Cultural Evolution as a Non-Stationary Stochastic Process

TL;DR

This paper introduces a model of cultural evolution based on non-stationary stochastic processes, capturing how cultural patterns form, persist, and abruptly change over time, inspired by biological evolution and social copying mechanisms.

Contribution

It develops a novel individual-based model combining biological evolution concepts with cultural copying, demonstrating metastable cultural stages and their abrupt transitions.

Findings

Cultural patterns form and merge over time.

The model reproduces periods of stability and sudden change.

Historical data on car manufacturers supports the model's relevance.

Abstract

We present an individual based model of cultural evolution, where interacting agents are coded by binary strings standing for strategies for action, blueprints for products or attitudes and beliefs. The model is patterned on an established model of biological evolution, the Tangled Nature Model (TNM), where a `tangle' of interactions between agents determines their reproductive success. In addition, our agents also have the ability to copy part of each other's strategy, a feature inspired by the Axelrod model of cultural diversity. Unlike the latter, but similarly to the TNM, the model dynamics goes through a series of metastable stages of increasing length, each characterized by mutually enforcing cultural patterns. These patterns are abruptly replaced by other patterns characteristic of the next metastable period. We analyze the time dependence of the population and diversity in the system, show how different cultures are formed and merge, and how their survival probability lacks, in the model, a finite average life-time. Finally, we use historical data on the number of car manufacturers after the introduction of the automobile to the market, to argue that ourmodel can qualitatively reproduce the flurry of cultural activity which follows a disruptive innovation.