Self-Regulated Personal Contracts as a Harm Reduction Approach to Generative AI in Undergraduate Programming Education

Aadarsh Padiyath; Barbara Ericson; Jessica Shen

arxiv: 2604.03256 · v1 · submitted 2026-03-11 · 💻 cs.CY

Self-Regulated Personal Contracts as a Harm Reduction Approach to Generative AI in Undergraduate Programming Education

Aadarsh Padiyath , Jessica Shen , Barbara Ericson This is my paper

Pith reviewed 2026-05-15 12:33 UTC · model grok-4.3

classification 💻 cs.CY

keywords generative AIprogramming educationself-regulated learningharm reductionstudent agencyChatGPTundergraduate courses

0 comments

The pith

A non-binding personal contract prompts intentional GenAI use in programming classes but often fails to sustain it under pressure.

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The paper examines whether undergraduates can exercise explicit agency over generative AI tools like ChatGPT by drafting their own non-binding guidelines tied to personal learning goals. Students in an intermediate Python course created usage rules at the semester start and reflected on them at several points over eleven weeks. Fifty-eight percent reported the exercise altered their thinking and supplied useful accountability. Yet many who endorsed their own guidelines later abandoned them when deadlines or convenience intervened. The central finding is that turning implicit choices explicit is feasible, but sustaining the rules demands repeated self-control that free access to the tools does not.

Core claim

The GenAI Contract intervention, grounded in harm reduction and self-regulated learning theory, led 58% of 217 students to report changed thinking and created helpful accountability structures. Awareness did not reliably produce sustained behavior change, because maintaining self-set guidelines required constant self-control across hundreds of individual decisions while using GenAI freely required none.

What carries the argument

The non-binding GenAI Contract: students articulate personal learning goals, write usage guidelines, and reflect on alignment at strategic points across the semester, graded only for completion.

If this is right

Students gain explicit awareness of how their learning goals relate to GenAI use.
The contract creates accountability structures that help some students resist deadline-driven shortcuts.
Behavior change remains inconsistent when external pressures outweigh the cost of self-regulation.
The approach is most effective for students still forming their relationship with the tools.
Grading solely for completion avoids enforcement conflicts while still prompting reflection.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

Similar contracts could be tested in other courses where GenAI tools create comparable tensions between convenience and learning goals.
Future versions might lower the self-control burden by adding optional reminders or brief peer check-ins at reflection points.
Repeated use of such contracts across multiple semesters could reveal whether the practice builds durable habits or increases decision fatigue.
Departments considering wider adoption would need to weigh voluntary participation against the risk that required versions lose student buy-in.

Load-bearing premise

Students can maintain enough ongoing self-control to follow their self-set guidelines across hundreds of decisions even when deadlines and effortless GenAI availability exert strong counter-pressure.

What would settle it

A follow-up study that logs actual student interactions with GenAI tools and compares them against each student's stated guidelines to count adherence rates under real course deadlines.

Figures

Figures reproduced from arXiv: 2604.03256 by Aadarsh Padiyath, Barbara Ericson, Jessica Shen.

**Figure 2.** Figure 2: The GenAI Contract template distributed to students across two time points: Step 3 during Week 5, post-Midterm 1; [PITH_FULL_IMAGE:figures/full_fig_p005_2.png] view at source ↗

read the original abstract

Students learning programming exercise agency in deciding when and how to use GenAI tools like ChatGPT. However, this agency is often implicit and shaped by deadline pressure and peer behavior rather than explicit and conscious learning goals. We designed a GenAI Contract grounded in harm reduction and self-regulated learning theory to scaffold intentional decision-making: students articulated personal learning goals, created usage guidelines, and reflected on alignment at strategic points across an eleven-week semester. The contract was non-binding and graded only for completion, emphasizing self-awareness over enforcement. We implemented this with N=217 students in an intermediate Python course. For students still forming their relationship with GenAI, it worked, as 58% of students reported the intervention changing their thinking and created helpful accountability structures. However, awareness did not always translate to sustained behavior change. Some students who valued their guidelines still abandoned them under various pressures. Maintaining guidelines required constant self-control across hundreds of decisions, while using GenAI freely requires none. Many students could not sustain this burden despite this self-awareness. We discuss supporting student agency when GenAI tools and learning goals create tension.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Desk Editor's Note private letter to a colleague

