SET-PAiREd: Designing for Parental Involvement in Learning with an AI-Assisted Educational Robot

TL;DR

The paper investigates how to support parental involvement in preschool learning through AI-assisted robots. It introduces SET, a card-based kit to map parental involvement contexts, and PAiREd, an LLM-assisted system that lets parents review and delegate AI-generated learning content to a robot. An in-home study with 20 families and children aged 3–5 reveals how contexts, content perceptions, and collaboration patterns shape effective parent–AI–robot learning, and it yields design implications for safety, control, and pedagogy. The work demonstrates a practical interaction paradigm that enables flexible, context-aware parental involvement in AI-enhanced learning at home.

Abstract

AI-assisted learning companion robots are increasingly used in early education. Many parents express concerns about content appropriateness, while they also value how AI and robots could supplement their limited skill, time, and energy to support their children's learning. We designed a card-based kit, SET, to systematically capture scenarios that have different extents of parental involvement. We developed a prototype interface, PAiREd, with a learning companion robot to deliver LLM-generated educational content that can be reviewed and revised by parents. Parents can flexibly adjust their involvement in the activity by determining what they want the robot to help with. We conducted an in-home field study involving 20 families with children aged 3-5. Our work contributes to an empirical understanding of the level of support parents with different expectations may need from AI and robots and a prototype that demonstrates an innovative interaction paradigm for flexibly including parents in supporting their children.

Figures (10)

• Figure 1: The SET Card-Based Activity Kit. (1) Banners are used to visualize a hierarchical diagram. (2) Experience banks allow participants to generate examples for each factor dimension. (3) Scenario cards guide participants to create real-life scenarios.
• Figure 2: PAiREd System Architecture Overview. The system consists of three key components: prompt instructions and data, the web application (which includes both the editor and activity interfaces), and the robot interaction module.
• Figure 3: Editor Interface of the PAiREd system. This interface allows users to navigate book content, modify LLM-generated learning content through regeneration, or manually edit it.
• Figure 4: Activity Interface of the PAiREd system. This interface enables parents to flexibly adjust their involvement and seamlessly share responsibilities with the robot through the mode-switching and role-delegation mechanisms.
• Figure 5: Mode-Switching and Role Delegation Mechanisms of the PAiREd system. The system offers four modes: robot takeover, robot-led, parent-led, and parent takeover. Parents can drag their icon to the "driver," "co-driver," or "exit" position to select the desired mode. Additionally, they can fine-tune the role delegation by assigning specific tasks to either themselves or the robot.