Emotional Contagion in Code: How GitHub Emoji Reactions Shape Developer Collaboration

Obada Kraishan · 2025

Read Emotional Contagion in Code: How GitHub Emoji Reactions Shape Developer Collaboration by Obada Kraishan (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2511-02515.txt.

Introduction

Obada Kraishan, a researcher, tracks how emotions ripple through developer discussions on GitHub, using the platform's emoji reactions as the signal. The dataset covers 2,098 issues and pull requests from fifty popular repositories, holding 106,743 emoji reactions. The question is whether these tiny signals merely decorate technical talk or steer it in fact.

The landscape turned out surprisingly warm: 57.4 percent of discussions carried positive sentiment, and positive emotional cascades outnumbered negative ones by twenty-three to one. Nearly half the items, 45.6 percent, showed an instant enthusiasm pattern of instant positive reinforcement. Statistical tests confirmed strong emotional contagion, with a correlation of 0.679 between early and later reactions and a very large effect size. Kraishan concludes that first reactions shape how a discussion unfolds, challenging the picture of technical talk as purely rational.

Catalog record

CREATORObada Kraishan
YEAR2025
COLLECTIONPAPERS
FORMATTXT
FILE SIZE37 KB
VERSION1.0
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2511-02515

Browse the full library · About the Tower of Babel

Full text

The opening of the work, from the library's hosted text:

