Estimation of Creator Influences Based on Cultural Evolution Models of Color Styles in Painting Arts

Abstract

We study the cultural transmission and evolutionary dynamics of color styles in painting arts. Creative cultures such as visual arts develop through the processes of knowledge transmission and modification, where the structure of knowledge transmission among creators may lead to notable patterns of evolution. Recent research has discovered interesting trends and dynamic patterns in the evolution of painting, but the relationship between these trends and the process of cultural transmission is still poorly understood. Here, we apply a clustering method to analyze the temporal changes in color distribution in oil paintings. As a result, synchronous frequency changes appear in some clusters of color styles. The synchronicity of these frequency changes indicates the existence of creators who produce paintings belonging to a biased set of clusters. Next, we construct a model representing the process of cultural transmission of color styles and demonstrate that an influencer-guided model, where color styles are primarily transmitted from one creator, explains historical data well. Furthermore, using this model, we derive a method for estimating the influence of creators solely from image data. We confirm that creators estimated to have high influence align to a notable extent with those frequently mentioned in the literature on art history.