Visual and Auditory Aesthetic Preferences Across Cultures

TL;DR

The paper addresses whether aesthetic preferences are universal or culturally learned across five modalities. It adopts a large-scale online approach with continuous $2$-D stimulus spaces to systematically sample shape, curvature, colour, harmony, and melody across 10 countries. The study provides evidence of universal patterns (e.g., shape symmetry, curvature tendencies, interval-based harmony) alongside modality-specific cultural variation (notably melody and some colour relations), highlighting a blend of shared perceptual organization and cultural learning. These findings advance understanding of how perceptual mechanisms interact with cultural experience to shape aesthetic judgments, with implications for design, education, and cross-cultural research.

Abstract

Research on how humans perceive aesthetics in shapes, colours, and music has predominantly focused on Western populations, limiting our understanding of how cultural environments shape aesthetic preferences. We present a large-scale cross-cultural study examining aesthetic preferences across five distinct modalities extensively explored in the literature: shape, curvature, colour, musical harmony and melody. We gather 401,403 preference judgements from 4,835 participants across 10 countries, systematically sampling two-dimensional parameter spaces for each modality. The findings reveal both universal patterns and cultural variations. Preferences for shape and curvature cross-culturally demonstrate a consistent preference for symmetrical forms. While colour preferences are categorically consistent, ratio-like preferences vary across cultures. Musical harmony shows strong agreement in interval relationships despite differing regions of preference within the broad frequency spectrum, while melody shows the highest cross-cultural variation. These results suggest that aesthetic preferences emerge from an interplay between shared perceptual mechanisms and cultural learning.

Figures (4)

• Figure 1: Schematics of experimental design. (A) In independent experiments for each modality, participants were asked to rate how much they like the seen or heard stimulus on a 7-point scale from 1 being "not at all" to 7 "very much", translated into their own language. Each presented stimulus was defined at random by sampling two points from the general space (see 'Defining Stimulus Space' in Methods). (B) Participants were recruited from 10 countries, including all continents which are coloured according to regions defined by WorldBank (www.worldbank.org).
• Figure 2: Preferred regions in modality spaces. (A) Using a fixed bandwidth to smooth the preference ratings, the most preferred regions in each modality space are highlighted in yellow. White diagonal lines indicate where values above and below are equal but in differing two parameter orders. (B) Cross-cultural similarity in regions of preference and their variability. Jensen-Shannon distance was used to measure the similarity between country-level matrices and dimensionality reduction was performed using UMAP. Countries positioned closer together share similar regions of preference. Insets show examples from different coordinates of these UMAPs to illustrate variations.
• Figure 3: Relational preference across modalities is assessed as follows: Shape = width-to-height aspect ratios; Curvature = the difference between control points $P1$ and $P2$; Colour = absolute difference in degrees between paired hues; Harmony and melody = pitch intervals between tone pairs in semitones. Each line represents a GAM-fitted curve per country, with colours denoting world regions.
• Figure 4: Agreement and disagreement between cultures. Between-country correlations in (A) preferred regions in modality spaces, and (B) relational preferences across modalities. Below each of these, we report the reliability using split-half correlations. Error bars indicate 95% CI.