Although humans fancy themselves unique beings and life can seem random, in reality, the collective behavior of societies can be quite predictable. 

That’s why, in recent decades, researchers have begun to apply traditional physics methods, such as data analysis and mathematical models, to study societal phenomena. While physics examines how matter influences each other, social physics examines how humans influence each other. Consider a cluster of particles as an analogy—if one moves and hits another, then that second particle will move and hit a third one, and so on. Social physics suggests this type of chain reaction can be seen in humans as well when considering the spread of ideas.¹

Social physics seeks to understand how collective behaviors such as group decision-making, cooperation, and innovation are sparked by individual actions. Using this lens, we can better grasp how information and ideas flow from person to person, and how these ideas shape societal norms and drive collective action. Researchers can observe individuals in group settings—including workplaces, social gatherings, or public spaces like parks or transit—to draw conclusions on how individual behavior is informed through social interactions.

In the age of big data, this flow of information can be illustrated by the way we interact through social media. Let’s take the example of a popular TikTok video. How does a video go viral in the first place? Well, it starts with individuals liking, commenting, and sharing the video, which spreads it to a wider group of users. Then some of those people in the wider group will share it with their own networks. Through interactions with one individual at a time, the video spreads the content far beyond the original source. In the end, the clip has reached a vast audience all across the internet, and “everyone” has seen or heard of it.

By analyzing individual actions such as shares, likes, or comments, social physics can trace how these small, seemingly isolated actions contribute to large-scale phenomena, revealing how information or behavior—whether that be in-person or online—propagates through society. 

Social Science: The study of how people interact with one another and their behavior within society.

Information Flow: Refers to how behaviors and beliefs spread by way of learning and social norms.²

Social Network: A group of people connected by interpersonal relationships that communicate with one another.

Social Network Analysis: A methodological approach using networks and graph theory to analyze social structures. It involves mapping and measuring relationships and the flow of information between people, groups, organizations, and computers.

Collective Behavior: The behaviors multiple people engage in outside of social norms and institutions, arising spontaneously and driven by group dynamics.  

Big data: A large collection of data in terms of volume, velocity, and variety. This data can be analyzed to understand individual behaviors and the flow of information between individuals and their wider social environment.

In the Age of Enlightenment, a movement in the 17th and 18th centuries, there grew a general interest in understanding how humans interact with each other in relation to the world around them. People began to recognize how individual actions influenced and shaped collective behavior, rather than being isolated from the environment in which it occurs.

In the late 18th century, Henri de Saint-Simon, a French social theorist, laid the groundwork for social physics. He thought that society, just like the natural world, could be studied scientifically and described through physical laws.³ Saint-Simon referred to this approach as social physiology. It wasn’t until 1819 that the term social physics was coined by philosopher Auguste Comte, who had been greatly influenced by Saint-Simon’s work. 

Comte believed that physics was the key to solving some of the biggest philosophical questions about individuals and societies. He defined social physics as “that science which occupies itself with social phenomena, considered in the same light as astronomical, physical, chemical, and physiological phenomena.”⁴ Comte went on to explain that such social phenomena were governed by laws that could be discovered through empirical methods.4 Eventually, Comte replaced the term social physics with sociology to emphasize more of a systematic approach to studying society. 

At this point, Saint-Simon and Comte had introduced the theoretical foundations of social physics and sociology, but this approach had not yet been put into practice. Social physics was finally institutionalized in the 1940s by John Q. Stewart, a Princeton astrophysics professor, who analyzed large sets of geo-coded data to understand human geographical phenomena such as population movements and migration patterns.⁵ 

Then, as the age of big data took off in the early 2000s, social physics ramped up significantly as a method for tracking and modeling social behaviors in real time. MIT professor Alex “Sandy” Pentland made the field popular through his books Social Physics: How Social Networks Can Make Us Smarter and Social Physics: How Good Ideas Spread. In these books, Pentland explains how big data can be leveraged to understand how information flows between social networks, and therefore, how ideas spread and innovation occurs.

As our access to data continues to grow in our modern society, social physics can provide insight into how our individual interactions lead to complex collective behavior. It can provide valuable information for public health, urban planning, and marketing, offering tools to address societal challenges by understanding the underlying patterns of human behavior. As the field continues to evolve, it not only enhances our comprehension of social dynamics but also empowers decision-makers to implement data-driven solutions for a more connected and informed society.

