Scaffolding Metacognition with GenAI: Exploring Design Opportunities to Support Task Management for University Students with ADHD

TL;DR

This paper investigates how Generative AI (GenAI) can scaffold metacognition to aid university students with ADHD in academic task management. Through 20 co-designed sessions with students and interviews with 5 ADHD intervention experts, the study identifies four core metacognitive challenges (awareness, initiation, attention, and emotion) and articulates three design directions: cognitive scaffolding to enhance task/self-awareness, reflective task execution to develop metacognitive abilities, and emotional regulation to sustain engagement. It highlights key opportunities and risks of GenAI, emphasizing a balanced, reflective role for AI that supports but does not automate or enable dependency. The findings offer practical design implications and a research agenda for neurodivergent populations, with considerations for privacy, user agency, and adaptive data use. Overall, GenAI can meaningfully support metacognition in ADHD task management when designed as a collaborative, self-regulation-promoting partner rather than a full automation substitute.

Abstract

For university students transitioning to an independent and flexible lifestyle, having ADHD poses multiple challenges to their academic task management, which are closely tied to their metacognitive struggles--difficulties in awareness and regulation of one's own thinking processes. The recently surged Generative AI shows promise to mitigate these gaps with its advanced information understanding and generation capabilities. As an exploratory step, we conducted co-design sessions with 20 university students diagnosed with ADHD, followed by interviews with five experts specialized in ADHD intervention. Adopting a metacognitive lens, we examined participants' ideas on GenAI-based task management support and experts' assessments, which led to three design directions: providing cognitive scaffolding to enhance task and self-awareness, promoting reflective task execution for building metacognitive abilities, and facilitating emotional regulation to sustain task engagement. Drawing on these findings, we discuss opportunities for GenAI to support the metacognitive needs of neurodivergent populations, offering future directions for both research and practice.

Figures (13)

• Figure 1: Overview of the individual co-design workshop with university students with ADHD (top) and interviews with ADHD experts (bottom). The individual co-design workshop process included (a) the introduction of the workshop process, (b) an interview about the daily academic task management challenges faced by participants, and (c) the co-design of GenAI solutions with the participants on an online collaborative whiteboard. The interview process with ADHD experts included (d) the introduction of the interview process, (e) discussions on the daily task management challenges faced by ADHD university students based on the experts' intervention experience, and the presentation of design ideas from student participants to seek advice and identify potential risks. For ease of reading, the original text in the screenshots has been translated into English and visually enhanced (e.g., larger font size and boldface). The original images are available in the Appendix.
• Figure 2: Overview of the metacognitive challenges faced by university students with ADHD across different stages of task management (Organization, Execution, Adjustment), and the opportunities for GenAI to provide support. The metacognitive challenges include a lack of task awareness (source, time, and prioritization) and self-awareness (cognitive ability) during task organization, limited metacognitive monitoring and control (task initiation, attention control, progress tracking, and task adjustment), and difficulties with emotion regulation during task execution. Participant design ideas suggest GenAI could support students in navigating these challenges by scaffolding task awareness and self-awareness, prompting reflection to strengthen monitoring and control, and facilitating productivity-oriented emotional regulation.
• Figure 3: Examples of participants' design ideas for leveraging GenAI to enhance task and self-awareness: (a) identifying tasks from chat logs and the clipboard (P2); (b) conducting priority analysis and generating task lists that can be manually adjusted later (P7). For ease of reading, the original design ideas presented in Chinese have been translated into English. The original images are available in the Appendix.
• Figure 4: Examples of participants' design ideas for leveraging GenAI to foster metacognitive abilities: (a) breaking down tasks into actionable subtasks and estimating workload based on conventional benchmarks and individuals' historical behavior (P15); (b) delivering multiple forms of motivational prompts (e.g., visual, audio, or vibration) to facilitate task initiation (P16); and (c) tracking the gap between actual vs. planned reading progress and suggesting how much work is required to achieve the goal (P11). For ease of reading, the original design ideas presented in Chinese have been translated into English. The original images are available in the Appendix.
• Figure 5: Examples of participants' design ideas for leveraging GenAI to facilitate emotional regulation for sustained task engagement: (a) serving as a study companion to help individuals stay focused (P7); and (b) providing emotional support through interactive, multimodal conversations with one's favorite fictional character as part of a reward mechanism (P10). For ease of reading, the original design ideas presented in Chinese have been translated into English. The original images are available in the Appendix.