Experimental Studies of Cultural Evolution

Abstract

Abstract This chapter reviews how experiments have played a growing role in the study of cultural evolution recently. Experiments occupy a middle ground between formal models and real-world observational and historical studies. Like models, they offer the advantages of control and manipulation in order to determine causal relations between variables. Like real-world studies, they involve actual human behaviour, albeit constrained within artificial boundaries. When interpreted within the context of this trade-off between external and internal validity, experiments can provide valuable insights into the processes and dynamics of cultural evolution. Given the centrality of social learning in theories of cultural evolution, experiments typically involve participants learning from at least one other participant. Various experimental designs have been employed including dyadic interactions, linear transmission chains, closed groups, replacement or migration methods, and economic games. Key findings relate to content and inductive biases that describe how the content of cultural traits affects their transmission and evolution; context biases concerning from whom traits are copied; studies probing the socio-cognitive and demographic factors that underpin cumulative cultural evolution; the burgeoning experimental study of non-human culture; and variation in learning strategies over the lifetime and across cultures. The chapter ends with some recommendations for future experimental research, including an awareness of the proper function and limitations of experimental methods; the need for closer links between experiments, theory, and real-world data; and the necessity for experiments to be both replicable and reproducible via open science practices, to ensure that new generations of scientists can build on reliable knowledge.

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 16 citations exist upstream.

Papers that cite the same references — the relational fallback while embeddings are still backfilling.