The paper tests a non-binding GenAI contract in a real undergrad Python class and reports honest mixed results on self-reported thinking changes, but lacks controls or baselines to support causal claims.

read the letter

The one thing to know is that this paper puts a non-binding personal contract into a real undergraduate programming course to help students manage their GenAI use, and it finds that while some students report it helped them think more carefully, many still dropped their own guidelines when deadlines arrived. The new part is the specific setup: students created their own learning goals and usage rules at the start, then reflected on how well they followed them at points during the eleven-week semester. The contract was only graded for completion, not for sticking to it, which fits the harm reduction angle. They ran this with 217 students and got back that 58% felt it changed their thinking and provided useful accountability. The authors are straightforward about the limits too, noting that self-awareness did not always produce sustained behavior change because following self-set rules takes constant effort while using the tools does not. That honesty is what the paper does well. It avoids overclaiming success and points out the practical difficulty of self-control across many small decisions. The soft spots are around the evidence. There is no control group or pre-intervention baseline, so the 58% change in thinking cannot be clearly tied to the contract itself rather than other things happening in the course or students' lives. The abstract gives no information on the survey design, response rate, or any statistical checks, which leaves the numbers hard to evaluate. The stress-test note is right on this: without those comparisons, the positive claims stay descriptive. This paper is for computer science educators who are dealing with GenAI in their classes and looking for middle-ground approaches that preserve student agency. A reader who wants a concrete example of how to structure such an intervention will get practical value from the description, even with the limitations. It deserves a serious referee because the topic is timely, the intervention is grounded in established theory, and the honest discussion of what did not work adds to the conversation. A revision could strengthen it with better measurement. I would recommend sending it to peer review.

Referee Report

3 major / 1 minor

Summary. The manuscript describes the design and deployment of a non-binding 'GenAI Contract' intervention in an undergraduate intermediate Python course with N=217 students. Grounded in harm reduction and self-regulated learning theory, students articulated personal learning goals and usage guidelines for generative AI tools, with periodic reflections across an 11-week semester. The contract was graded only for completion. The authors report that 58% of students indicated the intervention changed their thinking and created helpful accountability structures, while noting that awareness did not consistently produce sustained behavior change under deadline and tool-availability pressures.

Significance. If the self-reported changes can be causally attributed to the intervention, the work offers a practical, low-enforcement scaffold for promoting intentional GenAI use in CS education. It provides empirical data from a large real-world cohort on the gap between awareness and sustained self-regulation, which could inform curriculum design addressing the tension between tool accessibility and learning goals.

major comments (3)

[Results] Results section: the central claim that the contract 'worked' for 58% of students by changing thinking and creating accountability structures lacks causal grounding, as there is no control group, pre-intervention baseline survey, or objective usage logs to rule out course maturation, external media, or demand characteristics.
[Methods] Methods section: the 58% figure and the reported gap between awareness and behavior change are presented without details on survey instruments, response rates, statistical methods, or controls, limiting assessment of whether the data support the stated outcomes.
[Discussion] Discussion section: the observation that many students abandoned self-set guidelines under pressure is acknowledged, but the manuscript does not test or propose mechanisms to reduce the self-control burden, which is load-bearing for the harm-reduction framing.

minor comments (1)

[Abstract] Abstract: the phrasing 'it worked' for the 58% subgroup should be qualified as self-reported to avoid implying objective behavior change.

Simulated Author's Rebuttal

3 responses · 0 unresolved

We thank the referee for the constructive comments on causality, methods transparency, and mechanisms for reducing self-control demands. We have revised the manuscript to clarify the observational nature of the findings, expand methodological details, and propose supports for sustained adherence while preserving the harm-reduction framing.

read point-by-point responses

Referee: [Results] Results section: the central claim that the contract 'worked' for 58% of students by changing thinking and creating accountability structures lacks causal grounding, as there is no control group, pre-intervention baseline survey, or objective usage logs to rule out course maturation, external media, or demand characteristics.

Authors: We agree the design precludes causal claims. The 58% figure reflects students' self-reported perception that the contract changed their thinking about GenAI use and provided accountability. We have revised the Results and Discussion to remove any implication of proven effectiveness, explicitly label the data as self-reported, and add a limitations subsection addressing confounds such as course maturation, external influences, and demand characteristics. The contribution is the documentation of the awareness-to-behavior gap in a large naturalistic cohort, which can guide future controlled work. revision: yes
Referee: [Methods] Methods section: the 58% figure and the reported gap between awareness and behavior change are presented without details on survey instruments, response rates, statistical methods, or controls, limiting assessment of whether the data support the stated outcomes.

