# Social Physics **Entity class:** Computational social-science framework ## Definition **Social physics** is a framework associated with [[wiki/Alex Pentland|Alex “Sandy” Pentland]] for measuring how ideas, behaviors, and influence flow through social networks and how those flows produce collective outcomes. It uses digital traces, sensors, network structure, mathematical models, and machine learning to study populations from small groups to cities. For this collection, social physics is important because it makes explicit the move from isolated individuals to patterned interaction: communication, mobility, exposure, reinforcement, diffusion, and behavioral change become observable features of a dynamic network. ## Relationships - **developed within:** [[wiki/MIT Connection Science|MIT Connection Science]] and MIT's Human Dynamics research. - **methods:** [[wiki/Computational Social Science|Computational Social Science]], [[wiki/Network Science|Network Science]], [[wiki/Behavioral Sensing|Behavioral Sensing]], [[wiki/Information Diffusion|Information Diffusion]], and [[wiki/Population Modeling|Population Modeling]]. - **counterterrorism bridge:** [[wiki/Complex Contagion|Complex Contagion]] and [[wiki/Counterterrorism Predictive Graph|Counterterrorism Predictive Graph]]. ## Sources / Provenance - [MIT News, “Social physics,” March 4, 2014](https://news.mit.edu/2014/social-physics-0304). - [MIT News, “Inventing the ‘Google’ for predictive analytics,” December 20, 2017](https://news.mit.edu/2017/endor-inventing-google-predictive-analytics-1220). **As of:** 2026-09-23