Högskolan i Skövde

Social robots are designed in manners that encourage users to interact and communicate with them in socially appropriate ways, which implies that these robots should copy many social human behaviors to succeed in social settings. However, this approach has implications for what humans subsequently expect from these robots. There is a mismatch between expected capabilities and actual capabilities of social robots. Expectations of social robots are thus of high relevance for the field of Human-Robot Interaction (HRI). While there is recent interest of expectations in the HRI field there is no widely adapted or well formulated evaluation framework that offers a deeper understanding of how these expectations affect the success of the interaction. With basis in social psychology, user experience, and HRI, we have developed an evaluation framework for studying users’ expectations of social robots. We have identified three main factors of expectations for assessing HRI: affect, cognitive processing, and behavior and performance. In our framework, we propose several data collection techniques and specific metrics for assessing these factors. The framework and its procedure enables analysis of the collected data via triangulation to identify problems and insights, which can grant us a richer understanding of the complex facets of expectations, including if the expectations were confirmed or disconfirmed in the interaction. Ultimately, by gaining a richer understanding of how expectations affect HRI, we can narrow the social robot expectation gap and create more successful interactions between humans and social robots in society.