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

Ananya Bhattacharjee,Yuchen Zeng,Sarah Yi Xu,Dana Kulzhabayeva,Minyi Ma,Rachel Kornfield,Syed Ishtiaque Ahmed,Alex Mariakakis,Mary P Czerwinski,Anastasia Kuzminykh,Michael Liut,Joseph Jay Williams

2023-12-21

Abstract:Traditional interventions for academic procrastination often fail to capture the nuanced, individual-specific factors that underlie them. Large language models (LLMs) hold immense potential for addressing this gap by permitting open-ended inputs, including the ability to customize interventions to individuals' unique needs. However, user expectations and potential limitations of LLMs in this context remain underexplored. To address this, we conducted interviews and focus group discussions with 15 university students and 6 experts, during which a technology probe for generating personalized advice for managing procrastination was presented. Our results highlight the necessity for LLMs to provide structured, deadline-oriented steps and enhanced user support mechanisms. Additionally, our results surface the need for an adaptive approach to questioning based on factors like busyness. These findings offer crucial design implications for the development of LLM-based tools for managing procrastination while cautioning the use of LLMs for therapeutic guidance.

Human-Computer Interaction

What problem does this paper attempt to address?

### Problems the Paper Attempts to Solve This paper aims to explore how large language models (LLMs) can be used to personalize strategies to address academic procrastination. Specifically, the research focuses on the following aspects: 1. **Personalized Strategy Customization**: Understanding how users envision the role of LLMs in managing academic procrastination, particularly how they can provide structured, deadline-based planning advice tailored to individual circumstances. 2. **Challenges and Tensions**: Identifying potential challenges or points of tension in personalizing strategies using LLMs, such as the need for emotional support and factors that need to be considered in tool design. 3. **User Experience and Preferences**: Collecting key preferences and concerns regarding the design of academic procrastination management tools using LLMs through interviews with university students and experts in related fields, including the need for structured guidance and real-life case examples. 4. **Ethical Considerations**: Emphasizing the importance of carefully handling the boundaries of emotional support in developing such tools and ensuring that interventions align with validated strategies or clinical guidelines. In summary, this research aims to explore the potential and limitations of LLMs as auxiliary tools in academic settings, providing design guidance for the future development of more effective procrastination management tools that meet user needs.

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

"It Explains What I am Currently Going Through Perfectly to a Tee": Understanding User Perceptions on LLM-Enhanced Narrative Interventions

Large Language Model Agents for Improving Engagement with Behavior Change Interventions: Application to Digital Mindfulness

Insights from Social Shaping Theory: The Appropriation of Large Language Models in an Undergraduate Programming Course

Towards the Pedagogical Steering of Large Language Models for Tutoring: A Case Study with Modeling Productive Failure

Personalization of Large Language Models: A Survey

When Young Scholars Cooperate with LLMs in Academic Tasks: The Influence of Individual Differences and Task Complexities

Mental-LLM: Leveraging Large Language Models for Mental Health Prediction via Online Text Data

Student Data Paradox and Curious Case of Single Student-Tutor Model: Regressive Side Effects of Training LLMs for Personalized Learning

Impact of Guidance and Interaction Strategies for LLM Use on Learner Performance and Perception

Large Language Model Prompting Techniques for Advancement in Clinical Medicine

Toward Large Language Models as a Therapeutic Tool: Comparing Prompting Techniques to Improve GPT-Delivered Problem-Solving Therapy

Supporting Self-Reflection at Scale with Large Language Models: Insights from Randomized Field Experiments in Classrooms

Personalized Large Language Models

Challenges and Contributing Factors in the Utilization of Large Language Models (LLMs)

To prompt or not to prompt: Navigating the use of large language models for integrating and modeling heterogeneous data

Educational Personalized Learning Path Planning with Large Language Models

An Active Inference Strategy for Prompting Reliable Responses from Large Language Models in Medical Practice

The Dark Patterns of Personalized Persuasion in Large Language Models: Exposing Persuasive Linguistic Features for Big Five Personality Traits in LLMs Responses

Harnessing Large Language Models to Enhance Self-Regulated Learning via Formative Feedback

Apprentices to Research Assistants: Advancing Research with Large Language Models