BrickSmart: Leveraging Generative AI to Support Children's Spatial Language Learning in Family Block Play

TL;DR

BrickSmart introduces a GenAI-driven framework to empower parents to guide children's spatial language development during block play. Through a structured three-step workflow—Discovery & Design, Build & Learn, Explore & Expand—the system provides personalized building instructions, adaptive vocabulary guidance, progress tracking, and real-time parental prompts. In a comparative study with 24 parent–child pairs, BrickSmart yielded significant improvements in children's spatial vocabulary usage, learning gains, and engagement, while achieving high usability scores (SUS ≈ 71.5) and positive parental feedback. The work demonstrates the viability of AI-assisted guided play to augment early spatial cognition and offers design insights for future AI-enabled educational tools that balance automation with human agency.

Abstract

Block-building activities are crucial for developing children's spatial reasoning and mathematical skills, yet parents often lack the expertise to guide these activities effectively. BrickSmart, a pioneering system, addresses this gap by providing spatial language guidance through a structured three-step process: Discovery & Design, Build & Learn, and Explore & Expand. This system uniquely supports parents in 1) generating personalized block-building instructions, 2) guiding parents to teach spatial language during building and interactive play, and 3) tracking children's learning progress, altogether enhancing children's engagement and cognitive development. In a comparative study involving 12 parent-child pairs children aged 6-8 years) for both experimental and control groups, BrickSmart demonstrated improvements in supportiveness, efficiency, and innovation, with a significant increase in children's use of spatial vocabularies during block play, thereby offering an effective framework for fostering spatial language skills in children.

Figures (10)

• Figure 1: Workflow of BrickSmart. Step 1. Discover & Design: Children describe their desired scene using voice input, and BrickSmart assists parents to help refine these ideas. The model preview appears on the left. Step 2. Build & Learn: Parents and children construct the model following the instructions of BrickSmart. Parents are advised to incorporate spatial vocabulary during the building process and track the children's learning progress. Step 3. Explore & Expand: The assembled models are used for interactive play, where parents introduce more spatial vocabularies to the children as guided by BrickSmart.
• Figure 2: Overview of the personalized building instruction generation pipeline. Including 3D model generation, voxelization, optimization, and generation of formatted instructions.
• Figure 3: Children's Block Designs. A collection of diverse and creative block constructions by children using the BrickSmart system.
• Figure 4: Comparison of UEQ metrics between experimental and control groups. Error bars represent 95% confidence intervals (CI). $\ ^*$ stands for $p < 0.05$ and $\ ^{**}$ stands for $p < 0.01$. The same annotation applies to the rest of the paper.
• Figure 5: Comparison of NASA-TLX metrics between experimental and control groups. $\ ^*$ stands for $p < 0.05$