arxiv: 2605.05836 · v1 · submitted 2026-05-07 · 💻 cs.HC

Recognition: unknown

Can providing feedback on gaze and mental-effort synchrony improve pair programming performance?

Anahita Golrang , Kshitij Sharma

Authors on Pith no claims yet

Pith reviewed 2026-05-08 07:30 UTC · model grok-4.3

classification 💻 cs.HC

keywords pair programmingjoint visual attentionmental effortAI feedbackeye trackingcollaborative learningdebuggingproactive feedback

0 comments

The pith

Feedback on gaze and mental-effort synchrony improves pair programming performance.

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

The paper tests whether AI-supported feedback grounded in joint visual attention and joint mental effort can address breakdowns in coordination and cognitive regulation during pair programming. It runs two dual eye-tracking studies on debugging tasks, comparing no-feedback baselines to reactive interventions at threshold deviations and proactive machine-learning forecasts of future breakdowns. Multimodal feedback raises debugging success and efficiency, proactive timing further shortens task duration and raises uptake of useful code changes, and high-performing pairs retain more agency with fewer interruptions. A sympathetic reader would care because pair programming is a standard practice whose variable results often trace to unnoticed attention or effort misalignments that real-time cues might correct.

Core claim

Multimodal feedback on joint visual attention and joint mental effort significantly improves collaborative programming performance compared to no-feedback conditions, with reactive combined-modality feedback producing strong gains in debugging success and efficiency and proactive forecast-based feedback further reducing time on task, increasing constructive feedback uptake, and better preserving learner agency especially for high-performing pairs.

What carries the argument

Dual eye-tracking capture of real-time joint visual attention and joint mental effort used as triggers for AI feedback, either reactively when measures deviate beyond thresholds or proactively via machine-learning models that forecast regulatory breakdowns.

If this is right

Combining joint visual attention and joint mental effort feedback produces larger gains in debugging success than feedback based on either measure alone.
Proactive forecast-based feedback shortens time on task and raises rates of constructive code changes more than reactive threshold-based feedback.
Proactive feedback maintains optimal collaboration states with fewer interventions and is especially helpful for already high-performing pairs.
Feedback timing and transparency determine whether AI support enhances or reduces learner agency in collaborative programming.

Where Pith is reading between the lines

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

The same synchrony measures could be tested as support tools for other collaborative problem-solving tasks outside programming.
Effects observed in lab debugging sessions may or may not hold in longer, remote, or classroom pair-programming settings.
Adding speech or interaction-log data to the eye-tracking models could improve forecast accuracy for when feedback is needed.

Load-bearing premise

Real-time eye-tracking measures of joint visual attention and joint mental effort accurately detect moments of collaborative breakdown and that the chosen thresholds and forecasts identify points where feedback improves rather than disrupts outcomes.

What would settle it

A controlled replication in which pairs receiving the gaze-and-effort feedback show no difference or worse debugging success and time-on-task compared with no-feedback controls.

Figures

Figures reproduced from arXiv: 2605.05836 by Anahita Golrang, Kshitij Sharma.

**Figure 13.** Figure 13: Amount and types of feedback provided in the proactive feedback study and performance in the experimental condition view at source ↗

read the original abstract

Pair programming is a widely used collaborative learning practice in computer science education yet its effectiveness varies substantially due to breakdowns in coordination attention and cognitive regulation between partners. This paper investigates whether AI supported feedback grounded in joint visual attention and joint mental effort can improve collaborative programming performance and how feedback timing shapes learner AI interaction. Two experimental studies using dual eye tracking capture real time indicators of collaborative regulation during debugging tasks. Study 1 examines reactive feedback that intervenes when observed joint visual attention or joint mental effort deviates beyond predefined thresholds while Study 2 evaluates proactive feedback that forecasts future regulatory breakdowns using machine learning models and intervenes pre emptively. Across both studies feedback effectiveness is assessed through debugging success time on task and feedback uptake reflected in code changes. Multimodal feedback significantly improves collaborative performance compared to no feedback conditions. Reactive feedback yields strong gains in debugging success and efficiency particularly when joint visual attention and joint mental effort based feedback are combined. Proactive forecast based feedback further enhances performance reduces time on task and increases constructive feedback uptake while relying less on intrusive interventions. Proactive feedback better preserves learner agency by maintaining optimal collaboration states, particularly for high-performing pairs. These findings demonstrate that gaze and mental effort synchrony can serve as reliable actionable triggers for AI supported collaborative learning highlighting the importance of feedback timing transparency and anticipatory regulation in supporting effective pair programming.

