Third places visits and well-being: insights from longitudinal passive sensing data

Yu Fang, Kevin G. Saulnier, Jennifer Cleary, Zhenke Wu, Amy S.b. Bohnert, Srijan SenmedRxiv 2026 · DOI

TL;DR

Using passive smartphone sensing and wearable data from 410 medical interns over a year, researchers found that spending time in "third places"—locations that are neither home nor work—is strongly associated with better daily mood. This relationship is bidirectional: visiting third places boosts mood, and a better mood, in turn, predicts more social and geographical exploration the following day.

What Problem Does This Paper Address?

While urban sociologists have long theorized that "third places" (cafes, parks, libraries, gyms) are essential for social cohesion and mental health, empirical evidence has been thin. Most existing research relies on cross-sectional "snapshots" or self-reported surveys, which are prone to recall bias and cannot establish directionality. Furthermore, it has been difficult to determine if the benefits of third places are due to the locations themselves or simply the result of spending less time at a stressful job.

This study addresses these gaps by using longitudinal, objective data to determine if visiting third places actually improves well-being, how much of that effect is driven by physical activity or light exposure, and whether depressed individuals exhibit distinct patterns in how they navigate their environments.

Methods

The researchers utilized a longitudinal cohort design, following first-year physician residents—a group known for high stress and long work hours. They employed several key technical approaches:

  • Passive Sensing (SensorKit): They used Apple’s SensorKit "Visits" sensor to collect anonymized location data. Locations were categorized as ‘home’, ‘work’, ‘school’, ‘gym’, or ‘other’ (the proxy for third places) using a hybrid of automated detection and participant labeling.
  • Multi-Modal Data Fusion: Location data was paired with accelerometer-based step counts (from Fitbits or Apple Watches) and ambient light data (via the SensorKit Ambient Light sensor).
  • Linear Mixed Modeling: To distinguish between between-subject effects (do people who generally go out more feel better?) and within-subject effects (on days I go out more, do I feel better than usual?), the team used linear mixed models with random intercepts.
  • Mediation Analysis: They used structural equation modeling to quantify how much of the mood boost was specifically due to increased physical activity (steps) and outdoor light exposure (illuminance ≥ 1000 lux).

Datasets

The study analyzed 18,795 person-days of data from 410 participants across the United States.

  • Primary Cohort: First-year medical residents (interns) starting in 2023.
  • Mental Health Metrics: Daily mood scores (1–10 scale) and quarterly PHQ-9 (Patient Health Questionnaire) scores to assess depressive symptoms.
  • Location Data: Objective "Visits" data, providing arrival/departure times and anonymized unique identifiers for every location visited.

Key Insights and Ideas

  • The Virtuous Cycle of Mobility: The study found a bidirectional relationship. While visiting a third place improves mood later that day, a high mood score today is a significant predictor of visiting more unique locations tomorrow. This suggests that mental health and environmental engagement form a reinforcing feedback loop.
  • Independence from Work Hours: Crucially, the positive association between third places and mood remained significant even after adjusting for time spent at work. This implies that the benefit isn't just "not being at the hospital"; there is something inherently restorative about the third place itself.
  • The "Depression Gap" in Navigation: Between-subject analysis revealed that depressed individuals (PHQ-9 ≥ 10) spent significantly less time in third places (12.3% vs. 21.2%) and visited fewer unique locations daily compared to their non-depressed peers.
  • The Mechanism of Benefit: Physical activity and outdoor light exposure jointly explained about 19.2% to 27.6% of the association. This leaves a large portion of the effect unexplained, likely pointing toward social interaction and "sense of belonging" as the primary drivers.

Key Results

  • Daily Mobility: On average, participants visited 3.3 unique locations per day and spent 17.9% of their time in third places.
  • Mood Impact: Within individuals, every 10% increase in time spent at a third place was associated with a significant boost in same-day mood ($B=0.013, p<0.001$).
  • The "None" vs. "All" Effect: Participants reported significantly lower mood on days they visited zero third places and significantly higher mood on days spent exclusively in third places (e.g., vacation days).
  • Mediation: Step counts explained 10.7%–18.2% of the mood-location link, while outdoor light exposure explained 13.2%–14.2%.

Limitations and Future Work

  • Causality: As an observational study, it cannot definitively prove that third places cause better mood, though the longitudinal design and lagged analysis (mood today predicting behavior tomorrow) provide stronger evidence than previous cross-sectional work.
  • Selection Bias: The sample was limited to iPhone users who opted into SensorKit, which may not represent the broader population.
  • Unique Population: Medical interns face extreme stressors; the researchers note that these findings should be validated in other demographics (e.g., retirees, office workers) to ensure generalizability.
  • Missing Context: Passive sensing tells us where someone is, but not what they are doing or who they are with. Future work incorporating "digital phenotyping" of social interactions could fill the gap in the mediation analysis.

Summary updated

Abstract

Abstract Spending time in locations outside the home and workplace (termed “third places”), has been linked to better mental health. However, studies to date have typically been cross-sectional, based on self-reported location data and employed small sample sizes, limiting their ability to assess the presence and nature of the association between third places and mental health. To overcome these limitations, we collected 18,795 person-days of objective SensorKit location data passively from a national cohort of 410 first-year medical residents across the United States, to assess visits to third places and their associations with mood and depression over the course of one year. On average, participants visited 3.3 unique locations per day (SD=1.7) and spent 17.9% of their time at third places (SD = 26.5%). Within individuals, both a higher percentage of time spent at third places (B=0.013 [per 10% increase], p&lt;0.001) and a greater number of unique locations visited (B=0.032, p&lt;0.001) were associated with better mood later that same day, independent of the time spent at work. These associations were partially mediated by step counts and outdoor light exposure jointly (19.2% and 27.6%). Reverse-direction associations were observed, with better mood on one day predicting both more time spent at third places (B=0.052, p &lt;0.001) and more unique locations visited (B=0.032, p &lt;0.001) the following day. Between subjects, depressed subjects spent less percentage of time at third places (12.3% vs. 21.2%, t=-3.7, p &lt;0.001) and visited fewer unique places per day (2.9 vs. 3.4, t = –3.8, p &lt;0.001) compared to non-depressed subjects. These findings demonstrate the relationship between visiting third places and well-being, and suggest that interventions and policies aimed at encouraging third places visits have the potential to improve mental health.
BibTeX
@misc{fang2026third,
  author = {Yu Fang and Kevin G. Saulnier and Jennifer Cleary and Zhenke Wu and Amy S.B. Bohnert and Srijan Sen},
  title = {{Third places visits and well-being: insights from longitudinal passive sensing data}},
  year = {2026},
  doi = {10.64898/2026.03.12.26348188},
  howpublished = {medRxiv},
}
Figure