A few comments about genetics
Twin Studies:
Genetic studies using identical twins consistently report heritability estimates far higher than those produced by genome-wide association studies (GWAS). This gap is one of the more puzzling findings in behavioral genetics, and I believe it points to a fundamental misunderstanding of what "environment" actually means in the context of psychological and emotional development.
Researchers typically characterize shared versus non-shared environments in broad, surface-level terms: whether siblings had the same teachers, shared a bedroom, or were dressed alike. But these factors are largely irrelevant to how personality and temperament form. What actually shapes emotional development is something far more fine-grained — the moment-to-moment emotional signals a child receives from their parents, particularly those conveyed through facial expression.
Humans are remarkably sensitive to subtle variations in facial expression. Even small differences in how an emotion is displayed can produce meaningfully different responses in an observer. And facial structure matters here in a way that has been largely overlooked: when the same internal emotion is expressed by people with different faces, it does not look the same. Identical feelings produce subtly different external signals depending on the face expressing them.
This has significant implications for how children develop. Two children with different facial structures will express the same emotion somewhat differently. A parent, responding to what they perceive, will react in subtly different ways to each child. Each child then receives a slightly different emotional signal in return, which itself gets expressed differently on their respective faces, prompting yet another differentiated response from the parent. Over many such exchanges, these small differences accumulate into distinct emotional patterns, gradually shaping each child's temperament and personality in divergent directions.
Identical twins, however, share nearly identical facial structure. The same emotion therefore appears nearly identical on both faces, and the parent's responses are correspondingly similar. The feedback loop between expression and response remains far less divergent for identical twins than for non-identical siblings.
The heritability gap, in other words, is not a genetic mystery. It reflects the fact that researchers have been measuring the wrong environment. The relevant environment is not the one defined by classrooms and bedrooms, but the one built from thousands of subtle, reciprocal emotional exchanges between parent and child — exchanges shaped, in part, by the faces involved.
Newer Studies: What GWAS Shows, and What It Cannot Show
Modern studies often use genome-wide association studies (GWAS), which look for genetic variants that are statistically associated with a trait.
GWAS results point to many small influences, not single genes for complex traits. For traits like intelligence, personality, psychiatric diagnoses, and athletic performance, it finds many genetic differences, each one associated with only a very small difference in the trait.
A GWAS association does not prove a genetic cause. It can reflect ancestry or environment rather than a biological effect.
GWAS depends on how the trait is defined and measured. If a diagnosis contains many different underlying problems, or if a trait is measured loosely, the results become harder to interpret. The study can find associations while still leaving the real causes unclear.
Bottom line. GWAS supports the idea that genetics can contribute to complex traits, but it also shows that the effects are spread across many variants and are hard to translate into clear biological explanations or practical predictions. These limits leave plenty of room for large effects from childhood experience and for large differences in adult outcomes that are not locked in by DNA.
In Addition:
Genetic studies measure differences between people in their current environments, not what people could become under the best developmental conditions. So even if current genetic findings are fully accurate, they do not rule out most of the outcomes I list on the Introduction page.