Source-linked AI summary
Understanding metropolitan patterns of daily encounters
Lijun Sun, Kay W. Axhausen, Der-Horng Lee, Xianfeng Huang
TL;DR
The study addresses the lack of large-scale evidence linking individual behavior with collective face-to-face encounters. It uses public-transit smart-card records to construct time-resolved in-vehicle encounter networks, finding reproducible periodic encounters, behaviorally rooted encounter capability, and a citywide small-world structure of co-presence. The authors position these patterns as relevant to understanding collective behavior and diffusion or spreading processes.
Problem
Large-scale evidence describing both individual behavioral regularity and collective physical encounter patterns remains limited, although such encounters matter for temporal spreading processes.
Method
The study analyzes public-transit transaction records to construct time-resolved in-vehicle encounter networks across Singapore’s bus users.
Results
Repeated physical encounters show strong periodicity, encounter capability correlates with behavioral regularity, and repeated encounters form a highly connected metropolitan small-world network.
Takeaways & Limitations
The findings identify a citywide structure of co-presence and support studying collective regularity in social-network dynamics and diffusion or spreading processes.
Takeaways & Limitations
The results are embedded in Singapore’s specific social profile and data, and bus-use data do not portray everyday life in complete detail.
Abstract
from arXiv · showhide
Understanding of the mechanisms driving our daily face-to-face encounters is still limited; the field lacks large-scale datasets describing both individual behaviors and their collective interactions. However, here, with the help of travel smart card data, we uncover such encounter mechanisms and structures by constructing a time-resolved in-vehicle social encounter network on public buses in a city (about 5 million residents). This is the first time that such a large network of encounters has been identified and analyzed. Using a population scale dataset, we find physical encounters display reproducible temporal patterns, indicating that repeated encounters are regular and identical. On an individual scale, we find that collective regularities dominate distinct encounters' bounded nature. An individual's encounter capability is rooted in his/her daily behavioral regularity, explaining the emergence of "familiar strangers" in daily life. Strikingly, we find individuals with repeated encounters are not grouped into small communities, but become strongly connected over time, resulting in a large, but imperceptible, small-world contact network or "structure of co-presence" across the whole metropolitan area. Revealing the encounter pattern and identifying this large-scale contact network are crucial to understanding the dynamics in patterns of social acquaintances, collective human behaviors, and -- particularly -- disclosing the impact of human behavior on various diffusion/spreading processes.
INTRODUCTION
Large-scale evidence has been lacking for how individual mobility regularity and collective physical encounters are jointly structured. Using public-transit smart-card records, the study constructs city-scale encounter data to examine these patterns.
- INTRODUCTION: Physical encounters link people through co-presence in shared spatial and temporal settings, unlike non-physical contacts such as calls, emails, and online networks.Such encounters can involve acquaintances or unknown individuals and may contribute to social contagion as familiarity develops.
- INTRODUCTION: Individual mobility patterns display significant regularity and remarkable predictability despite earlier assumptions emphasizing behavioral variability.This motivates studying individual mobility regularity together with collective interactions.
- INTRODUCTION: Large-scale datasets rarely capture both individual regularity and collective encounter patterns at population scale.Existing proximity datasets generally cover spatially small settings, while larger studies often simulate mobility and behavior separately.
- INTRODUCTION: Daily bus demand shows two prominent peaks from Monday to Friday, indicating collective commuting behavior.The figure uses transaction records to construct an empirical encounter network for a city-scale transit service.
- INTRODUCTION: More than 20 million bus trips from 2,895,750 anonymous users over one week provide high-resolution data for extracting time-resolved in-vehicle encounters.The users represented about 55% of Singapore’s resident population.
RESULTS
Across metropolitan bus encounters, repeated meetings show strong temporal regularity, while individual encounter likelihood is rooted in behavioral regularity. Over time, these encounters form a strongly connected small-world network spanning the city.
- Temporal regularity: 27,892,055 intervals from 18,724,388 pairs show repeated encounters peaking at 24h, 48h, 72h, and 96h, covering about 75% of cases.The pattern indicates strong periodicity in joint encounters.
- Temporal regularity: Repeated encounters can be modeled as a Bernoulli process with success probability P_encounter ≈0.33, contributing to the decline in inter-event probabilities.The model describes repeated encounters over the population.
- Temporal regularity: 85% of consecutive encounters occur 23–25 hours apart, and encounter-time distributions remain indistinguishable across day gaps of 1–4 days.Most recurring encounters happen at approximately the same time of day.
- Encounter heterogeneity: Pairwise collective encounter strength has a power-law tail with exponent β ≈4.8±0.1, whereas individual encounter durations show exponentially decaying tails.This heterogeneity indicates that collective regularities dominate bounded individual encounter durations.
- Individual regularity: Lower behavioral time variation indicates more repetitive transit use, and larger encounter likelihood is strongly rooted in individual behavioral regularity.Encounter capability is measured after rescaling personal encounter weight by total travel time.
- Metropolitan network: ⟨l⟩=2.95, l_rand=2.63, diameter l_max=6, c=0.19, and c_rand=4.5 × 10^-4 characterize a well-connected small-world encounter network.Connections strengthen over time and extend across the whole city rather than remaining confined to one vehicle.
DISCUSSION
The study identifies weak, passive daily encounters that form a metropolitan-scale structure of co-presence, while noting that smart-card data capture only part of social interaction. These encounters reflect regular individual behavior and become highly connected over time.
- The findings are specific to transit use, a socially differentiated activity with limited time allocation and specific locations, and do not capture all social interactions.
- Individuals with repeated encounters form a highly connected aggregated network, whose largest component tracks the presented nodes and whose clustering suggests small-world structure.
- The resulting citywide co-presence structures may support research on collective behavior, social-network evolution, and diffusion or spreading processes.
- Transit data reveal weak, passive, indirectly enabled social links beyond people’s usual circles of friends and acquaintances.