Anticipating User Needs: Insights from Design Fiction on Conversational Agents for Computational Thinking

TL;DR

This paper investigates how to design conversational agents that scaffold computational thinking in introductory programming, addressing the gap where genAI tutoring tools focus on professional developers. Using Design Fiction with nine CS instructors, it elicits concrete expectations and concerns about an adaptive tutor named Atlas that can tailor guidance to a learner's background, provide multi-modal explanations, and collect useful student telemetry. The findings reveal five key capability areas—Programming Guidance, Code Elucidation, Student Telemetry, Course Administration, and UI/UX—and highlight benefits such as engagement and reduced isolation, alongside risks like mislearning and weakened instructor-student ties; credibility and customization emerge as critical criteria. The work offers design guidelines and future research directions, including student-centered studies and prototype development, to advance pedagogy-oriented AI tutors for computational thinking and programming education.

Abstract

Computational thinking, and by extension, computer programming, is notoriously challenging to learn. Conversational agents and generative artificial intelligence (genAI) have the potential to facilitate this learning process by offering personalized guidance, interactive learning experiences, and code generation. However, current genAI-based chatbots focus on professional developers and may not adequately consider educational needs. Involving educators in conceiving educational tools is critical for ensuring usefulness and usability. We enlisted nine instructors to engage in design fiction sessions in which we elicited abilities such a conversational agent supported by genAI should display. Participants envisioned a conversational agent that guides students stepwise through exercises, tuning its method of guidance with an awareness of the educational background, skills and deficits, and learning preferences. The insights obtained in this paper can guide future implementations of tutoring conversational agents oriented toward teaching computational thinking and computer programming.

Figures (2)

• Figure 1: Our research method consists of five main steps, with the outcome of one step being the input to the next.
• Figure 2: Categorization of expectations narrative data