Smartphone App May Be Able to Detect Depression
Two Dartmouth students who were about to fail their classes were given a reprieve, thanks to a smartphone app that detected mitigating mental health issues. According to a recent experiment, this app may be able to detect stress, loneliness, and possibly even clinical depression.
The two students were part of an experiment run by computer scientist Andrew Campbell, who tested the app on 48 Dartmouth students for whether it could gauge their states of mind. He aimed to discover why certain students thrive at the university while others withdraw and begin to perform poorly or even fail their classes. These two students, who had skipped lectures and left assignments incomplete, were facing F's in those classes as well as a semester's suspension, but were spared by the data on their smartphones and given the chance to complete the courses over the summer.
The app, called StudentLife, uses the smartphone's sensors to monitor every physical aspect of the student's life, including number and length of conversations with others, time spent indoors and outdoors, GPS location, physical activity levels, and amount of sleep. The app used the data to draw conclusions about the students' happiness, sadness, loneliness, and stress levels. From this data, the researchers concluded that "flourishing" students tended to have more and longer conversations with others, as well as stable sleep patterns, while depressed or unduly stressed students were socially withdrawn, spent more time alone, and experienced sleep disturbances. They interviewed the subjects daily in order to confirm that the predicted states of mind were correct, and reportedly, the app correctly inferred their mental states. They then looked for correlations between these mental states and academic performance.
"We found for the first time that passive and automatic sensor data, obtained from phones without any action by the user, significantly correlates student depression level, stress and loneliness with academic performance over the term," Campbell said.
They were also able to analyze other aspects of mental health as it relates to the life of a university student, finding that the students generally started off the semester with higher spirits, only to become more stressed and depressed as the workload increased and exams drew nearer. Campbell believes that the app can be used for continuous mental health assessment, and can be used as a therapeutic tool that is potentially more effective than self-reporting questionnaires.
This may very well be true, but there are several social and ethical problems inherent to this app. First, Campbell's aim seems to be to eliminate the need for confirmation interviews, and to gauge mental states purely based on smartphone data. While there are certainly risk factors for depression, taking for granted that any of these factors automatically spell depression is problematic, especially considering that people have different levels of extraversion. There is also a potential privacy issue, considering that the smartphone is taking note of every single physical action that the user takes.
"Privacy is the big issue here," says Cecilia Mascolo, who studies mobile sensing at the University of Cambridge. "You need to constrain this to a very specific application that will benefit people, and with the user always in control of their data."
Via New Scientist