Expertise in programming traditionally assumes a binary novice-expert divide. Learning resources typically target programmers who are learning programming for the first time, or expert programmers for that language. An underrepresented, yet important group of programmers are those that are experienced in one programming language, but desire to author code in a different language. For this scenario, we postulate that an effective form of feedback is presented as a transfer from concepts in the first language to the second. Current programming environments do not support this form of feedback. In this study, we apply the theory of learning transfer to teach a language that programmers are less familiar with-such as R-in terms of a programming language they already know-such as Python. We investigate learning transfer using a new tool called Transfer Tutor that presents explanations for R code in terms of the equivalent Python code. Our study found that participants leveraged learning transfer as a cognitive strategy, even when unprompted. Participants found Transfer Tutor to be useful across a number of affordances like stepping through and highlighting facts that may have been missed or misunderstood. However, participants were reluctant to accept facts without code execution or sometimes had difficulty reading explanations that are verbose or complex. These results provide guidance for future designs and research directions that can support learning transfer when learning new programming languages.
more »
« less
Evidence About Programmers for Programming Language Design (Dagstuhl Seminar 18061)
The report documents the program and outcomes of Dagstuhl Seminar 18061 "Evidence About Programmers for Programming Language Design". The seminar brought together a diverse group of researchers from the fields of computer science education, programming languages, software engineering, human-computer interaction, and data science. At the seminar, participants discussed methods for designing and evaluating programming languages that take the needs of programmers directly into account. The seminar included foundational talks to introduce the breadth of perspectives that were represented among the participants; then, groups formed to develop research agendas for several subtopics, including novice programmers, cognitive load, language features, and love of programming languages. The seminar concluded with a discussion of the current SIGPLAN artifact evaluation mechanism and the need for evidence standards in empirical studies of programming languages.
more »
« less
- NSF-PAR ID:
- 10073327
- Date Published:
- Journal Name:
- Dagstuhl reports
- Volume:
- 8
- Issue:
- 2
- ISSN:
- 2192-5283
- Page Range / eLocation ID:
- 1-25
- Format(s):
- Medium: X
- Sponsoring Org:
- National Science Foundation
More Like this
-
-
The impacts of many human factors on how people program are poorly understood and present significant challenges for work on improving programmer productivity and effective techniques for teaching and learning programming. Programming error messages are one factor that is particularly problematic, with a documented history of evidence dating back over 50 years. Such messages, commonly called compiler error messages, present difficulties for programmers with diverse demographic backgrounds. It is generally agreed that these messages could be more effective for all users, making this an obvious and high-impact area to target for improving programming outcomes. This report documents the program and the outputs of Dagstuhl Seminar 22052, “The Human Factors Impact of Programming Error Messages”, which explores this problem. In total, 11 on-site participants and 17 remote participants engaged in intensive collaboration during the seminar, including discussing past and current research, identifying gaps, and developing ways to move forward collaboratively to address these challenges.more » « less
-
more » « less
Abstract Source code is a form of human communication, albeit one where the information shared between the programmers reading and writing the code is constrained by the requirement that the code executes correctly. Programming languages are more syntactically constrained than natural languages, but they are also very expressive, allowing a great many different ways to express even very simple computations. Still, code written by developers is highly predictable, and many programming tools have taken advantage of this phenomenon, relying on language model
surprisal as a guiding mechanism. While surprisal has been validated as a measure of cognitive load in natural language, its relation to human cognitive processes in code is still poorly understood. In this paper, we explore the relationship between surprisal and programmer preference at a small granularity—do programmers prefer more predictable expressions in code? Usingmeaning‐preserving transformations , we produce equivalent alternatives to developer‐written code expressions and run a corpus study on Java and Python projects. In general, language models rate the code expressions developerschoose to write as more predictable than these transformed alternatives. Then, we perform two human subject studies asking participants to choose between two equivalent snippets of Java code with different surprisal scores (one original and transformed). We find that programmersdo prefer more predictable variants, and that stronger language models like the transformer align more often and more consistently with these preferences. -
Program comprehension is an important, but hard to measure cognitive process. This makes it difficult to provide suitable programming languages, tools, or coding conventions to support developers in their everyday work. Here, we explore whether functional magnetic resonance imaging (fMRI) is feasible for soundly measuring program comprehension. To this end, we observed 17 participants inside an fMRI scanner while they were comprehending source code. The results show a clear, distinct activation of five brain regions, which are related to working memory, attention, and language processing, which all fit well to our understanding of program comprehension. Furthermore, we found reduced activity in the default mode network, indicating the cognitive effort necessary for program comprehension. We also observed that familiarity with Java as underlying programming language reduced cognitive effort during program comprehension. To gain confidence in the results and the method, we replicated the study with 11 new participants and largely confirmed our findings. Our results encourage us and, hopefully, others to use fMRI to observe programmers and, in the long run, answer questions, such as: How should we train programmers? Can we train someone to become an excellent programmer? How effective are new languages and tools for program comprehension?more » « less
-
Programming can be an emotional experience, particularly for undergraduate students who are new to computer science. While researchers have interviewed novice programmers about their emotional experiences, it can be difficult to pinpoint the specific emotions that occur during a programming session. In this paper, we argue that electrodermal activity (EDA) sensors, which measure the physiological changes that are indicative of an emotional reaction, can provide a valuable new data source to help study student experiences. We conducted a study with 14 undergraduate students in which we collected EDA data while they worked on a programming problem. This data was then used to cue the participants’ recollections of their emotions during a retrospective interview about the programming experience. Using this methodology, we identified 21 distinct events that triggered student emotions, such as feeling anxiety due to a lack of perceived progress on the problem. We also identified common patterns in EDA data across multiple participants, such as a drop in their physiological reaction after developing a plan, corresponding with a calmer emotional state. These findings provide new information about how students experience programming that can inform research and practice, and also contribute initial evidence of the value of EDA data in supporting studies of emotions while programming.more » « less
- Free Publicly Accessible Full Text
-
Accepted Manuscript1.0
- Journal Article:
- https://doi.org/10.4230/DagRep.8.2.1