Authors: We have expanded the Methods section with full details on the survey instruments (exact wording of questions yielding the 58% response and behavior-change items), response rates among the 217 students, and the descriptive (non-inferential) analytic approach. We also state explicitly that no control conditions or baseline measures were collected. These additions allow readers to assess the data's scope and limitations directly. revision: yes
Referee: [Discussion] Discussion section: the observation that many students abandoned self-set guidelines under pressure is acknowledged, but the manuscript does not test or propose mechanisms to reduce the self-control burden, which is load-bearing for the harm-reduction framing.

Authors: We have revised the Discussion to propose specific mechanisms for lowering the self-control burden while retaining the harm-reduction emphasis: automated AI-interaction logging for effortless reflection, platform-integrated low-effort nudges aligned with personal guidelines, and simplified template-based guidelines that minimize ongoing monitoring. These are presented as directions for future work rather than tested interventions in the current study. revision: partial

Circularity Check

0 steps flagged

Empirical intervention study with no derivation chain or self-referential inputs

full rationale

The paper describes the design and deployment of a non-binding GenAI usage contract in a classroom setting with N=217 students, followed by self-reported outcomes (e.g., 58% reporting changed thinking). No equations, fitted parameters, predictive models, or mathematical derivations appear. The central claims rest on observed survey responses and qualitative reflections rather than any reduction to prior inputs, self-citations, or ansatzes. This is a standard empirical evaluation of an educational intervention and contains no circular steps.

Axiom & Free-Parameter Ledger

0 free parameters · 2 axioms · 1 invented entities

The central claim rests on the premise that a lightweight, student-authored contract can scaffold intentional GenAI use; this depends on the applicability of self-regulated learning theory and the assumption that self-awareness can overcome the low-effort alternative of unrestricted AI use.

axioms (2)

domain assumption Self-regulated learning theory provides an effective framework for guiding student decisions about GenAI tool use
The contract design explicitly draws on this theory to structure goal articulation and reflection.
ad hoc to paper A non-binding contract emphasizing self-awareness can promote better decision-making without external enforcement
The intervention deliberately avoids grading for compliance to focus on internal accountability.

invented entities (1)

GenAI Contract no independent evidence
purpose: A student-authored document for setting personal learning goals and usage guidelines for generative AI tools
Newly designed intervention introduced and tested in this study.

pith-pipeline@v0.9.0 · 5501 in / 1407 out tokens · 48872 ms · 2026-05-15T12:33:38.575020+00:00 · methodology

discussion (0)

Reference graph

Works this paper leans on

33 extracted references · 33 canonical work pages

[1]

Areej Ali, Aayushi Hingle Collier, Umama Dewan, Nora McDonald, and Aditya Johri. 2025. Analysis of generative AI policies in computing course syllabi. In Proceedings of the 56th ACM Technical Symposium on Computer Science Education V. 1. 18–24

work page 2025
[2]

Albert Bandura. 1982. Self-efficacy mechanism in human agency.American psychologist37, 2 (1982), 122

work page 1982
[3]

Seth Bernstein, Ashfin Rahman, Nadia Sharifi, Ariunjargal Terbish, and Stephen MacNeil. 2025. Beyond the Benefits: A Systematic Review of the Harms and Consequences of Generative AI in Computing Education. InProceedings of the 25th Koli Calling International Conference on Computing Education Research. 1–18

work page 2025
[4]

Samuel Boguslawski, Rowan Deer, and Mark G Dawson. 2025. Programming education and learner motivation in the age of generative AI: student and educator perspectives.Information and Learning Sciences126, 1/2 (2025), 91–109

work page 2025
[5]

High-Level

Kathryn Cunningham, Yike Qiao, Alex Feng, and Eleanor O’Rourke. 2022. Bring- ing" High-Level" Down to Earth: Gaining Clarity in Conversational Programmer Learning Goals. InProceedings of the 53rd ACM Technical Symposium on Computer Science Education-Volume 1. 551–557

work page 2022
[6]

Paul Denny, Stephen MacNeil, Jaromir Savelka, Leo Porter, and Andrew Luxton- Reilly. 2024. Desirable characteristics for ai teaching assistants in programming education. InProceedings of the 2024 on Innovation and Technology in Computer Science Education V. 1. 408–414

work page 2024
[7]

