We present the Robot Bookworm, a multi-session intervention co-designed with children and educators to foster reading motivation through personalized book discussions. The robot assigned each child a personally fitting book and engaged them in pedagogically structured discussions, with personalized book-aligned dialogic content selectively generated offline by a language model and moderated by people to ensure safety. We compared a personalized book discussion condition with a book-neutral control in a four-session, large-scale user study in two primary schools (N = 101, 8-11 y.o.). The intervention significantly increased reader-book relatedness and reading enjoyment, particularly for children with below-ceiling baseline enjoyment, but had no effect on intrinsic motivation. At a one-year follow-up, the quantitative effects were not sustained. However, children reported perceived positive shifts in attitudes towards reading, which they attributed to the Robot Bookworm.

This paper presents ongoing doctoral research on Dialogic Educational Child-Robot Interaction (DECRI), focused on designing child-centered hybrid dialogue systems that combine dialogic pedagogical structure with the adaptive potential of LLMs.We propose a generalisable framework that facilitates personalized educational dialogues, and validate its feasibility in a study (N=101) aimed at enhancing reading motivation.The framework integrates pedagogically grounded, co-designed dialogue strategies with LLM capabilities to support meaningful personalization.Future directions include expanding to small-group settings, leveraging local LLMs for enhanced personalization, co-design, and early user model development, and developing AI-based validation tools.The overarching goal is to responsibly design DECRI tailored to each child, while maintaining pedagogical and developmental integrity.

In this paper, we present a participatory design approach to age-based personalization for child-robot interaction. This is an important step towards social robots being effective across age groups. As a testbed for our approach, we used humor. Personalized humor is a powerful social motivator and is uniquely suited to build relatable and sustained child-robot interactions. Through a series of co-design workshops (n = 102 children), we identified humor concepts that fit the specific sense of humor for each of the four age groups (8–9, 9–10, 10–11, 11–12 y.o.), as well as humor concepts that resonated across these age groups. A user study showed that, overall, children found the interaction more amusing and a better fit for both their own sense of humor and that of their peer group when the robot used age-personalized humor compared to age-agnostic humor. The strength of the effects varied by age group, with the oldest group consistently scoring lower on the outcome measures, indicating that the design was not equally effective for all groups.