Social contexts shape the development of attention; however, little is known about joint attention beyond infancy. This study employed behavioral and eye-tracking measurements to investigate cultural variations in how caregivers direct 3- to 4-year-old children's attention and subsequent changes in children's attention to objects and contextual backgrounds in the United States (predominantly non-Hispanic Whites) and Japan (N = 60 mother-child dyads, 29 girls, 31 boys). The findings revealed that caregivers directed children's attention to culturally sensitive information, and significant cross-cultural differences in attention emerged after caregiver-child interaction, with Japanese children shifting their attention to the backgrounds. Results provide new insights into the role of social interaction and cultural diversity in the development of attention.