A theory limited in scope and evidence.

What promised to be a refreshing addition to cumulative cultural evolution, by moving the focus from cultural transmission to technological innovation, falls flat through a lack of thoroughness, explanatory power, and data. A comprehensive theory of cumulative cultural change must carefully integrate all existing evidence in a cohesive multi-level account. We argue that the manuscript fails to do so convincingly.

Flexible learning, rather than inveterate innovation or copying, drives cumulative knowledge gain.

Human technology is characterized by cumulative cultural knowledge gain, yet researchers have limited knowledge of the mix of copying and innovation that maximizes progress. Here, we analyze a unique large-scale dataset originating from collaborative online programming competitions to investigate, in a setting of real-world complexity, how individual differences in innovation, social-information use, and performance generate technological progress. We find that cumulative knowledge gain is primarily driven by pragmatists, willing to copy, innovate, explore, and take risks flexibly, rather than by pure innovators or habitual copiers. Our study also reveals a key role for prestige in information transfer.

Cultural adaptation is maximised when intelligent individuals rarely think for themselves.

Humans are remarkable in their reliance on cultural inheritance, and the ecological success this has produced. Nonetheless, we lack a thorough understanding of how the cognitive underpinnings of cultural transmission affect cultural adaptation across diverse tasks. Here, we use an agent-based simulation to investigate how different learning mechanisms (both social and asocial) interact with task structure to affect cultural adaptation. Specifically, we compared learning through refinement, recombination or both, in tasks of different difficulty, with learners of different asocial intelligence. We find that for simple tasks all learning mechanisms are roughly equivalent. However, for hard tasks, performance was maximised when populations consisted of highly intelligent individuals who nonetheless rarely innovated and instead recombined existing information. Our results thus show that cumulative cultural adaptation relies on the combination of individual intelligence and 'blind' population-level processes, although the former may be rarely used. The counterintuitive requirement that individuals be highly intelligent, but rarely use this intelligence, may help resolve the debate over the role of individual intelligence in cultural adaptation.

Innovation and cumulative culture through tweaks and leaps in online programming contests.

The ability to build progressively on the achievements of earlier generations is central to human uniqueness, but experimental investigations of this cumulative cultural evolution lack real-world complexity. Here, we studied the dynamics of cumulative culture using a large-scale data set from online collaborative programming competitions run over 14 years. We show that, within each contest population, performance increases over time through frequent 'tweaks' of the current best entry and rare innovative 'leaps' (successful tweak:leap ratio = 16:1), the latter associated with substantially greater variance in performance. Cumulative cultural evolution reduces technological diversity over time, as populations focus on refining high-performance solutions. While individual entries borrow from few sources, iterative copying allows populations to integrate ideas from many sources, demonstrating a new form of collective intelligence. Our results imply that maximising technological progress requires accepting high levels of failure.