That’s Not the Way I Would Explain It

Abbey Gandhi and Kasia Muldner. 2025. " That’s Not the Way I Would Explain It": A Teacher-Researcher’s Autoethnography of Learning to Program With ChatGPT. InProceedings of the 2025 ACM Conference on International Computing Education Research V. 1. 227–239

work page 2025
[8]

Peter M Gollwitzer. 1999. Implementation intentions: strong effects of simple plans.American psychologist54, 7 (1999), 493

work page 1999
[9]

Peter M Gollwitzer and Paschal Sheeran. 2006. Implementation intentions and goal achievement: A meta-analysis of effects and processes.Advances in experi- mental social psychology38 (2006), 69–119

work page 2006
[10]

Irene Hou, Sophia Mettille, Owen Man, Zhuo Li, Cynthia Zastudil, and Stephen MacNeil. 2024. The effects of generative ai on computing students’ help-seeking preferences. InProceedings of the 26th australasian computing education conference. 39–48

work page 2024
[11]

Majeed Kazemitabaar, Runlong Ye, Xiaoning Wang, Austin Zachary Henley, Paul Denny, Michelle Craig, and Tovi Grossman. 2024. Codeaid: Evaluating a classroom deployment of an llm-based programming assistant that balances student and educator needs. InProceedings of the 2024 chi conference on human factors in computing systems. 1–20

work page 2024
[12]

Michael S Kirkpatrick. 2025. Challenging AI as Critical Thinking. InInternational Conference on the Ethical and Social Impacts of ICT. Springer, 239–251

work page 2025
[13]

Shao-Heng Ko and Kristin Stephens-Martinez. 2025. Rethinking Computing Students’ Help Resource Utilization through Sequentiality.ACM Transactions on Computing Education25, 1 (2025), 1–34

work page 2025
[14]

Shao-Heng Ko, Kristin Stephens-Martinez, Matthew Zahn, Yesenia Velasco, Lina Battestilli, and Sarah Heckman. 2025. Student perceptions of the help resource landscape. InProceedings of the 56th ACM Technical Symposium on Computer Science Education V. 1. 596–602

work page 2025
[15]

Pamela K Kohler, Lisa E Manhart, and William E Lafferty. 2008. Abstinence-only and comprehensive sex education and the initiation of sexual activity and teen pregnancy.Journal of adolescent Health42, 4 (2008), 344–351

work page 2008
[16]

Ban it till we understand it

Sam Lau and Philip Guo. 2023. From" Ban it till we understand it" to" Resistance is futile": How university programming instructors plan to adapt as more students use AI code generation and explanation tools such as ChatGPT and GitHub Copilot. InProceedings of the 2023 ACM Conference on International Computing Education Research-Volume 1. 106–121

work page 2023
[17]

Mark Liffiton, Brad E Sheese, Jaromir Savelka, and Paul Denny. 2023. Codehelp: Using large language models with guardrails for scalable support in program- ming classes. InProceedings of the 23rd Koli Calling International Conference on Computing Education Research. 1–11

work page 2023
[18]

Rongxin Liu, Carter Zenke, Charlie Liu, Andrew Holmes, Patrick Thornton, and David J Malan. 2024. Teaching CS50 with AI: leveraging generative artificial intel- ligence in computer science education. InProceedings of the 55th ACM technical symposium on computer science education V. 1. 750–756

work page 2024
[19]

Dastyni Loksa, Lauren Margulieux, Brett A Becker, Michelle Craig, Paul Denny, Raymond Pettit, and James Prather. 2022. Metacognition and self-regulation in programming education: Theories and exemplars of use.ACM Transactions on Computing Education (TOCE)22, 4 (2022), 1–31

work page 2022
[20]

Lauren E Margulieux, James Prather, Brent N Reeves, Brett A Becker, Gozde Cetin Uzun, Dastyni Loksa, Juho Leinonen, and Paul Denny. 2024. Self-regulation, self-efficacy, and fear of failure interactions with how novices use llms to solve programming problems. InProceedings of the 2024 on Innovation and Technology in Computer Science Education V. 1. 276–282

work page 2024
[21]

G Alan Marlatt. 1996. Harm reduction: Come as you are.Addictive behaviors21, 6 (1996), 779–788

work page 1996
[22]

