The shape of the cultural transmission of shapes

Cultural evolutionary theory has long modeled offspring trait variance as constant across generations, an assumption borrowed from quantitative genetics that has remained largely untested. We examine this by comparing a Parental Variability-Dependent Inheritance (PVDI) model, in which offspring variance scales with parental variance, against the classic Galton-Pearson model, using physical artifacts transmitted across three generations in a cross-cultural experiment spanning Prague and two communities in Cameroon. The PVDI model outperformed alternatives in all tasks, though its expression was modulated by learner background rather than task domain. Cumulative improvement emerged in the engineering but not the aesthetic task, driven by individual-level innovation rather than performance-biased selection. These findings challenge a foundational assumption of the field, make a strong case for cross-cultural designs in cultural evolution research, and introduce a novel framework for the parametrization of the cultural transmission process.