similarity, psychologists investigate genetic influences.
Where genetically identical individuals differ, psychologists have evidence that genes cannot be the whole explanation.
Why “Raised Apart” Does Not Mean “Environment Removed”
This is one of the most important evaluation points for an A Level student.
It is tempting to imagine that identical twins raised apart have:
Same genes + completely different environments
Unfortunately, real life is rarely that tidy.
Twins raised apart may still share environmental characteristics.
For example:
adoption agencies may place children into broadly similar families;
both families may have similar socioeconomic backgrounds;
both twins may grow up within the same culture;
both may receive similar levels of education;
the twins shared the same prenatal environment before birth;
twins may discover one another and have contact later;
adoptive families are not randomly selected from every possible environment.
This creates the problem of selective placement.
Suppose both children are deliberately placed into stable, relatively well-educated homes.
Their environments may be more similar than the phrase “raised apart” suggests.
Therefore:
similarity between separated twins cannot automatically be attributed entirely to genetics.
And Twins Raised Together Create Another Problem
Traditional twin studies also make what is known as the equal environments assumption.
The argument goes something like this:
Identical twins share more genes than non-identical twins.
If identical twins are more psychologically similar, the difference can therefore be attributed partly to greater genetic similarity.
But what if identical twins are also treated more similarly?
Parents may:
dress identical twins similarly;
encourage the same activities;
put them in the same classes;
buy them the same toys;
expect them to behave similarly;
encourage a shared identity.
Other people may treat them more similarly simply because they look alike.
If the environments experienced by identical twins are more similar than those experienced by non-identical twins, some of the higher identical-twin correlation could reflect environment as well as genetics.
Again, nature and nurture become difficult to separate.
Van Leeuwen and the Twin-Family Study of Intelligence
An especially useful study for taking this topic further was conducted by Marieke van Leeuwen, Stéphanie van den Berg and Dorret Boomsma.
Rather than studying only pairs of twins, their research used an extended twin-family design.
Their sample involved 112 families, including twins, their siblings and their parents.
This is clever because researchers gain more comparisons.
They can examine similarities between:
identical twins;
non-identical twins;
ordinary siblings;
children and parents;
husbands and wives.
That gives researchers much more information than simply comparing identical and non-identical twins.
But How Do You Measure “Intelligence”?
This question deserves far more attention than it sometimes receives.
We talk casually about someone being “intelligent”, but psychologists need an operational definition.
They need something they can actually measure.
Van Leeuwen and colleagues used the Raven Progressive Matrices.
Participants are presented with patterns containing a missing element and have to determine which option correctly completes the pattern.
A simple example might look conceptually like this:
Triangle -> Square -> Pentagon -> ?
The participant has to identify the underlying rule rather than simply recall a fact they have learned.
Real Raven questions are considerably more sophisticated and use visual patterns rather than little sequences like this.
The test is particularly associated with abstract and non-verbal reasoning.
Van Leeuwen's team used performance on Raven matrices and estimated general intelligence using a Rasch measurement model.
That point matters.
The researchers did not somehow observe “intelligence” directly.
They observed performance on psychological tasks from which intelligence was estimated.
Is an IQ Test Really Measuring Intelligence?
This provides an excellent evaluation question.
Raven's matrices have advantages.
Because they rely less heavily on vocabulary than some intelligence tests, they may reduce some effects of language and formal knowledge.
But “less dependent on education” does not mean “independent of environment”.
Performance can still potentially be affected by:
schooling;
familiarity with tests;
concentration;
motivation;
anxiety;
fatigue;
understanding instructions;
experience solving abstract puzzles;
health;
nutrition;
developmental opportunities.
So when we say that researchers are investigating the inheritance of intelligence, we need to be more precise.
They are studying individual differences in measured cognitive performance.
That is not quite the same thing as discovering a single biological quantity called intelligence.
What Did Van Leeuwen's Study Find?
The researchers found that identical twins resembled one another more strongly in IQ than first-degree relatives such as non-identical twins, siblings and parent-child pairs.
After controlling for unreliability in the measurement scale, the model estimated that additive genetic effects accounted for around 67% of population variance in measured IQ.
