Cross-cultural evaluation of taste-sound correspondences in AI-generated music

2026-08-04

AI Summary

This study investigates cross-cultural consistency and perceptual mechanisms underlying taste–sound correspondences in AI-generated music. Conducting online experiments in Argentina, Italy, and Japan, the authors employed a fine-tuned MusicGen model to generate musical excerpts prompted by four basic tastes—sweet, sour, bitter, and salty—and analyzed responses using standardized ratings and exploratory factor analysis, enabling the first multi-national comparison of its kind. Results reveal that salty taste mappings exhibit the weakest cross-cultural consistency. Between-country rating differences are primarily attributable to scale-use tendencies (i.e., response style bias) rather than genuine shifts in perceptual structure, although latent perceptual dimensions show cultural specificity. The findings underscore the necessity of disentangling response biases from true perceptual reconfiguration and demonstrate that the fine-tuned model is preferred more strongly in Argentina and Italy.

Abstract

Sonic seasoning research has shown that listeners attribute systematic gustatory and emotional meaning to sound, and text-to-music generative artificial intelligence has recently been used to render gustatory prompts as musical stimuli. Whether the taste-sound correspondences acquired by such models hold beyond the cultural context in which they were validated remains untested. We extended a single-country study to a three-country online experiment conducted in Argentina, Italy, and Japan (N = 361). Participants first indicated their preference between base and fine-tuned MusicGen excerpts generated from four taste prompts (sweet, sour, bitter, salty), and then rated fine-tuned excerpts on twelve taste, emotion, and thermal descriptors. Preference for the fine-tuned model was confirmed in Argentina and Italy but not in Japan, and the salty prompt yielded the weakest correspondence in all three cohorts. Ratings differed substantially between countries, yet the main effect of country was no longer detectable once ratings had been standardized within participant, whereas the interactions characterizing the mapping of prompts onto descriptors remained essentially unchanged. Much of the apparent cross-cultural divergence is therefore attributable to differences in scale use; a structural component nevertheless persists. In addition an exploratory factor analysis indicated that the twelve descriptors were organized along different latent dimensions in each cohort. These results indicate that cross-cultural variation in AI-mediated sonic seasoning operates at two levels: the overall level at which taste is attributed to a given stimulus, and the relational structure of those attributions. Evaluations of generative music systems across populations should accordingly distinguish response-style bias from genuine perceptual reorganization.

Problem

• cross-cultural
• taste-sound correspondences
• AI-generated music
• sonic seasoning
• cultural variation

Innovation

• cross-cultural evaluation
• generative AI music
• taste-sound correspondence
• sonic seasoning
• response-style bias