Skip to main content Skip to navigation

Improving Student-AI Interaction Through Pedagogical Prompting: An Example in Computer Science Education

Project Overview

The document explores the integration of generative AI, particularly large language models (LLMs), into education through a novel approach known as pedagogical prompting. This method encourages students to learn how to interact effectively with AI, thereby enhancing their educational experience, especially in computer science. A study conducted by the authors involved both instructors and students, focusing on the development and evaluation of an intervention aimed at improving students' ability to formulate prompts for LLMs. The findings indicate that this intervention not only improved students' prompt-crafting skills but also led to better learning outcomes and heightened engagement with AI tools. Overall, the document highlights the potential of generative AI in transforming educational practices by equipping students with the skills necessary to leverage AI technology effectively for their learning.

Key Applications

Pedagogical prompting intervention using an interactive system

Context: Early undergraduate computer science education (CS1/CS2), targeting novice students

Implementation: Developed an interactive system for teaching prompt construction through scenario-based learning activities.

Outcomes: Significant improvements in students' ability to write pedagogical prompts, increased confidence in using LLMs for learning, and positive attitudes towards future use of pedagogical prompts.

Challenges: Initial dependency on direct answers from AI, potential cognitive overload from prompt construction, and varied student engagement levels.

Implementation Barriers

Cognitive Barrier

Students may struggle with the cognitive load required to construct effective prompts, leading to frustration and disengagement.

Proposed Solutions: Implement structured scaffolding to guide prompt construction and reduce cognitive load.

Behavioral Barrier

Students often default to seeking direct solutions from AI instead of using it pedagogically.

Proposed Solutions: Educators should emphasize the importance of pedagogical prompting and provide direct instruction on its use.

Project Team

Ruiwei Xiao

Researcher

Xinying Hou

Researcher

Runlong Ye

Researcher

Majeed Kazemitabaar

Researcher

Nicholas Diana

Researcher

Michael Liut

Researcher

John Stamper

Researcher

Contact Information

For information about the paper, please contact the authors.

Authors: Ruiwei Xiao, Xinying Hou, Runlong Ye, Majeed Kazemitabaar, Nicholas Diana, Michael Liut, John Stamper

Source Publication: View Original PaperLink opens in a new window

Project Contact: Dr. Jianhua Yang

LLM Model Version: gpt-4o-mini-2024-07-18

Analysis Provider: Openai

Let us know you agree to cookies