Language and Robots: A Cross-Cultural Perspective on Human-Robot Interaction Between Monolingual and Bilingual Speakers

Abstract

Robots, supported by advanced human-robot interaction (HRI) platforms, are set to assume an increasingly vital role in society. Nevertheless, the design of these robots necessitates a thorough consideration of cultural and linguistic differences inherent in individual preferences and lifestyles. Prior research has delineated different interaction patterns between monolinguals and bilinguals, with the latter's ability to switch languages potentially affecting their interaction with AI. On the other hand, bilinguals have the ability to switch between two languages, which can enhance cognitive flexibility and potentially influence their interaction with AI. This research explores whether bilingual speakers communicate differently in human-robot interactions depending on the language they use. The hypothesis posits that the language used by the bilingual speaker determines how the conversation is conducted. Language is an obvious indicator of cultural differences, so this study conducted a 30-minute game session with avatar manipulation for 10 Korean, 10 English, and 30 bilingual participants. Data regarding speech and in-game behaviors were analyzed through utterance counts, interruption frequencies, and sentiment analysis to compare and linguistic variations. The findings from this study offer valuable insights for the AI industry in the development of new guidelines and highlight the importance of linguistic differences in shaping interactions between humans and artificial agents in cooperative environments.