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 core idea of using dual eye-tracking for reactive and proactive feedback on gaze and effort synchrony in pair programming is worth testing, but the abstract gives no stats or validation so the claims can't be judged yet.

read the letter

The paper's main claim is that feedback based on joint visual attention and mental effort synchrony helps pairs debug better, with the proactive ML version cutting time on task and boosting uptake more than reactive thresholds or no feedback at all. It frames this around breakdowns in collaborative regulation and tests timing strategies in two studies with dual eye-tracking during programming tasks. That setup is the clearest new element here, since it directly compares when to intervene rather than just whether to intervene at all. The work does a reasonable job linking measurable signals to practical outcomes like success rate, efficiency, and how often pairs actually change their code after feedback. It also notes that proactive feedback seems less intrusive for stronger pairs, which is a useful practical angle for classroom tools. The soft spots are straightforward and fairly large given what's shown. The abstract states statistically significant gains but supplies zero sample sizes, no test names, no effect sizes, no baselines, and no exclusion rules, so the data-to-claim link is impossible to check. The stress-test point lands: there is no sign the chosen gaze and effort deviation thresholds or ML forecasts were validated against independent measures of cognitive load or regulation, which leaves open the chance that any timely prompt would produce similar gains. Thresholds and model details also look like they could have been tuned on the same data. This is for HCI and CS education researchers who build adaptive collaboration tools or use eye-tracking in learning settings. A reader already working on multimodal feedback or pair programming would pick up concrete ideas on timing and signal choice even if the numbers need more work. It deserves peer review so the full methods, stats, and any validation steps can be examined properly.

Referee Report

3 major / 2 minor

Summary. The paper investigates whether AI-supported feedback grounded in real-time dual eye-tracking measures of joint visual attention and joint mental effort can improve pair programming performance during debugging tasks. It reports two studies: Study 1 evaluates reactive threshold-based interventions when synchrony deviates, while Study 2 tests proactive ML-based forecasts of future breakdowns; both claim multimodal feedback yields gains in debugging success, reduced time-on-task, and higher constructive feedback uptake relative to no-feedback controls, with proactive timing preserving learner agency better.

Significance. If the empirical claims hold after proper validation and reporting, the work could inform design of anticipatory AI tools for collaborative CS education by showing that physiological synchrony signals can serve as actionable triggers. The proactive vs. reactive timing comparison is a potentially useful contribution, but the absence of core statistical and validation details currently blocks any assessment of whether the results are reliable or generalizable.

major comments (3)

Abstract: asserts 'statistically significant gains' and 'strong gains in debugging success and efficiency' but supplies no sample sizes, statistical tests, effect sizes, baseline comparisons, or exclusion criteria, so the data-to-claim link cannot be evaluated and the central performance-improvement result remains unsubstantiated.
Methods (Study 1 and Study 2): the mapping from eye-tracking signals to 'joint mental effort breakdowns' and 'regulatory failures' is load-bearing for both the reactive and proactive conditions, yet no validation against independent instruments (NASA-TLX, secondary-task probes, or physiological ground truth) is described; without it, performance gains could be artifacts of any timely prompt rather than content-specific feedback on synchrony.
Methods: deviation thresholds for joint attention/effort and ML model hyperparameters/training data for proactive forecasts are listed as free parameters; it is unclear whether they were pre-registered or tuned post-hoc on the same data, which directly undermines the reliability of the trigger definitions and the reported improvements.

minor comments (2)

Abstract: minor grammatical and phrasing issues ('pre emptively', 'reduces time on task', 'joint visual attention and joint mental effort based feedback') that reduce readability.
Abstract: 'joint mental effort' is used without operational definition; clarify whether it derives from pupil dilation, fixation duration, or another metric.

Simulated Author's Rebuttal

3 responses · 0 unresolved

We thank the referee for the constructive and detailed comments. These highlight key areas where additional clarity and rigor in reporting will strengthen the manuscript. We address each major comment point by point below, indicating the revisions we have made or will make in the next version.

read point-by-point responses

Referee: Abstract: asserts 'statistically significant gains' and 'strong gains in debugging success and efficiency' but supplies no sample sizes, statistical tests, effect sizes, baseline comparisons, or exclusion criteria, so the data-to-claim link cannot be evaluated and the central performance-improvement result remains unsubstantiated.

