The third international workshop on Eye Movements in Programming will focus on advancing the methodological, theoretical and applied aspects of eye-movements in programming.
Previously we focused on understanding expert (2013) and novice (2014) source code reading behaviors, and we launched the first distributed data collection in this discipline. By analyzing the pre-shared datasets and discussion of bottom-up data-to-models approaches, the workshop produced new methods, tools and understanding of source code reading gaze patterns.
The third edition aims at investigating the opposite direction. We invite contributions departing from theoretical perspectives and grounds to present new hypotheses about gaze behaviour in comprehension, debugging, and other tasks of programming. These may include, but are not limited to, affective computing in programming, vision based models, readability, and new theories of program comprehension.
Contributions are expected to present implications to industrial programming practice or programming education. In preparation for the workshop session, participants submit a short paper (about two pages) discussing how various aspects of comprehension leave their trace in gaze. All submissions will undergo a light review by the program committee.
A technical workshop report including the position papers will be published. At the workshop we will discuss the proposed operationalizations of comprehension and how to validate them against the data gathered during the Distributed Collection of Eye Movement Data in Programming (http://bit.ly/emip_data). Furthermore, we will prepare a joint publication with the results. To participate send an email to Roman Bednarik.
Participation is independent of attending ACM Koli Calling. Workshop web-site: emipws.org.
The aim of the position papers is to discuss aspects of a cognitive model or process, a strategy, a certain element of program comprehension, etc. The arguments should be derived from literature (psychology, computer science etc.) or experience (e.g. teaching novices, working with programmers in industry).
Authors are encouraged to describe the model and operationalize it, so it is possible to validate the occurrences of this particular visual behavior in actual gaze data. At the workshop session, we will discuss how to evaluate the presented models on the dataset from the Distributed Data Collection; the dataset includes over a hundred of programmers of different expertise from total novice to expert with many years of professional experience reading Java (and Python).
ACM 2 column format
www.acm.org/sigs/publications/proceedings-templates (Option 2 for LaTeX)