Predicting personality from patterns of behavior collected with smartphones

Abstract

Smartphones enjoy high adoption rates around the globe. Rarely more than an arm’s length away, these sensor-rich devices can easily be repurposed to collect rich and extensive records of their users’ behaviors (e.g., location, communication, media consumption), posing serious threats to individual privacy. Here we examine the extent to which individuals’ Big Five personality dimensions can be predicted on the basis of six different classes of behavioral information collected via sensor and log data harvested from smartphones. Taking a machine-learning approach, we predict personality at broad domain (rmedian= 0.37) and narrow facet levels (rmedian= 0.40) based on behavioral data collected from 624 volunteers over 30 consecutive days (25,347,089 logging events). Our cross-validated results reveal that specific patterns in behaviors in the domains of 1) communication and social behavior, 2) music consumption, 3) app usage, 4) mobility, 5) overall phone activity, and 6) day- and night-time activity are distinctively predictive of the Big Five personality traits. The accuracy of these predictions is similar to that found for predictions based on digital footprints from social media platforms and demonstrates the possibility of obtaining information about individuals’ private traits from behavioral patterns passively collected from their smartphones. Overall, our results point to both the benefits (e.g., in research settings) and dangers (e.g., privacy implications, psychological targeting) presented by the widespread collection and modeling of behavioral data obtained from smartphones.

Testing the Efficacy of a Multicomponent, Self-Guided, Smartphone-Based Meditation App: Three-Armed Randomized Controlled Trial

“…For example, intervention components could be provided in response to passively assessed mood states (eg, using data streams routinely gathered through phone sensors). This would require not only the validation of passive measures [ 129 , 130 ] but also studies that clarify the optimal pairing of intervention components to mood. Microrandomized trials may be an ideal design for this purpose.…”

Descriptive, Predictive and Explanatory Personality Research: Different Goals, Different Approaches, but a Shared Need to Move beyond the Big Few Traits

“…Several recent articles describe how personality and its associations with other variables can be assessed through objectively measured behaviour or digital traces of behaviour (e.g. Cooper et al, 2020;Hall & Matz, 2020;Stachl et al, 2020a;Wiernik et al, 2020). These approaches offer great potential for non-invasively collecting personality-related information about large numbers of people and possibly over extended periods of time, hence allowing measurement of short-term and even longer-term changes in personality.…”

“…Predictive personality research may not only use personality traits as predictors but also as outcomes. A wealth of recent research has explored the possibility to extract personality-relevant information not only from traditional sources like self-reports but also from digital traces that people leave behind such as social media or credit card records, mobile sensor data or diaries (Kosinski, Stillwell, & Graepel, 2013;Stachl et al, 2020a;Weston, Gladstone, Graham, Mroczek, & Condon, 2019;Wiernik et al, 2020). Typically, such data are given psychological meaning by first collating them into scores that approximate self-reported personality traits (e.g.…”

“…As a result, these methods often approximate subjectively rated traits rather than provide entirely new information, and any issues with self‐reports can spill over to their digital approximations (Tay, Woo, Hickman, & Saef, 2020). Currently, typical correlations between self‐reported traits and their digital approximations are in the range from .30 to .40 (Stachl et al, 2020a; Tay, Woo, Hickman, & Saef, 2020), so the gap between them remains non‐trivial. It may narrow as research progresses, though.…”

Beyond Traditional Measures of Personality with BISCUIT and BARE: A New Statistical Learning Technique and Behavioral Item Pool to Push Personality Psychology Forward

“…A major challenge of an EAR-type study is the huge quantity of raw data to be coded, and it would not be feasible for humans to code these data if the time frame were one year. To some extent, machine/statistical learning methods can be used to code phone data, such as summarizing GPS location to identify when a participant has visited a grocery store (Harari et al, 2016;Stachl et al, 2020). However, SLTs will need to advance before they are able to perform tasks akin to text analysis (Iliev et al, 2014;Chen and Wojcik, 2016), turning thousands of hours of video and audio into frequency variables of how often someone has meditated, slapped someone, or had a hangover in the past year.…”