Michal A Naisteter and Justin A Sitron. 2010. Minimizing harm and maximizing pleasure: Considering the harm reduction paradigm for sexuality education. American Journal of Sexuality Education5, 2 (2010), 101–115

work page 2010
[23]

Aadarsh Padiyath. 2024. Do I Have a Say in This, or Has ChatGPT Already Decided for Me?XRDS: Crossroads, The ACM Magazine for Students31, 1 (2024), 52–55

work page 2024
[24]

Aadarsh Padiyath, Xinying Hou, Amy Pang, Diego Viramontes Vargas, Xingjian Gu, Tamara Nelson-Fromm, Zihan Wu, Mark Guzdial, and Barbara Ericson. 2024. Insights from Social Shaping Theory: The Appropriation of Large Language Models in an Undergraduate Programming Course. InProceedings of the 2024 ACM Conference on International Computing Education Research...

work page doi:10.1145/3632620.3671098 2024
[25]

Jennifer Parham, Leo Gugerty, and DE Stevenson. 2010. Empirical evidence for the existence and uses of metacognition in computer science problem solving. In Proceedings of the 41st ACM technical symposium on Computer science education. 416–420

work page 2010
[26]

James Prather, Paul Denny, Juho Leinonen, Brett A Becker, Ibrahim Albluwi, Michelle Craig, Hieke Keuning, Natalie Kiesler, Tobias Kohn, Andrew Luxton- Reilly, et al. 2023. The robots are here: Navigating the generative ai revolution in computing education. InProceedings of the 2023 working group reports on innovation and technology in computer science edu...

work page 2023
[27]

James Prather, Brent N Reeves, Juho Leinonen, Stephen MacNeil, Arisoa S Randri- anasolo, Brett A Becker, Bailey Kimmel, Jared Wright, and Ben Briggs. 2024. The widening gap: The benefits and harms of generative ai for novice programmers. InProceedings of the 2024 ACM Conference on International Computing Education Research-Volume 1. 469–486

work page 2024
[28]

Annapurna Vadaparty, Daniel Zingaro, David H Smith IV, Mounika Padala, Chris- tine Alvarado, Jamie Gorson Benario, and Leo Porter. 2024. Cs1-llm: Integrating llms into cs1 instruction. InProceedings of the 2024 on Innovation and Technology in Computer Science Education v. 1. 297–303

work page 2024
[29]

Zilin Wang, Di Zou, Ruofei Zhang, Lap-Kei Lee, Haoran Xie, and Fu Lee Wang

work page
[30]

ChatGPT-enhanced self-regulated learning in programming education: impacts on motivation, self-efficacy, and learning outcomes.Interactive Learning Environments(2025), 1–26

work page 2025
[31]

Ramazan Yilmaz and Fatma Gizem Karaoglan Yilmaz. 2023. Augmented intelli- gence in programming learning: Examining student views on the use of ChatGPT for programming learning.Computers in Human Behavior: Artificial Humans1, 2 (2023), 100005

work page 2023
[32]

Cynthia Zastudil, Magdalena Rogalska, Christine Kapp, Jennifer Vaughn, and Stephen MacNeil. 2023. Generative ai in computing education: Perspectives of students and instructors. In2023 IEEE Frontiers in Education Conference (FIE). IEEE, 1–9

work page 2023
[33]

Barry J Zimmerman. 2002. Becoming a self-regulated learner: An overview. Theory into practice41, 2 (2002), 64–70

work page 2002

[1] [1]

Areej Ali, Aayushi Hingle Collier, Umama Dewan, Nora McDonald, and Aditya Johri. 2025. Analysis of generative AI policies in computing course syllabi. In Proceedings of the 56th ACM Technical Symposium on Computer Science Education V. 1. 18–24

work page 2025

[2] [2]

Albert Bandura. 1982. Self-efficacy mechanism in human agency.American psychologist37, 2 (1982), 122

work page 1982

[3] [3]

Seth Bernstein, Ashfin Rahman, Nadia Sharifi, Ariunjargal Terbish, and Stephen MacNeil. 2025. Beyond the Benefits: A Systematic Review of the Harms and Consequences of Generative AI in Computing Education. InProceedings of the 25th Koli Calling International Conference on Computing Education Research. 1–18

work page 2025

[4] [4]