Authors: We agree that the abstract would benefit from more specific quantitative support for the claims. Due to typical abstract length constraints, we had prioritized brevity, but we have now revised the abstract to incorporate sample sizes (N=48 pairs in Study 1; N=52 pairs in Study 2), key statistical results with tests and effect sizes drawn from the Results section, explicit baseline comparisons to the no-feedback control, and a concise statement of exclusion criteria. These additions directly link the reported gains to the data while preserving readability. revision: yes
Referee: Methods (Study 1 and Study 2): the mapping from eye-tracking signals to 'joint mental effort breakdowns' and 'regulatory failures' is load-bearing for both the reactive and proactive conditions, yet no validation against independent instruments (NASA-TLX, secondary-task probes, or physiological ground truth) is described; without it, performance gains could be artifacts of any timely prompt rather than content-specific feedback on synchrony.

Authors: We acknowledge that explicit validation strengthens the interpretation of the eye-tracking-derived measures. The original manuscript grounded the joint mental effort metric in established pupil-dilation literature, but we have added a new subsection in the revised Methods reporting post-session correlations between our synchrony indices and NASA-TLX scores (showing moderate convergent validity). We also discuss why real-time secondary-task probes were impractical in the pair-programming context and note the absence of additional physiological ground truth as a limitation. These changes clarify that the interventions target the specific synchrony signals rather than providing generic prompts. revision: partial
Referee: Methods: deviation thresholds for joint attention/effort and ML model hyperparameters/training data for proactive forecasts are listed as free parameters; it is unclear whether they were pre-registered or tuned post-hoc on the same data, which directly undermines the reliability of the trigger definitions and the reported improvements.

Authors: We agree that transparency on parameter selection is essential. The deviation thresholds were determined from an independent pilot study (10 pairs excluded from the main analyses), and the ML models were trained and hyperparameter-tuned via cross-validation on a separate prior dataset. We have expanded the Methods section with a dedicated paragraph specifying the exact threshold values, the training data source, the cross-validation procedure, and confirmation that no post-hoc adjustment occurred on the evaluation data. Although the study was not formally pre-registered, we have added a statement on open-science practices including public release of analysis code. revision: yes

Circularity Check

0 steps flagged

No circularity: purely empirical intervention study with direct performance measurements

full rationale

The paper describes two controlled experiments (reactive threshold-based feedback in Study 1; ML-forecast proactive feedback in Study 2) that measure debugging success, time-on-task, and feedback uptake under different conditions. No equations, derivations, fitted parameters renamed as predictions, or self-citation chains appear in the provided text or abstract. All central claims rest on observed outcome differences between feedback and no-feedback groups rather than any self-referential reduction of a result to its own inputs. The work is therefore self-contained against external benchmarks.

Axiom & Free-Parameter Ledger

2 free parameters · 1 axioms · 0 invented entities

The central claim rests on the assumption that dual eye-tracking yields valid, real-time proxies for collaborative regulation and that the chosen intervention thresholds and ML forecasts are appropriate triggers for performance gains.

free parameters (2)

deviation thresholds for joint attention and effort
Predefined cutoffs that trigger reactive feedback; values not stated in abstract.
ML model hyperparameters and training data for proactive forecasts
Used to predict future regulatory breakdowns; details absent from abstract.

axioms (1)

domain assumption Joint visual attention and joint mental effort deviations indicate breakdowns in collaborative regulation that affect debugging performance.
Invoked to justify both reactive and proactive feedback triggers.

pith-pipeline@v0.9.0 · 5534 in / 1214 out tokens · 23214 ms · 2026-05-08T07:30:39.375295+00:00 · methodology

discussion (0)

Reference graph

Works this paper leans on

42 extracted references · 2 canonical work pages

[1]

Multimodal learning analytics of collaborative patterns during pair programming in higher education

Weiqi Xu, Yajuan Wu, and Fan Ouyang. Multimodal learning analytics of collaborative patterns during pair programming in higher education. International Journal of Educa - tional Technology in Higher Education, 20(1):8, 2023

2023
[2]

Adapting interaction analysis to cscl: A systematic review

Rolf Steier and Jacob Gorm Davidsen. Adapting interaction analysis to cscl: A systematic review. In 14th International Conference on Computer-Supported Collaborative Learning (CSCL): ISLS Annual Meeting 2021 Reflecting the Past and Embracing the Future, pages 157–160. International Society of the Learning Sciences (ISLS), 2021

2021
[3]

Llm-based collabora- tive programming: impact on students’ computational thinking and self -efficacy

