Contrasting Cognitive Styles in Vision-Language Models: Holistic Attention in Japanese Versus Analytical Focus in English

TL;DR

This study explores how vision-language models trained on Japanese and English data reflect culturally influenced visual processing styles, revealing that models internalize and reproduce cultural perceptual differences.

Contribution

It demonstrates that VLMs trained on different languages exhibit culturally grounded attentional patterns, linking language, culture, and visual cognition in AI models.

Findings

Japanese-trained models show holistic attention patterns.

English-trained models focus on individual objects.

Models reflect cultural perceptual differences embedded in training data.

Abstract

Cross-cultural research in perception and cognition has shown that individuals from different cultural backgrounds process visual information in distinct ways. East Asians, for example, tend to adopt a holistic perspective, attending to contextual relationships, whereas Westerners often employ an analytical approach, focusing on individual objects and their attributes. In this study, we investigate whether Vision-Language Models (VLMs) trained predominantly on different languages, specifically Japanese and English, exhibit similar culturally grounded attentional patterns. Using comparative analysis of image descriptions, we examine whether these models reflect differences in holistic versus analytic tendencies. Our findings suggest that VLMs not only internalize the structural properties of language but also reproduce cultural behaviors embedded in the training data, indicating that cultural cognition may implicitly shape model outputs.