Intermediatepopulation
Twin Studies
Overview
Twin studies compare monozygotic (identical, 100% genes) and dizygotic (fraternal, ~50%) twins: greater MZ similarity implies genetic influence. Falconer's formula (h^2 = 2*(rMZ - rDZ)) estimates heritability. Findings: schizophrenia ~0.8, height ~0.8-0.9, personality ~0.4-0.5, shared environment often near zero for adult personality. Adoption and separated-twin studies strengthen causal inference.
Learning objectives
- Explain the MZ/DZ logic
- Apply Falconer's formula
- Summarize major findings
- Critique assumptions and limits
The design
MZ twins: one zygote, ~100% shared genes. DZ twins: two zygotes, ~50% like siblings. Both share womb, family, cohort: the comparison isolates genetic sharing. rMZ > rDZ -> genetic influence; rMZ ~ rDZ -> shared environment dominates.
Genetic sharing varies while family environment is matched: nature's controlled trial.
Falconer's formula
h^2 = 2(rMZ - rDZ). Example: schizophrenia rMZ ~ 0.5, rDZ ~ 0.2 -> h^2 ~ 0.6 (liability scale ~0.8). Shared environment c^2 = 2rDZ - rMZ; non-shared e^2 = 1 - rMZ. Simple, powerful, assumption-laden.
A (additive genes) + C (shared environment) + E (non-shared): the variance decomposition twins enable.
Landmark findings
Schizophrenia/bipolar: h^2 ~ 0.7-0.8. Height: ~0.8-0.9. Big Five personality: ~0.4-0.5, mostly non-shared environment. IQ: rises with age (~0.2 childhood -> ~0.8 adulthood). Minnesota separated-twin studies found striking MZ similarities reared apart.
Schizophrenia h^2 ~ 0.8 yet most high-risk individuals never develop it: risk, not fate.
Criticisms
Equal-environment assumption (MZ treated more alike?), chorionicity differences (MZ twins sharing placentas), assortative mating inflating DZ similarity, generalizability (twins aren't singletons), and the missing-GWAS gap: twin h^2 exceeds SNP h^2, implicating rare variants and non-additive effects.
Twin estimates exceed molecular ones: the field's central puzzle, partly resolving via rare variants.
Key points
- MZ (100% genes) vs DZ (~50%): rMZ > rDZ implies genetic influence
- Falconer's h^2 = 2*(rMZ - rDZ); ACE decomposition
- Findings: schizophrenia/height ~0.8, personality ~0.4-0.5
- Assumptions (equal environments) and missing heritability are key caveats
Common misconceptions
✓ Truth: They estimate variance components under assumptions; molecular studies must identify actual variants.
✓ Truth: They differ in somatic mutations, epigenetics, and X-inactivation: 'identical' is approximate.
Practice questions
Q1. If rMZ = 0.8 and rDZ = 0.5 for a trait, Falconer's h^2 is
- 0.3
- 0.6
- 0.8
- 1.0
Falconer's formula gives h² = 2 × (rMZ − rDZ) = 2 × (0.8 − 0.5) = 0.6.
Q2. Greater MZ than DZ similarity suggests
- Shared environment only
- Genetic influence
- Nothing
- Measurement error
The design isolates genetic sharing as the difference.
Q3. A common criticism of twin studies is
- Too many twins
- The equal-environment assumption may fail
- Genes don't exist
- Math errors
If MZ pairs share more similar environments, h^2 is overestimated.
Click a question to reveal the answer and explanation.
Frequently asked questions
Why does IQ heritability rise with age?
Active rGE: genotypes increasingly select environments (reading, peers), amplifying genetic differences.
What did separated-twin studies show?
MZ reared apart remain strikingly similar (Minnesota study): genes matter even across households.
Do twins generalize to singletons?
Mostly yes for heritability, but twin-specific prenatal factors (shared womb, lower birth weight) add caveats.