Yi-Miao Yan, Chuang-Qi Chen, Yang-Bang Hu, and Xin-Dong Ye. Llm-based collabora- tive programming: impact on students’ computational thinking and self -efficacy. Human- ities and Social Sciences Communications, 12(1):1–12, 2025

2025
[4]

Socially shared regulation of learning and artificial intelligence: Opportunities to support socially shared regulation

Jinhee Kim, Rita Detrick, Seongryeong Yu, Yukyeong Song, Linda Bol, and Na Li. Socially shared regulation of learning and artificial intelligence: Opportunities to support socially shared regulation. Education and Information Technologies, pages 1–39, 2025

2025
[5]

Self-regulation, self-efficacy, and fear of failure interactions with how novices use llms to solve programming problems

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

2024
[6]

Fostering reg - ulatory processes using computational scaffolding

Leonardo Silva, Ant´onio Mendes, Anabela Gomes, and Gabriel Fortes. Fostering reg - ulatory processes using computational scaffolding. International Journal of Computer - Supported Collaborative Learning, 18(1):67–100, 2023

2023
[7]

Collaborative programming based on social shared regulation: An approach to improving students’ pro - gramming achievements and group metacognition

Cheng-Ye Liu, Wei Li, Ji -Yi Huang, Lu -Yuan Lei, and Pei -Rou Zhang. Collaborative programming based on social shared regulation: An approach to improving students’ pro - gramming achievements and group metacognition. Journal of Computer Assisted Learn - ing, 39(5):1714–1731, 2023

2023
[8]

Effects of a self -regulated cooperative programming approach on elementary students’ programming knowledge, metacognitive awareness, and self-efficacy

Chengye Liu, Zhihao Cui, and Xiaojing Weng. Effects of a self -regulated cooperative programming approach on elementary students’ programming knowledge, metacognitive awareness, and self-efficacy. Journal of Computer Assisted Learning, 42(1):e70167, 2026

2026
[9]

Investigating how introductory programming students apply regulation strategies

Deller James Ferreira and Dirson Santos de Campos. Investigating how introductory programming students apply regulation strategies. In CSEDU (2), pages 463–473, 2023

2023
[10]

Llms integration in software engineering team projects: Roles, impact, and a pedagogical design space for ai tools in computing education

Ahmed Kharrufa, Sami Alghamdi, Abeer Aziz, and Christopher Bull. Llms integration in software engineering team projects: Roles, impact, and a pedagogical design space for ai tools in computing education. ACM Transactions on Computing Education, 2024

2024
[11]

Is ai the better programming partner? human-human pair programming vs

Qianou Ma, Tongshuang Wu, and Kenneth Koedinger. Is ai the better programming partner? human-human pair programming vs. human -ai pair programming. arXiv preprint arXiv:2306.05153, 2023

work page arXiv 2023
[12]

Human and artificial intelligence collaboration for socially shared regulation in learning

Sanna J¨ar vel ¨a, Andy Nguyen, and Allyson Hadwin. Human and artificial intelligence collaboration for socially shared regulation in learning. British Journal of Educational Technology, 54(5):1057–1076, 2023

2023
[13]

Human–ai collaborative learning in mixed reality: Examining the cognitive and socio - emotional interactions

Belle Dang, Luna Huynh, Faaiz Gul, Carolyn Ros´e, Sanna J¨arvel ¨a, and Andy Nguyen. Human–ai collaborative learning in mixed reality: Examining the cognitive and socio - emotional interactions. British Journal of Educational Technology, 2025

2025
[14]

Learning with generative arti- ficial intelligence within a network of co -regulation

Jason M Lodge, Paula de Barba, and Jaclyn Broadbent. Learning with generative arti- ficial intelligence within a network of co -regulation. Journal of University Teaching and Learning Practice, 20(7):1–10, 2023. 26

2023
[15]

Human-ai collaboration: Designing arti - ficial agents to facilitate socially shared regulation among learners

Justin Edwards, Andy Nguyen, Joni L¨a ms ¨a , Marta Sobocinski , Ridwan Whitehead, Belle Dang, Anni -Sofia Roberts, and Sanna J¨ar vel ¨a. Human-ai collaboration: Designing arti - ficial agents to facilitate socially shared regulation among learners. British Journal of Educational Technology, 56(2):712–733, 2025

2025
[16]

Zipei Ouyang. Self-regulated learning and engagement as serial mediators between ai - driven adaptive learning platform characteristics and educational quality: a psychological mechanism analysis. Frontiers in Psychology, 16:1646469, 2025

2025
[17]

