Advancedpopulation
Heritability Explained
Overview
Heritability (h^2) is the fraction of phenotypic variance due to additive genetic variance IN A SPECIFIC POPULATION AND ENVIRONMENT: not a property of individuals, not fixed across contexts, and not a measure of immutability. Estimated from twin, adoption, and parent-offspring studies. Misunderstood constantly: high heritability does not mean unchangeable (height rose hugely via nutrition) or within-group causes explain between-group differences.
Learning objectives
- Define heritability precisely
- Explain how it is estimated
- Debunk the major misconceptions
- Explain gene-environment interplay
The definition, carefully
h^2 = VA/VP in a given population in a given range of environments. Height h^2 ~ 0.8 in modern well-nourished populations: 80% of height VARIATION traces to genetic differences. It says nothing about why YOU are your height, or whether heights could change environmentally.
Three words that prevent 90% of heritability misreadings.
Estimation
Parent-offspring regression slope ~ h^2/2; twin comparisons (2*(rMZ - rDZ)); adoption studies separating genes and rearing environment; modern SNP heritability (GREML/GCTA) from unrelated individuals' marker similarity. Each method has assumptions and biases.
Twin studies assume MZ and DZ pairs share environments equally: violated if identical twins are treated more alike.
The great misconceptions
High h^2 does not mean immutable: Dutch heights gained ~20 cm in 150 years via nutrition despite h^2 ~ 0.8. Within-group heritability says nothing about between-group differences (the heritability of group differences requires separate analysis). h^2 = 0.8 does not mean '80% genetic' for an individual.
Heritability ~0.8 AND massive environmental gains coexist: variance fractions ≠ fixed biology.
Gene-environment interplay
G×E interaction: genotypes respond differently to environments (stress-sensitive variants). rGE correlation: genotypes shape environments (active/evocative/passive). Heritability RISES in uniform environments and FALLS in variable ones: it measures context, not essence.
Change the environment's range and h^2 changes: it is a local statistic, not a constant.
Key points
- h^2 = VA/VP: a population variance fraction in a context
- Estimated via relatives, twins, adoptees, SNP methods
- High heritability ≠ immutable, ≠ individual, ≠ between-group explanation
- G×E and rGE complicate every estimate
Common misconceptions
✓ Truth: Heritability describes variation within studied environments; interventions can still shift means substantially.
✓ Truth: Within-group h^2 provides no logical basis for attributing between-group differences to genes.
Practice questions
Q1. Heritability is best defined as
- Genetic determination of individuals
- The fraction of phenotypic variance due to additive genetic variance in a population
- Immutability
- A constant
Population-, variance-, and context-specific by definition.
Q2. High heritability does NOT imply
- Genetic influence on variation
- That the trait cannot change environmentally
- Additive variance
- Familial resemblance
Means shifted hugely (height, IQ) despite high h^2: variance ≠ fixed potential.
Q3. Twin heritability estimates assume
- MZ and DZ pairs share environments equally
- Genes don't matter
- No environment
- Clones
The equal-environment assumption: violations inflate h^2 estimates.
Click a question to reveal the answer and explanation.
Frequently asked questions
Why did Dutch height rise if h^2 is 0.8?
Heritability partitions variation around the current mean; improving nutrition shifted the whole mean upward.
What is SNP heritability?
h^2 estimated from genome-wide markers in unrelated people (GREML): captures common-variant additive effects.
Can heritability be 0 or 1?
Yes in principle: 0 if no genetic variation (clones in one environment), 1 if no environmental variation.