The rapid evolution of contemporary society is exacerbating complexity and uncertainty, precipitating challenges across all sectors. In the realm of education, a critical issue is the growing prevalence of multi-layered and ill-structured problems tha...
The rapid evolution of contemporary society is exacerbating complexity and uncertainty, precipitating challenges across all sectors. In the realm of education, a critical issue is the growing prevalence of multi-layered and ill-structured problems that defy resolution through traditional, fragmented subject-matter knowledge alone. Addressing these evolving challenges necessitates a paradigm shift from knowledge-centered instruction to competency-based education grounded in real-world problem-solving. This imperative has driven the emergence of competency-based approaches, which prioritize problem-solving in authentic contexts over the passive acquisition of knowledge, thereby fundamentally reshaping educational methodologies. In particular, the advancement of Artificial Intelligence (AI), a core driver of the information society, is redefining the mechanisms of knowledge generation and problem resolution, underscoring the urgent need for competencies that enable problem-solving through human-AI convergence.
Consequently, contemporary education demands novel pedagogical paradigms that transcend traditional disciplinary boundaries to integrate multidisciplinary domains with AI. However, educational settings are currently hindered by inconsistent conceptual definitions of AI integrated education. Furthermore, instructors encounter significant challenges in implementation due to a paucity of concrete methodologies for infusing AI into various subjects. To resolve these practical impediments, there is an urgent imperative to develop a systematic instructional design model. Such a model is essential for guiding the formulation of learning objectives, the curation of educational content, and the execution of structured teaching-learning procedures.
The purpose of this study is to develop a model for designing AI integrated instruction between subject matters and AI to develop problem-solving competency as an educational demand of a changing society, and to verify its validity. To achieve this, the research questions are as follows: First, how is the AI integrated instructional design models for problem-solving competency? Second, what are the design principles of the AI integrated instructional design model for problem-solving competency? Third, are the developed model and principles valid?
This study adopted the Design and Development Research methodology, structured into four distinct phases. In the first phase, instructional components were identified by reviewing relevant literature and analyzing current cases of AI integrated instruction. In the second phase, a draft of the AI integrated Instructional Systems Design (ISD) model and its guiding principles was developed based on a conceptual framework derived from these components. In the third phase, the model's internal validity was secured through two rounds of expert reviews and a usability evaluation involving field instructors. Finally, in the fourth phase, a field evaluation was performed to verify external validity and to confirm the model's practical applicability and effectiveness in educational environments.
The findings of this study are summarized as follows. First, the AI integrated instructional design model aimed at enhancing problem-solving competency is composed of two distinct frameworks: a conceptual model and a procedural model. The conceptual model identifies learning content, learning objectives, and learning activities as its core components. Learning content is organized along a continuum based on the degree of structure, ranging from structured to ill-structured problems. Learning objectives consist of knowledge understanding, knowledge application, and knowledge creation, where 'knowledge' encompasses both subject matter expertise and AI-related knowledge. Learning activities are structured into stages: understanding and defining the problem, analyzing the problem, exploring the problem, solving the problem, and reflection and evaluation. Furthermore, the procedural model—specifically the AI integrated Instructional Systems Design (ISD) model—was designed considering the unique characteristics and applicability of AI integrated education. It comprises the following phases: Analysis, Design, Prototype Development and Usability Testing, Development, Implementation, and Evaluation.
Second, the design principles for this model consist of nine guiding principles: (1) Principle of Alignment among Objectives, Content, Methods, and Evaluation; (2) Principle of Interdisciplinary Content Organization; (3) Principle of AI Integration; (4) Principle of Questioning and Problem Generation; (5) Principle of Problem-Based Learning Design; (6) Principle of Facilitating Thought Expansion; (7) Principle of Inducing Critical Thinking; (8) Principle of Evaluation and Reflection; and (9) Principle of Long-term Practice.
Third, the validity of the developed model and principles was rigorously established. Internal validity was verified through two rounds of expert reviews and one usability evaluation, while external validity was confirmed via field evaluation.
Theoretically, this study is significant as it expands the paradigm of integration beyond traditional interdisciplinary boundaries to encompass AI, while also enriching the limited body of Design and Development Research with empirical cases of integrated education. From a practical standpoint, it offers a robust model and guiding principles that empower field instructors to mitigate conceptual ambiguity, thereby facilitating the development and implementation of systematic and effective AI integrated instruction.