A study led by Katharina Kühne at the University of Potsdam, recently published in Frontiers in Robotics and AI, explores the complex dynamics of how social robots should communicate, specifically focusing on the use of dialects versus standard language. The research highlights the varying preferences people have regarding the speech patterns of robots designed for human interaction, emphasizing the importance of both trustworthiness and competence in these artificial entities.
Kühne’s research delves into the nuanced aspects of human-robot interaction, examining how the use of a dialect as opposed to a standard language by a robot can influence perceptions of its trustworthiness and competence. The study suggests that while standard language use is often associated with intelligence, employing a dialect, which may be perceived as friendly or familiar, can enhance the comfort level in human-robot interactions. This dichotomy is particularly relevant in different settings – a dialect might be preferred for creating a sense of connection in casual or intimate settings, like elderly homes, whereas standard language might be more appropriate in formal or service-oriented environments.
Credit: University of Potsdam
To empirically test these hypotheses, the study involved 120 participants from Berlin and Brandenburg. They were exposed to videos where a robot with a male human voice spoke either in standard German or the Berlin dialect, a variant perceived as working-class and often used in media for a casual, informal impression. The participants, diverse in demographics, assessed the robot’s trustworthiness and competence. The study also recorded the type of device used by participants to watch these videos, considering the potential impact of the viewing medium on their perceptions.
The findings indicated a correlation between perceived competence and trustworthiness, with a general preference for the robot speaking standard German. However, participants who were more familiar and comfortable with the Berlin dialect tended to favor the robot using the dialect. Intriguingly, the research found that the type of device used to view the videos influenced perceptions, with those using mobile devices like phones or tablets rating the standard German-speaking robot lower, possibly due to increased distractions and cognitive load.
The study opens avenues for further research into the role of cognitive load in human-robot interactions and the potential benefits of dialect use in specific contexts. The findings also underscore the importance of in-group identity and the perceived prestige of dialects in shaping preferences for robot communication styles. The research team plans to extend their work to more real-life scenarios to deepen the understanding of these dynamics. This study contributes significantly to the field of social robotics, offering insights into how robotic communication can be optimized for different social contexts and individual preferences.