LADICA: A Large Shared Display Interface for Generative AI Cognitive Assistance in Co-Located Team Collaboration

TL;DR

The paper addresses the challenge of meaningfully augmenting co-located team collaboration with AI by introducing LADICA, an AI-enhanced large shared display interface. It proposes a three-layer cognitive scaffolding—idea repository, affinity lens, and discussion reference—integrated with LLMs to support meta- and macro-cognitive processes during ideation, analysis, and live discussion. Through a formative study and a 14-participant lab user study, the work demonstrates that LADICA can enhance idea generation, mutual awareness, and multi-perspective analysis while highlighting user concerns about AI dominance and information overload. The work advances practical AI-assisted collaboration by preserving human initiative and providing scalable external cognition on shared displays, with implications for co-located teamwork in education and industry settings.

Abstract

Large shared displays, such as digital whiteboards, are useful for supporting co-located team collaborations by helping members perform cognitive tasks such as brainstorming, organizing ideas, and making comparisons. While recent advancement in Large Language Models (LLMs) has catalyzed AI support for these displays, most existing systems either only offer limited capabilities or diminish human control, neglecting the potential benefits of natural group dynamics. Our formative study identified cognitive challenges teams encounter, such as diverse ideation, knowledge sharing, mutual awareness, idea organization, and synchronization of live discussions with the external workspace. In response, we introduce LADICA, a large shared display interface that helps collaborative teams brainstorm, organize, and analyze ideas through multiple analytical lenses, while fostering mutual awareness of ideas and concepts. Furthermore, LADICA facilitates the real-time extraction of key information from verbal discussions and identifies relevant entities. A lab study confirmed LADICA's usability and usefulness.

Figures (11)

• Figure 1: Conceptual framework of LADICA. Teams build external cognition by offloading their ideas and information shared among the team as external representations on the large shared display (left), and these external representations act as scaffolds to support teams' cognitive process, enabling them to discuss and coordinate more effectively (right). During the offloading stage, teams create three layers of representation, idea repository, affinity lens, and discussion reference. Each representation layer scaffolds meta- and marco-cognitive processes in corresponding team activities (group ideation, analysis and discussion-based knowledge-building) respectively.
• Figure 2: The interaction flow of group goal decomposition. The team can activate the feature by pressing button on toolbar. Then they can enter their group goal on query bar and LADICA will suggest sub-goals for group collaboration. The team can then press button to expand the sub-goal into a topic group for group ideation.
• Figure 3: The interaction flow of relation-based expansion of an idea. Users can activate this feature by pressing the button next to an idea note. Then, they can choose a relation type to explore, and LADICA will generate potential thinking aspects associated with the idea based on the selected relation type.
• Figure 4: The interaction flow of query-based expansion of an idea. Users can activate this feature by pressing button next to an idea note. They can then enter a query, and LADICA will provide a list of thinking direction hints based on the user's query. Users can press to automatically revise the idea based on the hint or to add it as a new note for further exploration.
• Figure 5: The interaction flow of discussion-based key information extraction and retrieval. Users can start recording the ongoing discussion by pressing "Start Recording" under the "Speech" menu. The live transcription is displayed simultaneously. By pressing "Get relevant ideas," LADICA will identify existing ideas related to the discussion, where each card on the right showing an existing idea relevant to the ongoing discussion and corresponding transcription; By pressing "Extract key information", LADICA will extract key information from the discussion and also highlight related existing idea notes, as shown in cards on the right in the lower figure.