daSilva AW, Huckins JF, Wang W, Wang R, Campbell AT, Meyer ML. Daily perceived stress predicts less next day social interaction: Evidence from a naturalistic mobile sensing study. Emotion. 2021;21(8):1760-1770. doi:10.1037/emo0000994
This experiment assessed the relationship between perceived stress and social interaction using naturalistic observational methods. While animal models of stress and social interaction have been well characterized, the link in humans remains less clear. To assess this, participants’ (n = 88 university students) daily behavior was monitored for roughly two months (M = 67 days). Measures of perceived daily stress were taken using random administration of The Mobile Photographic Stress Meter (MPSM). The MPSM provides participants 16 images, one of which is selected, to describe their current stress level. To determine social interaction in a naturalistic way, passive sensor data was collected from participants’ mobile phones. The StudentLife app was used to collect data on social interaction using microphone sensors and a previously validated conversation classifier. The time spent around conversation, not necessarily engaging in conversation, in a day was used as the measure of social interaction. In addition to this data, information on sleep duration, location, and movement were also collected. Increased perceived stress was negatively related to social interaction. When stress rates were higher, the next day, social interactions were generally lower. In addition, higher daily stress predicted lower movement and sleep levels the next day. The impact of a stressful day on social interaction, movement, and sleep did not extend beyond the next day. Slightly elevated stress levels were seen two, but not three, days after the initial higher stress day. This study provides novel data showing that perceived stress in humans does decrease social interaction the next day. This data also supports naturalistic passive sensor data collection, which can be built upon in future research.