50 years ago, film director John Cassavetes observed that lifestyle is a driver of social division. Our article suggests that this holds true today. However, we add that lifestyle divisions are not random but cluster (and are perceved to cluster) around political identities.

We find that liberals and conservatives behave differently in everyday life, but not as differently as people think. (Observers tended to overestimate lifestyle differences between liberals and conservatives in the same community).

We quantified the association between political identity and 61 social, movement, work, and leisure behaviors collected from smartphone sensors/logs and ecological momentary assessments in a sample of students on a college campus.

osf.io/preprints/ps...

This work suggests that to understand authoritarian movements upending the economic, social, and political fabric of countries around the world, researchers have much to learn from the inherently quotidian—the stuff of everyday life.

Regarding the most important individual behaviors for predicting authoritarianism, Facebook app use was king. It comprised 5 out of the 10 most important behaviors overall. It was also the most important type of app use, and app use was the most important type of sensor modality.

AGGRESSION & EMOTIONALITY: Authoritarians used positive and negative emotion words more often and with more variability, and they used anger- and death-related words more. The exception was anxiety! Those LOW on authoritarianism used more anxiety words.

AUTHORITY & HIERARCHY (2): Gender-related hierarchy seemed to play a particularly important role here. Individuals high on authoritarianism used more male references (e.g., he, him, man), suggesting a more male-dominated social circle and/or a focus on men and masculinity.

AUTHORITY & HIERARCHY (1): Authoritarians’ language revealed a preoccupation with status (e.g., using more clout-related words) and a tenuous relationship with submission (e.g., greater variability in assent-related words).

[In the beeswarm plots on the right, positive SHAP values indicate the behavior predicts higher authoritarianism; negative SHAP values indicate the behavior predicts lower authoritarianism].

LIMITED EXPOSURE: Authoritarians have an active but constrained social circle (e.g., receiving more text messages from fewer numbers), an interest in familiar pop culture (e.g., listening to Europop music), and a focus on intergroup dynamics (e.g., “they”-related language).

Behavioral patterns reflecting limited exposure to unknown people/places/cultures were most important for predicting individuals' authoritarianism, followed by preference for simple information, respect for authority & hierarchy, and tendency towards aggression & emotionality.

Everyday behavioral patterns measured with smartphones predicted individuals’ authoritarianism twice as accurately as demographics (i.e., age, gender, education, nationality).

We used prior literature to derive theory-informed behaviors that a) reflect authoritarians’ key psychological attributes, and b) can be measured with smartphones sensors/logs (i.e., app usage, keyboard typing, music listening, unlocking/locking, calling/texting, GPS location).

Thanks for your questions! More details in the pre-print: osf.io/k4d5q_v1. We didn't measure wellbeing in our study, but liberals tended to engage in behaviors that prior research has found to be less beneficial for wellbeing, a finding we are following up on. The "fun" comment was meant as a joke.