Using Psycho-Physiological Measures to Assess Task Difficulty in Software Development

Author(s): Thomas Fritz, Andrew Begel
Venue: International Conference on Software Engineering
Date: 2014

Type of Experiement: Controlled Experiment
Sample Size: 15
Class/Experience Level: Professional
Participant Selection: lived in the greater Seattle, WA, interest in participating in user studies at microsoft, 2 years of software experience, knew C#, no bifocal glasses
Data Collection Method: Observation, Survey


Simple mistakes are often made by Software Engineers. This study introduces a new idea: stop the programmers before they get a chance to make those mistakes. It investigates an approach to classify the difficulty of code comprehension tasks using data from psycho-physiological sensors.

Psycho-physiological measures such as the eyes, brain, and skin, have been explored and linked to psychological, and specifically cognitive, processes and states. A brief overview of the process of the study was a pre-questionnaire, a series of tasks using external support tracking software, and a post test questionnaire. These tasks were short (several minutes) code comprehension tasks.

In conclusion, this study shows that a tool can detect when a new participant will perceive his tasks to be difficult with a precision of over 70% and a recall over 62%. Now that this shown to be feasible, researchers have the opportunity to develop new programming support tools, allowing them to potentially intervene in time to stop bugs from entering the code.