Using Eye Tracking to Investigate Reading Patterns and Learning Styles of Software Requirement Inspectors to Enhance Inspection Team Outcome

Abstract

Background -- Inspecting requirements and design artifacts to find faults saves rework effort significantly. While inspections are effective, their overall team performance rely on inspectors' ability to detect and report faults. Our previous research showed that individual inspectors have varying LSs (i.e., they vary in their ability to process information recorded in requirements document). To extend the results of our previous LS research, this paper utilizes the concept of eye tracking (to record eye movements of inspectors) along with their LSs to detect reading patterns of inspectors during requirements inspections. Aim -- The objective of this research is to analyze the reading trends of effective and efficient inspectors using eye movement and LS data of individual inspectors and virtual inspection teams. Method -- The current research uses data (LS, eye tracking, and inspection) from thirteen inspectors to find its impact on inspection effectiveness and efficiency. Results -- Results from this study show that, inspectors who detect more faults during inspection, focus significantly more at the fault region to find and report faults as opposed to comprehending requirements information. Results also showed Inspection teams with diverse inspectors outperform similar teams and spend more time in comprehending information at the fault region. Additionally, results showed that inspectors with SEQ LS significantly tends to focus more at fault locations and are preferred for inspection. Conclusion -- These results can aid the selection of inspectors during the inspection process thus improving software quality