Using developer eye movements to externalize the mental model used in code summarization tasks

Abid, Nahla J.; Maletic, Jonathan I.; Sharif, Bonita

doi:10.1145/3314111.3319834

Cited by 24 publications

(3 citation statements)

References 22 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…A survey by [53] chronicles up to 2015 and is expanded upon by [37]. An exemplar body of work is led by Sharif et al, with advancements in both knowledge of how programmers read code [1,2,13,17] and practical uses for this knowledge [52,[54][55][56]. Eye tracking data is widely recognized as representing the attention given to dierent parts of source code by programmers.…”

Section: Eye Tracking In Soware Engineering (Se) Researchmentioning

confidence: 99%

“…One way to measure the attention that humans pay to source code is by tracking the movement of a person's eyes when reading that code. A plethora of eye tracking studies has demonstrated consistent processes that programmers follow when reading code [1,2,13,47,53], and it has even been demonstrated that programmers who are blind seek the same information from code, just using a dierent mechanical process [5]. In general, people are very ecient at extracting key features they need to understand code.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Towards Modeling Human Attention from Eye Movements for Neural Source Code Summarization

Bansal

Sharif

McMillan

2023

Proc. ACM Hum.-Comput. Interact.

Self Cite

View full text Add to dashboard Cite

Neural source code summarization is the task of generating natural language descriptions of source code behavior using neural networks. A fundamental component of most neural models is an attention mechanism. The attention mechanism learns to connect features in source code to specific words to use when generating natural language descriptions. Humans also pay attention to some features in code more than others. This human attention reflects experience and high-level cognition well beyond the capability of any current neural model. In this paper, we use data from published eye-tracking experiments to create a model of this human attention. The model predicts which words in source code are the most important for code summarization. Next, we augment a baseline neural code summarization approach using our model of human attention. We observe an improvement in prediction performance of the augmented approach in line with other bio-inspired neural models.

show abstract

Section: Eye Tracking In Soware Engineering (Se) Researchmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Towards Modeling Human Attention from Eye Movements for Neural Source Code Summarization

Bansal

Sharif

McMillan

2023

Proc. ACM Hum.-Comput. Interact.

Self Cite

View full text Add to dashboard Cite

show abstract

“…Program comprehension (Bednarik & Tukiainen, 2006) and code summarization tasks (Abid, Maletic, & Sharif, 2019) can be also approached by eye movement analysis. Approaches from research on natural-language text reading is also applicable to source code, however not without review (Busjahn, Bednarik, & Schulte, 2014).…”

Section: Eye Tracking In Development Tasksmentioning

confidence: 99%

A Comparison of Paper Sketch and Interactive Wireframe by Eye Movements Analysis, Survey, and Interview

Kieffer,

Vanderdonckt

2023

Proceedings of the Annual Hawaii International Conference on System Sciences

View full text Add to dashboard Cite

Eye movement-based analyses have been extensively performed on graphical user interface designs, mainly on high-fidelity prototypes such as coded prototypes. However, practitioners usually initiate the development life cycle with low-fidelity prototypes, such as mock-ups or sketches. Since little or no eye movement analysis has been performed on the latter, would eye tracking transpose its benefits from high-to low-fidelity prototypes and produce different results? To bridge this gap, we performed an eye movement-based analysis that compares gaze point indexes, gaze event types and durations, fixation and saccade indexes produced by N =8 participants between two treatments, a paper prototype vs. a wireframe. The paper also reports a qualitative analysis based on the answers provided by these participants in a semi-directed interview and on a perceived usability questionnaire with 14 items. Due to its interactivity, the wireframe seems to foster a more exploratory approach to design (e.g., testing and navigating more extensively) than the paper prototype.

show abstract

A practical guide on conducting eye tracking studies in software engineering

et al. 2020

View full text Add to dashboard Cite

Using developer eye movements to externalize the mental model used in code summarization tasks

Cited by 24 publications

References 22 publications

Towards Modeling Human Attention from Eye Movements for Neural Source Code Summarization

Towards Modeling Human Attention from Eye Movements for Neural Source Code Summarization

A Comparison of Paper Sketch and Interactive Wireframe by Eye Movements Analysis, Survey, and Interview

A practical guide on conducting eye tracking studies in software engineering

Contact Info

Product

Resources

About