They also reported a correlation of about r = 0.33 between spouses' IQ scores. Their modelling favoured the idea of phenotypic assortment — people with similar intelligence tending to partner with one another — rather than similarity arising purely because people from similar social backgrounds meet each other.
That is important because the classical twin model can become more complicated if mating is not random for the characteristic being studied.
But perhaps an even more interesting result was evidence suggesting that genetics and environment did not simply operate as two independent forces.
The researchers estimated that a further portion of variation was associated with gene-environment interaction.
And this leads us beyond the simple nature-versus-nurture argument.
Perhaps Nature Versus Nurture Is the Wrong Question
Students often begin this topic imagining two competing explanations.
Nature
Your genes determine your intelligence.
Nurture
Your upbringing determines your intelligence.
But modern behavioural genetics suggests something much more complicated.
Genes can influence how people respond to environments.
And environments can influence how genetic differences are expressed.
This is called gene-environment interaction.
A Practical Example: Same School, Different Effect
Imagine two children attend exactly the same mathematics lesson.
Same teacher.
Same textbook.
Same classroom.
Same explanation.
We might describe that as a shared environment.
But it does not follow that the lesson has the same psychological effect on both children.
One student may understand the pattern quickly, become interested, attempt harder questions and receive positive feedback.
The other may struggle initially, become frustrated and avoid the subject.
One shared environmental event has produced two different experiences.
This is why simply listing “genes” and “environment” as separate causes can be misleading.
Genes Can Also Help Create Environments
Consider another possibility.
A child who finds reading relatively easy may start reading voluntarily.
Because they read more, they develop a larger vocabulary.
That makes reading increasingly enjoyable.
They start choosing more difficult books.
Teachers notice their ability and recommend additional material.
The original differences may have contained a genetic component.
But that genetic influence has helped produce a particular environment.
The environment then strengthens the behaviour.
This is an example of why psychologists discuss gene-environment correlation.
Genes do not operate in isolation from experience.
They can partly influence the experiences people seek, receive and create.
An Important Thought Experiment for Students
Imagine we discovered two genetically identical babies.
We raise Baby A in an environment containing:
excellent nutrition;
stimulating conversation;
books;
good healthcare;
high-quality schooling;
opportunities to explore;
adults who encourage curiosity.
Now imagine Baby B experiences:
severe nutritional deprivation;
little stimulation;
interrupted schooling;
chronic stress;
serious illness;
few learning opportunities.
Would anyone seriously expect their cognitive development to be identical simply because their genes were identical?
Of course not.
Genes affect development within an environment.
This is why high heritability does not mean that environmental interventions are useless.
Bouchard and colleagues themselves explicitly noted that evidence for substantial heritability did not reduce the importance of education and other interventions.
Environment Can Have Measurable Effects
Adoption research gives us another way of approaching the question.
A large Swedish study compared siblings where one child had been raised by biological parents while another had been adopted into a different family.
The researchers found that adoption into more advantaged socioeconomic circumstances was associated with higher measured cognitive ability at age 18.
This provides evidence that rearing environment can influence cognitive outcomes, even when genetic effects are substantial.
So the evidence does not support the simplistic conclusion:
“Intelligence is genetic.”
Nor does it support:
“Intelligence is produced entirely by upbringing.”
Instead, intelligence appears to emerge from a complicated developmental system involving both.
Another Surprise: Heritability Can Change with Age
There is another finding that often surprises students.
The estimated heritability of intelligence does not necessarily remain constant throughout life.
A longitudinal Dutch twin study found that estimated heritability of full-scale IQ increased from about 34% at ages 9-11 to around 65% at ages 12-14, while estimated shared environmental influence decreased.
At first this can sound extraordinary.
Surely we accumulate more environmental experiences as we get older?
Yes.
But as children gain independence, genetically influenced preferences may increasingly affect which environments they choose.
The child interested in music practises more music.
The child fascinated by numbers chooses mathematical activities.
The strong reader reads more.
Small initial differences can become amplified through experience.
Once again, genes and environment are interacting rather than taking turns.
Why Twin Studies Are Powerful
Twin studies have several major strengths.
They provide naturally occurring comparisons
It would obviously be completely unethical to deliberately separate identical twins merely to conduct an experiment.
Researchers instead study naturally occurring circumstances.
