The gravitational law of social interaction

Moshe Levy, Jacob Goldenberg

Research output: Contribution to journalArticlepeer-review

Abstract

While a great deal is known about the topology of social networks, there is much less agreement about the geographical structure of these networks. The fundamental question in this context is: how does the probability of a social link between two individuals depend on the physical distance between them? While it is clear that the probability decreases with the distance, various studies have found different functional forms for this dependence. The exact form of the distance dependence has crucial implications for network searchability and dynamics: Kleinberg (2000) [15] shows that the small-world property holds if the probability of a social link is a power-law function of the distance with power -2, but not with any other power. We investigate the distance dependence of link probability empirically by analyzing four very different sets of data: Facebook links, data from the electronic version of the Small-World experiment, email messages, and data from detailed personal interviews. All four datasets reveal the same empirical regularity: the probability of a social link is proportional to the inverse of the square of the distance between the two individuals, analogously to the distance dependence of the gravitational force. Thus, it seems that social networks spontaneously converge to the exact unique distance dependence that ensures the Small-World property.

Original languageAmerican English
Pages (from-to)418-426
Number of pages9
JournalPhysica A: Statistical Mechanics and its Applications
Volume393
DOIs
StatePublished - 1 Jan 2014

Keywords

  • Distance dependence
  • Gravitational law
  • Link probability
  • Searchability
  • Small world
  • Social networks

All Science Journal Classification (ASJC) codes

  • Statistical and Nonlinear Physics
  • Statistics and Probability

Fingerprint

Dive into the research topics of 'The gravitational law of social interaction'. Together they form a unique fingerprint.

Cite this