Extracting hierarchical features of cultural variation using network-based clustering

Area of Science:

• Cultural evolution studies
• Network analysis
• Computational social science

Background:

• Cultural variation arises from hierarchical descent processes, including population structure and trait transmission.
• Traditional methods may not fully capture the complexities of hierarchically structured cultural data.
• Understanding these structures is key to uncovering factors driving cultural evolution.

Purpose of the Study:

• To adapt a network-based hierarchical clustering method for analyzing cultural variation.
• To demonstrate the method's utility in uncovering insights into cultural evolution.
• To identify geographic and other influences on cultural patterns.

Main Methods:

• Adaptation of a network-based hierarchical clustering algorithm.
• Construction of similarity networks to depict community structures.
• Application to diverse datasets: regional pronunciation, global folklore, global phonemics, and temporal first names.

Main Results:

• The method successfully identified community structures within cultural datasets.
• Insights were gained into geographic influences on pronunciation and phonemic variation.
• Patterns in folklore and first name variation were illuminated, highlighting temporal and other factors.

Conclusions:

• Network-based hierarchical clustering is a valuable tool for analyzing cultural variation.
• The approach effectively reveals underlying structures and influences on cultural evolution.
• This method offers a novel perspective for understanding the dynamics of cultural change.