They allow genetic similarity to vary systematically
Identical and non-identical twins provide different degrees of genetic relatedness while often having broadly similar family backgrounds.
They can produce quantitative evidence
Researchers can calculate correlations and construct statistical models rather than relying simply on anecdotal similarities.
Reared-apart twins are particularly informative
When identical twins remain similar despite different childhood homes, purely shared-family explanations become less convincing.
But Twin Studies Have Important Limitations
Good Psychology requires evaluation, not simply memorising a result.
1. Twins raised apart are extremely rare
Large representative samples are difficult to obtain.
Researchers may therefore work with unusual groups of participants.
That raises questions about generalisability.
2. “Apart” does not mean completely different environments
Adoption placement, social class and culture may make the homes more similar than expected.
3. Prenatal environment is shared
Identical twins raised in different homes still spent their prenatal development together.
4. Identical twins may be treated unusually similarly
This can complicate comparisons with non-identical twins.
5. Intelligence itself is difficult to operationalise
An IQ score is a measurement obtained from particular psychological tasks.
It should not automatically be treated as a complete measurement of every aspect of human intellectual ability.
6. Correlations do not prove individual causation
A high correlation between identical twins does not identify particular genes or explain the biological mechanism producing a behaviour.
Do Not Confuse Heritability with Inevitability
This is probably the single most important sentence in this whole topic:
A heritable characteristic can still be strongly influenced by environment.
Height provides an obvious analogy.
Human height is substantially heritable.
But serious childhood malnutrition can reduce adult height.
Genes influence the developmental range.
Environment affects how development actually proceeds.
Intelligence is considerably more complicated than height, but the same basic warning applies.
A population-level heritability estimate tells us something about why people differ under the conditions in which that population was studied.
It does not tell us what a particular individual could achieve under different circumstances.
Why I Think Twin Studies Are Such a Good A Level Psychology Topic
What I like about twin research is that it initially appears to offer a beautifully simple experiment.
Identical twins have the same genes.
Separate them.
See what happens.
But the more carefully we examine the idea, the more complicated it becomes.
What counts as the environment?
Are two adoptive families really independent environments?
How do we measure intelligence?
Does a test score represent intelligence itself?
Can genes influence which environments a person experiences?
Can environmental effects depend upon genotype?
Suddenly a simple nature-nurture comparison has become a lesson in research methods, correlations, operationalisation, validity, ethics, biological psychology and individual differences.
That is exactly what makes Psychology interesting.
Turning This into an A Level Exam Argument
Suppose the question asked:
“Discuss the contribution of twin studies to our understanding of intelligence.”
A strong argument might develop like this:
Point: Twin studies provide evidence for genetic influences on intelligence.
Evidence: Identical twins share virtually all their genes, whereas non-identical twins share about half of their segregating genetic variation. Greater similarity in identical twins therefore supports genetic influence.
Example: Bouchard's study of twins raised apart reported substantial similarity in intelligence despite separate rearing environments.
Evaluation: However, twins raised apart may still experience similar cultural and socioeconomic environments through selective placement, so environmental similarity cannot be eliminated completely.
Further evidence: Extended twin-family research such as Van Leeuwen et al. found substantial additive genetic influence on measured IQ.
Further evaluation: Nevertheless, evidence for gene-environment interaction suggests that separating behaviour neatly into genetic and environmental percentages may oversimplify development.
That is much stronger than writing:
“Bouchard proved intelligence is inherited.”
He did not.
The Bigger Lesson: Genes Are Not a Script
Perhaps the greatest contribution of twin research is not that it finally settles the nature-nurture debate.
It is that it shows why the original debate was too simplistic.
Identical twins raised in different families can sometimes remain remarkably similar.
That tells us that genetic variation matters.
But identical twins are never psychologically identical.
That tells us something else matters too.
Environment matters.
Individual experience matters.
Development matters.
And increasingly, research suggests that genes and environments continually influence one another.
So instead of asking:
“Is intelligence inherited or learned?”
perhaps the better psychological question is:
“How do genetic differences and environmental experiences interact during development to produce the differences in intelligence that we observe?”
That is a much harder question.
But it is also a much better one.
And that is often where the most interesting Psychology begins.
No comments:
Post a Comment