Using smartphones for research is nothing new, but how much insight about human behavior can these devices provide? A new study interprets data gathered from subjects' own phones, potentially mimicking randomized trials.
Researcher Fani Tsapeli from the University of Birmingham, along with her colleague Mirco Musolesi from the University College London, demonstrated the use of smartphones in conducting large-scale behavioral studies, with the goal of evaluating what causes increased stress levels in subjects.
They published their findings in the journal EPJ Data Science.
What Smartphones Say of Human Behavior
The researchers argued that smartphones, along with wearable devices, are an indispensable feature of daily life. "Their improved sensing and computing capabilities bring new opportunities for human behavior monitoring and analysis," they wrote.
Previous studies using smartphones focused on finding factors in the phone data-extracted features, but the pure correlation analysis did not offer enough insight into human behavior.
In this study, the researchers used data from StudentLife, a research project at Hanover's Dartmouth College in the United States. Details of the subjects' location taken from raw GPS data were used to determine if they are working or socializing.
The data also included activity levels such as walking, running, or commuting via public transport, as obtained from their raw accelerometer figures.
The researchers found that exercising as well as spending time outside the home or workplace had a positive impact on the subjects' stress levels. On the contrary, reduced working time had little impact on stress.
The results cannot be applied to the general population due to the small sampling. But the approach, according to the authors, has been validated and exhibits great potential for further studies.
Smartphones Know When Users Are Feeling Down
Equipped with potent sensing capabilities, smartphones can signal behavioral changes linked to depression and similar emotions, as found in a study published in the Journal of Medical Internet Research in mid-2015.
The researchers recruited 28 subjects - 14 with depression symptoms and the rest without - to accomplish a health questionnaire. They then gathered data on GPS location and phone usage recorded via built-in sensors on the subjects' phone for two weeks.
Afterwards, the researchers created algorithms estimating some behavioral markers that might be linked to depression, analyzing the association between those markers and the degree of depressive symptoms.
The findings: a couple of behavioral markers - including those capturing length of phone usage and time spent in various locations - greatly correlated with depression scores on the questionnaire.
Depressed participants had irregular movement patterns, meaning they left the house for work at different times every day. They were also less mobile and spent most of their time in fewer places.
Mobile Health Interventions
Smartphones serve as important tools for healthcare as well. Mobile health (mHealth) interventions, for instance, have become integral in national healthcare systems in Australia and many European nations.
mHealth can assist in diagnosis. A patient, for example, can provide a picture of a skin injury to a doctor through a mobile device. In the mental care realm, the tool monitors patients by sending out regular questionnaires about mood and daily activities through SMS or other specific apps.
In the U.S., about 6.9 percent of adults have at least one major depressive phase every year, with over two-thirds of depressed individuals seeking psychological support.
Over 70 percent of them, however, struggle with costs such as transportation, stigma issues, and the lack of motivation to seek traditional mental health therapy.
Photo: Maurizio Pesce | Flickr
