Investigating the Impact of Programmers' Emotions on Code Quality
Disciplines
Bioelectrical and Neuroengineering
Abstract (300 words maximum)
This study explores the relationship between programmers' emotional states and the quality of the code they produce. In an experimental setting, ten volunteers were tasked with coding while their brain activity was measured using non-invasive EEG technology, focusing on the left and right prefrontal cortex, regions often linked to emotional processing. The Frontal Asymmetry Index (FAI) was employed to assess emotional states, complemented by self-reported data from the SPANE questionnaires, which captured both positive and negative emotions. Code quality was assessed using a rubric designed to identify task failures. Preliminary results suggest that positive emotions may correlate with poorer code quality. However, the small sample size (n=10) resulted in inconclusive findings regarding the connection between self-reported emotions and code quality. Expanding this research with a larger sample is essential to better understand the potential link between emotional states and programmers’ performance.
Academic department under which the project should be listed
CCSE - Information Technology
Primary Investigator (PI) Name
Dr. Maria Valero de Clemente
Additional Faculty
Adriane Randolph
arandol3@kennesaw.edu
Professor
College of Business
Department of Information Systems and Security
Investigating the Impact of Programmers' Emotions on Code Quality
This study explores the relationship between programmers' emotional states and the quality of the code they produce. In an experimental setting, ten volunteers were tasked with coding while their brain activity was measured using non-invasive EEG technology, focusing on the left and right prefrontal cortex, regions often linked to emotional processing. The Frontal Asymmetry Index (FAI) was employed to assess emotional states, complemented by self-reported data from the SPANE questionnaires, which captured both positive and negative emotions. Code quality was assessed using a rubric designed to identify task failures. Preliminary results suggest that positive emotions may correlate with poorer code quality. However, the small sample size (n=10) resulted in inconclusive findings regarding the connection between self-reported emotions and code quality. Expanding this research with a larger sample is essential to better understand the potential link between emotional states and programmers’ performance.