Samuel Boguslawski, Rowan Deer, and Mark G Dawson. 2025. Programming education and learner motivation in the age of generative AI: student and educator perspectives.Information and Learning Sciences126, 1/2 (2025), 91–109

work page 2025

[5] [5]

High-Level

Kathryn Cunningham, Yike Qiao, Alex Feng, and Eleanor O’Rourke. 2022. Bring- ing" High-Level" Down to Earth: Gaining Clarity in Conversational Programmer Learning Goals. InProceedings of the 53rd ACM Technical Symposium on Computer Science Education-Volume 1. 551–557

work page 2022

[6] [6]

Paul Denny, Stephen MacNeil, Jaromir Savelka, Leo Porter, and Andrew Luxton- Reilly. 2024. Desirable characteristics for ai teaching assistants in programming education. InProceedings of the 2024 on Innovation and Technology in Computer Science Education V. 1. 408–414

work page 2024

[7] [7]

That’s Not the Way I Would Explain It

Abbey Gandhi and Kasia Muldner. 2025. " That’s Not the Way I Would Explain It": A Teacher-Researcher’s Autoethnography of Learning to Program With ChatGPT. InProceedings of the 2025 ACM Conference on International Computing Education Research V. 1. 227–239

work page 2025

[8] [8]

Peter M Gollwitzer. 1999. Implementation intentions: strong effects of simple plans.American psychologist54, 7 (1999), 493

work page 1999

[9] [9]

Peter M Gollwitzer and Paschal Sheeran. 2006. Implementation intentions and goal achievement: A meta-analysis of effects and processes.Advances in experi- mental social psychology38 (2006), 69–119

work page 2006

[10] [10]

Irene Hou, Sophia Mettille, Owen Man, Zhuo Li, Cynthia Zastudil, and Stephen MacNeil. 2024. The effects of generative ai on computing students’ help-seeking preferences. InProceedings of the 26th australasian computing education conference. 39–48

work page 2024

[11] [11]

Majeed Kazemitabaar, Runlong Ye, Xiaoning Wang, Austin Zachary Henley, Paul Denny, Michelle Craig, and Tovi Grossman. 2024. Codeaid: Evaluating a classroom deployment of an llm-based programming assistant that balances student and educator needs. InProceedings of the 2024 chi conference on human factors in computing systems. 1–20

work page 2024

[12] [12]

Michael S Kirkpatrick. 2025. Challenging AI as Critical Thinking. InInternational Conference on the Ethical and Social Impacts of ICT. Springer, 239–251

work page 2025

[13] [13]

Shao-Heng Ko and Kristin Stephens-Martinez. 2025. Rethinking Computing Students’ Help Resource Utilization through Sequentiality.ACM Transactions on Computing Education25, 1 (2025), 1–34

work page 2025

[14] [14]

Shao-Heng Ko, Kristin Stephens-Martinez, Matthew Zahn, Yesenia Velasco, Lina Battestilli, and Sarah Heckman. 2025. Student perceptions of the help resource landscape. InProceedings of the 56th ACM Technical Symposium on Computer Science Education V. 1. 596–602

work page 2025

[15] [15]

Pamela K Kohler, Lisa E Manhart, and William E Lafferty. 2008. Abstinence-only and comprehensive sex education and the initiation of sexual activity and teen pregnancy.Journal of adolescent Health42, 4 (2008), 344–351

work page 2008

[16] [16]

Ban it till we understand it

Sam Lau and Philip Guo. 2023. From" Ban it till we understand it" to" Resistance is futile": How university programming instructors plan to adapt as more students use AI code generation and explanation tools such as ChatGPT and GitHub Copilot. InProceedings of the 2023 ACM Conference on International Computing Education Research-Volume 1. 106–121

work page 2023

[17] [17]

Mark Liffiton, Brad E Sheese, Jaromir Savelka, and Paul Denny. 2023. Codehelp: Using large language models with guardrails for scalable support in program- ming classes. InProceedings of the 23rd Koli Calling International Conference on Computing Education Research. 1–11

work page 2023

[18] [18]

Rongxin Liu, Carter Zenke, Charlie Liu, Andrew Holmes, Patrick Thornton, and David J Malan. 2024. Teaching CS50 with AI: leveraging generative artificial intel- ligence in computer science education. InProceedings of the 55th ACM technical symposium on computer science education V. 1. 750–756

work page 2024

[19] [19]

