Social Robots for Long-Term Interaction: A Survey

Paper
Design Frameworks

Abstract As the field of HRI evolves, it is important to understand how users interact with robots over long periods. This paper reviews the current research on long-term interaction between users and social robots. We describe the main features of these robots and highlight the main findings of the existing long-term studies. We also present a set of directions for future research and discuss some open issues that should be addressed in this field.

Introduction. Human-Robot Interaction (HRI) is a multidisciplinary field concerned with the “analysis, design, modelling, implementation and evaluation of robots for human use” [18]. While a lot of work has been done in studying how users interact with robots within a single interaction, only in the last decade the first long-term studies, in which the same user (or group of users) interacts with a robot several times, have started to appear. There are several reasons for this. First, longitudinal studies are much more laborious and timeconsuming than short-term studies [20], especially in naturalistic environments. Second, only recently technology has been robust enough to allow for some degree of autonomy when users interact with robots for extended periods of time. Finally, the appearance of the first commercial social robots (e.g., Pleo and Paro) and robots for domestic use such as iRobot’s Roomba, together with demographic trends such as the ageing of the world population, are also fostering research in this area.

Discussion / Conclusion. All the presented studies (see Table 1 for a summary) found positive results regarding the long-term effects of social robots in therapeutic or health-related scenarios. However, the users who took part in these studies were very different (elderly, autistic children and adults), and thus further research is needed to consolidate these results. Moreover,

Synthesis notes from this paper's topics 8

This paper has no direct semantic matches yet — here are notes from the topic(s) it's filed under.

Lines of inquiry this paper opens 24

Research framings built by reading the notes related to this paper — the questions it feeds into.

How should dialogue recommender systems manage conversation history and state? How can AI alignment serve diverse human preferences at scale? How can recommendation systems balance personalization with stability and coverage? How can AI systems learn from failures without cascading errors? How do knowledge graphs enable efficient multi-hop reasoning over alternatives? How should personalization be implemented to improve AI assistant effectiveness? How should conversational agents balance goal-driven initiative with user control? How should dialogue systems best leverage conversation history for retrieval? How do social dynamics and selection effects compound in rating aggregates? How can persona representations reduce language model variance and improve task accuracy? How do language models inherit human biases from training data? What dimensions of recommendation quality do standard metrics miss? How do formal dialogue structures reveal conversation coherence mechanisms? How can we distinguish genuine user preferences from measurement artifacts?