Cultural Differences in an Artificial Representation of the Human Emotional Brain System: A Deep Learning Study

Abstract

Cultural differences between countries are customarily calculated by comparing mean scores on multiple dimensions, which ignores heterogeneity within societies. This article proposes a fundamentally new way of measuring cultural differences by evaluating the functioning of the emotional brain—that is, by comparing how people think, rather than what they think. Deep learning as a supervised machine learning technique is used to mathematically represent the emotional brain of the Volksgeist (“folk spirit”) with a complex system of nonlinear combinations of value priorities, opinions, and other factors with subjective feelings of well-being. Cross-validating how well one Volksgeist's artificial brain fits another country provides a novel measure for emotional cultural differences, which also considers heterogeneity within a society. This measure is conceptually and statistically unrelated to the Kogut–Singh index calculated from mean scores on cultural dimensions. However, it shows good face validity and supports international marketing and business theory as a negative and highly significant predictor for bilateral trade flows within Europe, using a classic gravity model.