Dastyni Loksa, Lauren Margulieux, Brett A Becker, Michelle Craig, Paul Denny, Raymond Pettit, and James Prather. 2022. Metacognition and self-regulation in programming education: Theories and exemplars of use.ACM Transactions on Computing Education (TOCE)22, 4 (2022), 1–31

work page 2022

[20] [20]

Lauren E Margulieux, James Prather, Brent N Reeves, Brett A Becker, Gozde Cetin Uzun, Dastyni Loksa, Juho Leinonen, and Paul Denny. 2024. Self-regulation, self-efficacy, and fear of failure interactions with how novices use llms to solve programming problems. InProceedings of the 2024 on Innovation and Technology in Computer Science Education V. 1. 276–282

work page 2024

[21] [21]

G Alan Marlatt. 1996. Harm reduction: Come as you are.Addictive behaviors21, 6 (1996), 779–788

work page 1996

[22] [22]

Michal A Naisteter and Justin A Sitron. 2010. Minimizing harm and maximizing pleasure: Considering the harm reduction paradigm for sexuality education. American Journal of Sexuality Education5, 2 (2010), 101–115

work page 2010

[23] [23]

Aadarsh Padiyath. 2024. Do I Have a Say in This, or Has ChatGPT Already Decided for Me?XRDS: Crossroads, The ACM Magazine for Students31, 1 (2024), 52–55

work page 2024

[24] [24]

Aadarsh Padiyath, Xinying Hou, Amy Pang, Diego Viramontes Vargas, Xingjian Gu, Tamara Nelson-Fromm, Zihan Wu, Mark Guzdial, and Barbara Ericson. 2024. Insights from Social Shaping Theory: The Appropriation of Large Language Models in an Undergraduate Programming Course. InProceedings of the 2024 ACM Conference on International Computing Education Research...

work page doi:10.1145/3632620.3671098 2024

[25] [25]

Jennifer Parham, Leo Gugerty, and DE Stevenson. 2010. Empirical evidence for the existence and uses of metacognition in computer science problem solving. In Proceedings of the 41st ACM technical symposium on Computer science education. 416–420

work page 2010

[26] [26]

James Prather, Paul Denny, Juho Leinonen, Brett A Becker, Ibrahim Albluwi, Michelle Craig, Hieke Keuning, Natalie Kiesler, Tobias Kohn, Andrew Luxton- Reilly, et al. 2023. The robots are here: Navigating the generative ai revolution in computing education. InProceedings of the 2023 working group reports on innovation and technology in computer science edu...

work page 2023

[27] [27]

James Prather, Brent N Reeves, Juho Leinonen, Stephen MacNeil, Arisoa S Randri- anasolo, Brett A Becker, Bailey Kimmel, Jared Wright, and Ben Briggs. 2024. The widening gap: The benefits and harms of generative ai for novice programmers. InProceedings of the 2024 ACM Conference on International Computing Education Research-Volume 1. 469–486

work page 2024

[28] [28]

Annapurna Vadaparty, Daniel Zingaro, David H Smith IV, Mounika Padala, Chris- tine Alvarado, Jamie Gorson Benario, and Leo Porter. 2024. Cs1-llm: Integrating llms into cs1 instruction. InProceedings of the 2024 on Innovation and Technology in Computer Science Education v. 1. 297–303

work page 2024

[29] [29]

Zilin Wang, Di Zou, Ruofei Zhang, Lap-Kei Lee, Haoran Xie, and Fu Lee Wang

work page

[30] [30]

ChatGPT-enhanced self-regulated learning in programming education: impacts on motivation, self-efficacy, and learning outcomes.Interactive Learning Environments(2025), 1–26

work page 2025

[31] [31]

Ramazan Yilmaz and Fatma Gizem Karaoglan Yilmaz. 2023. Augmented intelli- gence in programming learning: Examining student views on the use of ChatGPT for programming learning.Computers in Human Behavior: Artificial Humans1, 2 (2023), 100005

work page 2023

[32] [32]

Cynthia Zastudil, Magdalena Rogalska, Christine Kapp, Jennifer Vaughn, and Stephen MacNeil. 2023. Generative ai in computing education: Perspectives of students and instructors. In2023 IEEE Frontiers in Education Conference (FIE). IEEE, 1–9

work page 2023

[33] [33]

Barry J Zimmerman. 2002. Becoming a self-regulated learner: An overview. Theory into practice41, 2 (2002), 64–70

work page 2002