Reading focus
Syllabus: Where Do Individual Differences Come From?
Compare cognitive and personality domains with the all-trait summary, and examine what the resemblance patterns leave unidentified.
Why you’re reading it. Learn to evaluate broad genetic evidence without turning an average across heterogeneous traits into a universal constant.
A question to consider. Maya and Leo share a household. What would a small shared-environment estimate actually say about resemblance in a population, and why would it not show that their household had no influence?
Why this reading
A broad synthesis makes it possible to examine the reach of twin evidence across many traits. It also reveals why a striking overall average is less informative than understanding what was measured, which populations were studied, and how resemblance was modeled.
Before you read
Identical twins share nearly all their inherited DNA; fraternal twins share, on average, half their segregating genetic variants. Twin models use differences in resemblance to estimate components of variation under assumptions about genetic and environmental similarity. The nonshared environmental component also includes measurement error; it is not a catalogue of experiences unique to each person.
Read for
- Compare trait domains and sample coverage before interpreting the overall heritability average.
- Distinguish a pattern compatible with a simple model from evidence that every omitted influence is absent.
- Examine what twin correlations can identify and which combinations of genetic and shared environmental effects remain difficult to separate.
Consider afterwards
- A report treats a small shared-environment estimate as proof that families do not matter. What does the model estimate, and which effects of family life would that conclusion leave unresolved?
Editorial context
The synthesis covers much more than psychology, and its average is not a value to assign to every personality trait or cognitive ability. Much of the evidence comes from the United States and Europe. The reported twin-pair total combines partly overlapping observations, rather than counting unique pairs. Compatibility with an additive model does not establish the absence of shared environments or more complex processes; the paper discusses limits to identifying them. Twin estimates do not specify molecular mechanisms or predict an individual’s response to an intervention.