Sewall Wright's (1943) concept of isolation by distance is as germane to cultural transmission as genetic transmission. Yet there has been little research on how the spatial scale of social learning—the geographic extent of cultural transmission—affects cultural diversity. Here, we employ agent-based simulation to study how the spatial scale of unbiased social learning affects selectively neutral cultural diversity over a range of population sizes and densities. We show that highly localized unbiased cultural transmission may be easily confused with a form of biased cultural transmission, especially in low-density populations. Our results have important implications for how archaeologists infer mechanisms of cultural transmission from diversity estimates that depart from the expectations of neutral theory.