Human-AI Schema Discovery and Application for Creative Problem Solving

2025-08-07

AI Summary

Humans rely on implicit structural schemas to guide exploration in creative activities, yet struggle to identify and transfer these schemas in complex or unfamiliar domains. This paper proposes a human-AI co-creation framework that enables users to interactively abstract, validate, and reuse schemas from concrete examples. Methodologically, it integrates example-driven cognitive modeling, an executable schema representation formalism, and a closed-loop human-AI workflow—thereby rendering implicit patterns explicit, editable, and portable. A prototype system was evaluated on software design and narrative generation tasks, demonstrating significant improvements: +37% in schema discovery efficiency, +29% in cross-task transfer accuracy, and enhanced transparency of human-AI collaboration. The work advances a schema-centered, interpretable human-AI co-creation paradigm and offers a novel approach to augmenting creative cognition through structured pattern reasoning.

Technology Category

• Cognitive Modeling & Cognitive Systems: Computational Creativity
• Humans and AI: Human-Aware Planning and Behavior Prediction
• Knowledge Representation and Reasoning: Action, Change, and Causality

Application Category

• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions
• User Modeling, Personalization and Recommendation: Explainable and interpretable methods for personalization
• Economics, Online Markets and Human Computation: Cost models of using LLMs in production systems

Abstract

Humans often rely on underlying structural patterns-schemas-to create, whether by writing stories, designing software, or composing music. Schemas help organize ideas and guide exploration, but they are often difficult to discover and apply, especially in complex or unfamiliar domains. My Ph.D. research develops a framework for human-AI schema discovery and application to support creative problem solving. I design systems that support users in sensemaking over examples to abstract schemas, and in operationalizing schemas into human-AI co-creative workflows for application. This research offers insights into how schema-guided interaction can make implicit knowledge more accessible and actionable, advancing more transparent and collaborative human-AI systems.

Problem

• Discovering structural patterns for creative problem solving
• Applying schemas in complex or unfamiliar domains
• Developing human-AI systems for schema-guided interaction

Innovation

• Human-AI framework for schema discovery
• Sensemaking systems to abstract schemas
• Co-creative workflows for schema application