Understanding the Role of Large Language Models in Personalizing and Scaffolding Strategies to Combat Academic...

Area of Science:

• Psychology
• Human-Computer Interaction
• Artificial Intelligence

Background:

• Traditional academic procrastination interventions lack individualization.
• Large language models (LLMs) offer potential for personalized support.
• User expectations and LLM limitations in this context require investigation.

Purpose of the Study:

• To explore user needs and expectations for LLM-based academic procrastination interventions.
• To identify design implications for developing effective LLM tools for procrastination management.

Main Methods:

• Conducted interviews and focus group discussions with university students and experts.
• Utilized a technology probe to present personalized procrastination advice generation.
• Gathered qualitative feedback on user experience and desired features.

Main Results:

• LLMs should provide structured, deadline-oriented steps for managing procrastination.
• Enhanced user support mechanisms are crucial for LLM-based interventions.
• Adaptive questioning, considering user busyness, is necessary for personalized advice.

Conclusions:

• LLM-based tools require specific design features for effective academic procrastination management.
• Findings inform the development of personalized digital interventions.
• Caution is advised regarding the use of LLMs for therapeutic guidance in this domain.