Emotional Contagion in Code: How GitHub Emoji Reactions Shape Developer Collaboration Obada Kraishan Abstract Developer communities increasingly rely on emoji reactions to communicate, but we know little about how these emotional signals spread and influence technical discussions. We analyzed 2,098 GitHub issues and pull requests across 50 popular repositories, examining patterns in 106,743 emoji reactions to understand emotional contagion in software development. Our findings reveal a surprisingly positive emotional landscape: 57.4% of discussions carry positive sentiment, with positive emotional cascades outnumbering negative ones 23:1. We identified five distinct patterns, with “instant enthusiasm” affecting 45.6% of items—nearly half receive immediate positive reinforcement. Statistical analysis confirms strong emotional contagion ( r = 0.679 r=0.679 , p < 0.001 p<0.001 ) with a massive effect size ( d = 2.393 d=2.393 ), suggesting that initial reactions powerfully shape discussion trajectories. These findings challenge assumptions about technical discourse being purely rational, demonstrating that even minimal emotional signals create measurable ripple effects. Our work provides empirical evidence that emoji reactions are not mere decoration but active forces shaping collaborative outcomes in software development. K eywords emotional contagion \cdot GitHub \cdot developer collaboration \cdot emoji reactions \cdot social coding \cdot sentiment analysis 1 Introduction Software development, while technical, involves significant emotional interactions among developers. Interactions such as code reviews and bug reports often involve emotional expressions like frustration, excitement, confusion, and celebration [ 1 , 2 ] . GitHub’s emoji reactions have fundamentally changed how emotions are expressed in development communities, turning simple acknowledgments like thumbs-up or heart emojis into catalysts for broader emotional shifts. To illustrate the impact of emoji reactions, consider a typical scenario: a developer submits their first pull request to a popular open-source project. Within minutes, a maintainer adds a rocket reaction. Other contributors follow with thumbs-up and heart reactions. The person who submits feels welcomed and energized. They engage more actively in the discussion, and the positive atmosphere attracts additional reviewers. Such interactions create an emotional cascade, a phenomenon known as emotional contagion in social psychology, occurring countless times daily across GitHub’s millions of repositories [ 3 , 4 ] . Regardless of their widespread use, the collective impact of emoji reactions remains surprisingly underexplored. Do positive reactions genuinely create more positive discussions? Can a single negative reaction derail an otherwise constructive conversation? More fundamentally, do these tiny emotional signals actually influence real outcomes like whether code gets merged or issues get resolved? This paper examines emotional contagion through GitHub emoji reactions. We examine how emotions spread through developer communities by analyzing thousands of issues and pull requests across popular repositories. Our investigation reveals that emotional contagion in code collaboration is not only real but remarkably powerful, with initial reactions creating ripple effects that shape entire discussions. Our investigation centers on three research questions: • RQ1: Do emoji reactions create measurable emotional contagion in GitHub discussions? • RQ2: What patterns of emotional expression appear from reaction behaviors? • RQ3: Do emotional signals correlate with collaboration outcomes? Our work makes three key contributions. First, we provide quantitative evidence that emotional contagion occurs in technical discussions. Second, we identify distinct emotional patterns that emerge in developer interactions. Third, we establish that these emotional signals correlate with concrete outcomes, challenging the assumption that technical quality alone drives development decisions. 2 Related Work 2.1 Emotion in Software Engineering The role of emotions in software development has gained significant attention due to their impact on productivity and code quality. Graziotin et al. [ 5 ] showed that developer happiness directly correlates with productivity, while Müller and Fritz [ 6 ] found that emotional states significantly impact code quality. These studies primarily focused on individual developers, measuring emotions through surveys, biometric sensors, or self-reporting. More recent work has examined emotions at the team level. Girardi et al. [ 7 ] analyzed sentiment in code review comments, finding that negative emotions lean to cluster around certain types of changes. Ortu et al. [ 8 ] connected emotional expressions in issue discussions to project success metrics. However, their analysis relied on textual sentiment rather than explicit emotional signals like reactions. Our work extends this line of research by examining how single emoji clicks, as minimal emotional cues, spread through developer networks. The introduction of emoji reactions on platforms like GitHub and Slack has created new channels for emotional expression in technical contexts. Claes et al. [ 9 ] studied emoji use in developer chat, finding that even senior developers regularly use emoji to soften criticism or celebrate achievements. However, their work focused on direct communication rather than the broader emotional signals conveyed through emoji reactions. Miller et al. [ 10 ] examined emoji interpretation across cultures, revealing potential miscommunication risks, though they did not investigate emotional transmission effects. 2.2 Emotional Contagion Theory Emotional contagion, defined as the spread of emotions between individuals, has been extensively documented in face-to-face interactions [ 11 , 12 ] . Hatfield et al. [ 13 ] identified three stages: mimicry (unconsciously copying others’ expressions), feedback (physiological responses to our own expressions), and convergence (aligning emotional states). While originally studied in physical settings, the stages of emotional contagion have also been observed in digital environments. A significant example of digital emotional contagion came from Kramer et al.’s Facebook experiment [ 14 ] , which showed that manipulating news feed sentiment influenced users’ subsequent posts. Other studies have found similar effects on Twitter [ 15 , 16 ] and in online gaming communities [ 17 ] . Building on these findings, Goldenberg and Gross [ 18 ] recently proposed that digital emotions might spread even more easily than in-person ones, as online disinhibition reduces emotional regulation. In professional contexts, emotional contagion affects team dynamics and performance. Barsade [ 19 ] showed that positive emotional contagion in work groups leads to improved cooperation and reduced conflict. Totterdell et al. [ 20 ] found that team mood convergence predicted better performance in professional sports teams. These findings suggest that emotional contagion could similarly impact collaboration outcomes in various professional settings, though this connection remains unexplored. 2.3 GitHub as a Social Platform GitHub, initially designed as a version control platform, has evolved into a social network where technical and social elements increasingly intertwine, as recognized by researchers [ 21 , 22 ] . Dabbish et al. [ 23 ] introduced the concept of “social coding,” showing how GitHub’s transparency features influence collaboration patterns. Marlow et al. [ 24 ] found that developers form impressions of others based on their GitHub activity, using social signals to evaluate technical competence. Beyond individual impressions, the social dynamics of GitHub encompass broader interactions that influence project outcomes. Casalnuovo et al. [ 25 ] demonstrated that social factors predict project success better than technical metrics alone. Tsay et al. [ 26 ] found that social distance between contributors affects pull request acceptance rates. These studies suggest that emotional dynamics, though often overlooked, might play a crucial role in influencing development outcomes alongside technical and social factors. Building on the understanding of GitHub as a social platform, recent work has begun examining its specific social features. Zhang et al. [ 27 ] analyzed the impact of GitHub Stars on project visibility, while Borges and Valente

Read the complete text