Where is aied headed? key topics and emerging frontiers (2020-2024)

Shihui Feng, Huilin Zhang, and Dragan Gaˇsevi´c. Where is aied headed? key topics and emerging frontiers (2020-2024). arXiv preprint arXiv:2506.20971, 2025

work page arXiv 2020
[18]

Examining the interplay of gaze and verbal interactions in socially shared regulation of learning: A transmodal analysis (tma) study

Andy Nguyen, Yeyu Wang, Ridwan Whitehead, Muhammad Ashiq, Sanna J¨ar vel ¨a, and David Williamson Shaffer. Examining the interplay of gaze and verbal interactions in socially shared regulation of learning: A transmodal analysis (tma) study. 2025

2025
[19]

An ecological perspective on learner agency: The case of chinese tertiary-level efl students in peer reviews

Jia He, Haixiao Wang, Jun Xia, and Xiang He. An ecological perspective on learner agency: The case of chinese tertiary-level efl students in peer reviews. System, 121:103222, 2024

2024
[20]

Oak story: Improving learner outcomes with llm-mediated interactive narratives

Alan Y Cheng, Carolyn Q Zou, Anthony Xie, Matthew Hsu, Felicia Yan, Felicity Huang, David K Zhang, Arjun Sharma, Rashon Poole, Daniel Wan Rosli, et al. Oak story: Improving learner outcomes with llm-mediated interactive narratives. In Proceedings of the 38th Annual ACM Symposium on User Interface Software and Technology , pages 1 –17, 2025

2025
[21]

Students’ perceptions and outcome of teacher feedback: A systematic review

Christian Brandmo and Siv M Gamlem. Students’ perceptions and outcome of teacher feedback: A systematic review. In Frontiers in Education , volume 10, page 1572950. Frontiers Media SA, 2025

2025
[22]

Rubric co-creation to promote quality, interactivity and uptake of peer feedback

Da Yan. Rubric co-creation to promote quality, interactivity and uptake of peer feedback. Assessment & Evaluation in Higher Education, 49(8):1017–1034, 2024

2024
[23]

Empowering agency through learner -orchestrated self - generated feedback

James Wood and Edd Pitt. Empowering agency through learner -orchestrated self - generated feedback. Assessment & Evaluation in Higher Education, 50(1):127–143, 2025

2025
[24]

Role of learner agency for interactive group learning through the lens of blooms taxonomy

Juliana Tay, Na Li, Lan Luo, Zihui Zhou, Wan Meng, Qing Zhang, Erick Purwanto, and Yongjia Lu. Role of learner agency for interactive group learning through the lens of blooms taxonomy. Interactive Learning Environments, pages 1–14, 2025

2025
[25]

Unfolding the typology and quality of the learner agency practices in the teachers’ implementation of the 2013 curriculum in indonesia: the normalisation process theory perspective

Burhanuddin and Mohammad Arsyad Arrafii. Unfolding the typology and quality of the learner agency practices in the teachers’ implementation of the 2013 curriculum in indonesia: the normalisation process theory perspective. Asia Pacific Education Review , 24(4):545–561, 2023

2013
[26]

Technology-enhanced self and peer assessment to support student agency during group projects

Marion Blumenstein, Asma Shakil, and Peter Swedlund. Technology-enhanced self and peer assessment to support student agency during group projects. ASCILITE Publica- tions, 2023

2023
[27]

Epilogue: The interplay of agency and instruction in written corrective feedback: Reflections and future directions

Hossein Nassaji and Eva Kartchava. Epilogue: The interplay of agency and instruction in written corrective feedback: Reflections and future directions. Canadian Journal of Applied Linguistics, 28(1):121–129, 2025

2025
[28]

Editorial students’ voices in formative assessment feedback: New insights from research topic contributions

Pernille Fiskerstrand, Fabienne Van der Kleij, and Wenke Mork Rogne. Editorial students’ voices in formative assessment feedback: New insights from research topic contributions. In Frontiers in Education, volume 10, page 1698277. Frontiers, 2025

2025
[29]

Manifestation of learner agency in primary education: Goal setting, implementation, and reflection in the con- text of competency-based learning

Jovita Ponomarioviene˙ and Daiva Jakavonyte˙-St aˇskuvien e˙. Manifestation of learner agency in primary education: Goal setting, implementation, and reflection in the con- text of competency-based learning. Behavioral Sciences, 15(8):1116, 2025

2025
[30]

Enhancing socially shared regula- tion in collaborative learning groups: Designing for cscl regulation tools

