Abstract:
Heritability is a central concept in the long-standing debate about nature versus nurture in biological and social sciences. However, existing notions of heritability are based on strong assumptions and do not use explicit causal models. We propose a new, counterfactual definition of heritability by adopting the potential outcomes model in causal inference. Our counterfactual heritability measures the importance of genetic inheritance by the average magnitude of squared difference between an individual and their hypothetical ``non-identical twin'' that is exposed to the same environment. We provide bounds on the counterfactual heritability that can, in principle, be computed from observational data. We then compare counterfactual heritability and its associated bounds with common notions of heritability in population-based studies, twin and sibling studies, and plant breeding experiments. Our results and comparisons highlight the importance of clarifying the causal structural assumptions and counterfactual comparisons in reasoning about heritability.

Brief bio: Hongyuan Cao is a professor of statistics at Florida State University. She got her Ph.D. from UNC-Chapel Hill. Her research develops rigorous and scalable statistical methods for high dimensional biomedical data, with applications in genomics, clinical studies, and public health. She focuses on multiple testing, replicability, and complex survival and longitudinal data to ensure reliable and reproducible scientific discovery. More recently, she integrates statistical inference with clinical AI to build trustworthy, interpretable, and causally grounded models for precision health and health care decision-making. She is an elected fellow of ASA.