Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Thinking Like a Student: AI-Supported Reflective Planning in a Theory-Intensive Computer Science Course

Noa Izsak

TL;DR

The paper addresses the challenge of effectively supporting students in a theory-intensive CS course by reframing reinforcement sessions through an instructor-facing AI tool. It uses a large language model as a reflective planning partner to simulate a learner, uncover conceptual bottlenecks, and shape a structured, repeatable session format without altering the core syllabus. Evaluation from 24 participants shows increased confidence and perceived understanding, especially in abstract topics like the pumping lemma, supported by qualitative themes of clarity, pacing, and a supportive environment. The study demonstrates that AI-assisted instructional design—when carefully reviewed and integrated by instructors—can enhance pedagogy in theoretically dense domains and be adapted to other demanding CS subjects.

Abstract

In the aftermath of COVID-19, many universities implemented supplementary "reinforcement" roles to support students in demanding courses. Although the name for such roles may differ between institutions, the underlying idea of providing structured supplementary support is common. However, these roles were often poorly defined, lacking structured materials, pedagogical oversight, and integration with the core teaching team. This paper reports on the redesign of reinforcement sessions in a challenging undergraduate course on formal methods and computational models, using a large language model (LLM) as a reflective planning tool. The LLM was prompted to simulate the perspective of a second-year student, enabling the identification of conceptual bottlenecks, gaps in intuition, and likely reasoning breakdowns before classroom delivery. These insights informed a structured, repeatable session format combining targeted review, collaborative examples, independent student work, and guided walkthroughs. Conducted over a single semester, the intervention received positive student feedback, indicating increased confidence, reduced anxiety, and improved clarity, particularly in abstract topics such as the pumping lemma and formal language expressive power comparisons. The findings suggest that reflective, instructor-facing use of LLMs can enhance pedagogical design in theoretically dense domains and may be adaptable to other cognitively demanding computer science courses.

Thinking Like a Student: AI-Supported Reflective Planning in a Theory-Intensive Computer Science Course

TL;DR

The paper addresses the challenge of effectively supporting students in a theory-intensive CS course by reframing reinforcement sessions through an instructor-facing AI tool. It uses a large language model as a reflective planning partner to simulate a learner, uncover conceptual bottlenecks, and shape a structured, repeatable session format without altering the core syllabus. Evaluation from 24 participants shows increased confidence and perceived understanding, especially in abstract topics like the pumping lemma, supported by qualitative themes of clarity, pacing, and a supportive environment. The study demonstrates that AI-assisted instructional design—when carefully reviewed and integrated by instructors—can enhance pedagogy in theoretically dense domains and be adapted to other demanding CS subjects.

Abstract

In the aftermath of COVID-19, many universities implemented supplementary "reinforcement" roles to support students in demanding courses. Although the name for such roles may differ between institutions, the underlying idea of providing structured supplementary support is common. However, these roles were often poorly defined, lacking structured materials, pedagogical oversight, and integration with the core teaching team. This paper reports on the redesign of reinforcement sessions in a challenging undergraduate course on formal methods and computational models, using a large language model (LLM) as a reflective planning tool. The LLM was prompted to simulate the perspective of a second-year student, enabling the identification of conceptual bottlenecks, gaps in intuition, and likely reasoning breakdowns before classroom delivery. These insights informed a structured, repeatable session format combining targeted review, collaborative examples, independent student work, and guided walkthroughs. Conducted over a single semester, the intervention received positive student feedback, indicating increased confidence, reduced anxiety, and improved clarity, particularly in abstract topics such as the pumping lemma and formal language expressive power comparisons. The findings suggest that reflective, instructor-facing use of LLMs can enhance pedagogical design in theoretically dense domains and may be adaptable to other cognitively demanding computer science courses.

Paper Structure

This paper contains 26 sections, 4 figures, 1 table.

Table of Contents

  1. Introduction
  2. Related Work
  3. AI in the Computer Science Curriculum.
  4. LLMs in Computing Education.
  5. Misconception and Conceptual Detection.
  6. Gap and Contribution.
  7. Design of the Intervention
  8. Course Context
  9. Pre-Redesign Challenges
  10. Redesign Principles
  11. Use of ChatGPT as a Reflective Planning Tool
  12. Session Format
  13. First Academic Hour:
  14. Second Academic Hour:
  15. Evaluation
  16. ...and 11 more sections

Figures (4)

  • Figure 1: Flowchart diagram of session structure, indicating via color coding at each step whether the activity was led jointly by students and instructor, exclusively by students, or exclusively by the instructor.
  • Figure 2: Average self-reported confidence before and after AI-informed reinforcement sessions, with standard deviation error bars ($N=24$ maximum). Gains were observed across all topics, particularly in Cardinality and Turing Machines.
  • Figure 3:
  • Figure 4: Would You Recommend These Sessions?