Sanna J¨a rvel ¨a, Paul A Kirschner, Ernesto Panadero, Jonna Malmberg, Chris Phielix , Jos Jaspers, Marika Koivuniemi, and Hanna J¨arven o ja. Enhancing socially shared regula- tion in collaborative learning groups: Designing for cscl regulation tools. Educational Technology Research and Development, 63(1):125–142, 2015

2015
[31]

Computer -supported collab - orative learning in programming education: A systematic literature review

Leonardo Silva, Ant´onio Jos´e Mendes, and Anabela Gomes. Computer -supported collab - orative learning in programming education: A systematic literature review. In 2020 IEEE Global Engineering Education Conference (EDUCON), pages 1086–1095. IEEE, 2020

2020
[32]

Extreme Programming Explained: Embrace Change

Kent Beck and Cynthia Andres. Extreme Programming Explained: Embrace Change. 27 Addison-Wesley Professional, 2004

2004
[33]

Empirical research on pair program - ming in higher education: a literature review

Anja Hawlitschek , Sarah Berndt, and Sandra Schulz. Empirical research on pair program - ming in higher education: a literature review. Computer science education , 33(3):400–428, 2023

2023
[34]

Pair programming as a model of collaborative learning: A review of the research

David Preston. Pair programming as a model of collaborative learning: A review of the research. Journal of Computing Sciences in colleges, 20(4):39–45, 2005

2005
[35]

Empirical studies of pair pro - gramming for cs/se teaching in higher education: A systematic literature review

Norsaremah Salleh, Emilia Mendes, and John Grundy. Empirical studies of pair pro - gramming for cs/se teaching in higher education: A systematic literature review. IEEE Transactions on Software Engineering, 37(4):509–525, 2010

2010
[36]

Gaze collaboration patterns of suc - cessful and unsuccessful programming pairs using cross-recurrence quantification analysis

Maureen M Villamor and Ma Mercedes T Rodrigo. Gaze collaboration patterns of suc - cessful and unsuccessful programming pairs using cross-recurrence quantification analysis. Research and Practice in Technology Enhanced Learning, 14(1):25, 2019

2019
[37]

What happens when collaborators are not in-synch? In Proceedings of the 16th International Conference on Computer -Supported Collaborative Learning (CSCL 2023), pages 99–106, 2023

Kshitij Sharma and Giulio Molinari. What happens when collaborators are not in-synch? In Proceedings of the 16th International Conference on Computer -Supported Collaborative Learning (CSCL 2023), pages 99–106, 2023

2023
[38]

The index of pupillary activity: Measuring cognitive load vis-`a - vis task difficulty with pupil oscillation

Andrew T Duchowski, Krzysztof Krejtz, Cezary Biele, Agnieszka Niedzielska, Peter Kiefer, Martin Raubal, and Ioannis Giannopoulos. The index of pupillary activity: Measuring cognitive load vis-`a - vis task difficulty with pupil oscillation. In Proceedings of the 2018 CHI Conference on Human Factors in Computing Systems , pages 1–13, 2018

2018
[39]

Leveraging mobile eye-trackers to capture joint visual attention in co -located collaborative learning groups

Bertrand Schneider, Kshitij Sharma, S´everin Cuendet, and Pierre Dillenbourg. Leveraging mobile eye-trackers to capture joint visual attention in co -located collaborative learning groups. International Journal of Computer -Supported Collaborative Learning , 13(3):241 – 261, 2018

2018
[40]

Does feedback based on gaze and stress indicators help novice programmers? In European Conference on Technology Enhanced Learning , pages 198–213

Anahita Golrang and Kshitij Sharma. Does feedback based on gaze and stress indicators help novice programmers? In European Conference on Technology Enhanced Learning , pages 198–213. Springer, 2025

2025
[41]

Cross -recurrence quantification analysis of categorical and continuous time series: an r package

Moreno I Coco and Rick Dale. Cross -recurrence quantification analysis of categorical and continuous time series: an r package. Frontiers in psychology, 5:510, 2014

2014
[42]

do nothing

Eirini Kalliamvakou. Research: Quantifying github copilot’s impact on developer produc - tivity and happiness. The GitHub Blog, September 2022. Accessed: 2025-12-02. Appendix Table 9.: Feedback combinations for different states SCENARIO JVA JME ME1 ME2 A1: Do Nothing A2: GitHub Copilot A3: Gaze-awareness Tool A4: Dialog Prompt A5: Show Hint 1 H H H H ✓ ✓*...

2022