Henri de Saint-Simon

French social theorist who believed that society could be studied scientifically and provide valuable insights into how we can improve such. He referred to this as social physiology, conceptualizing society as an organic entity developing according to natural laws.6 Saint-Simon is also well known for predicting the industrialization of the world, and suggesting that businessmen and other leaders would come to control society as opposed to the feudal system or the military.⁷ 

Auguste Comte

Considered the father of sociology, Auguste Comte was a French philosopher who coined the term social physics in 1819. Comte believed that positivism, a scientific approach that emphasized empirical evidence and objectivity, could be applied to social studies to better understand social order and change. Comte rejected the idea that religion or royalty were the powers that dictated people’s behavior. Instead, he was interested in discovering the forces that hold society together and drive social change.⁸ 

John Q. Stewart

American engineer, astrophysicist, and proponent of social physics, Stewart implemented methods of physics to study human geography. He applied the laws of physics to population movement to better understand migration, likening human behavior to Isaac Newton’s law of gravity, where the “gravitational pull” between two populations depends on their size and distance from one another. Stewart’s theories laid the groundwork for geographic information systems that are used today in economics and urban planning.⁹

Alex “Sandy” Pentland

American computer scientist and MIT professor, Pentland is the author of Social Physics: How Social Networks Can Make Us Smarter and Social Physics: How Good Ideas Spread, which both explore how ideas spread within society and lead to change and innovation. Pentland emphasizes the role of data analysis in understanding culture and society and enacting change.¹⁰ In 2012, Forbes magazine named Pentland one of the most powerful data scientists in the world for his contributions to social physics.¹¹

By applying physics methods to understanding social issues, social physics can provide us with data-driven insights into how people collectively behave in a society. Leveraging big data and technological advancements, social physics analysis reveals patterns in human behavior and demonstrates how individuals impact one another. Revealing these patterns has implications in a variety of fields—policymakers can use the data to identify how misinformation spreads to prevent its flow, city planners can use it to track popular commutes and allocate resources efficiently, and public health officials can use it to track how diseases are transmitted to target interventions.

Public Policy

Policymakers, such as politicians and corporation directors, are tasked with making decisions to positively influence society. Both quantitative and qualitative research about human behavior is instrumental in making evidence-based decisions that impact the wider public. Social physics provides these insights and allows governments to develop policies according to the needs of citizens.¹² 

For example, during an election, researchers can track the spread of misinformation by collecting data from social media and mobile phones. Identifying the origins of misinformation can then help us target communications to those who have been exposed to diminish its impact.¹³

Urban Planning

By analyzing patterns in mobility and communications, social physics reveals how we can improve urban infrastructure, leading to more efficient and livable urban environments.

For example, by collecting and analyzing data about people’s work commutes within a city, social physics can show which routes are most popular and at what times. Policymakers can then use that data to determine public transport schedules. 

Spread of Innovation

While we may believe that innovation occurs thanks to one individual’s genius contributions, social physics shows us that information and ideas flow through our social networks and impact our collective behavior. By mapping societies, social physics reveals how ideas flow, are shaped, and then drive change, including political revolutions, cultural movements, and economic fluctuations. 

An example of how innovation is tracked occurred in an INET study, where researchers analyzed job postings and patents to understand how technological innovations emerged, were adopted, and then impacted jobs.

The study highlighted that technological innovations predominantly emerged in specific regions such as Silicon Valley, New York, Seattle, and Boston, which correlated with the creation of higher-earning jobs. This analysis demonstrates the interconnectedness of innovation, geography, and labor markets, and can help businesses better harness the potential of innovation. For example, they may partner with universities in these regions to establish programs that equip students with technical skills to ensure that there is a pipeline of talent that will be able to maximize the impact of new technologies.¹⁴

Public Health

Social physics plays an important role in understanding and addressing complex issues in public health. When we think about health, our first instinct is likely to examine it from an individual, biological lens. However, the social world in which that body operates can also have a significant impact on our health.¹⁵

For example, using social physics models, public health officials can study patterns of disease transmission. This was incredibly important during the COVID-19 pandemic. Researchers were able to track people’s movement using GPS technology and identify outbreaks in their early stages, to be able to effectively intervene.¹⁶

Ethical & Privacy Concerns

Social physics relies on the collection and analysis of data from personal sources such as social media, mobile devices, and credit cards. This type of personal information often includes sensitive data that can be linked to individuals, raising privacy concerns.  

The mapping and analysis of our social networks can become especially dangerous if in the wrong hands. When third parties gain insight into such data, it can help them spread propaganda and false information that can impact our collective behavior. Social media companies like Facebook and Instagram will often sell users’ data to companies (usually unbeknownst to users), which provides businesses with valuable insights into how to target potential customers in their advertising efforts.

Oversimplification of Human Behavior

Social physics relies largely on quantitative data and thus has been criticized for simplifying human behavior to mathematical models. Human behavior is inherently complex, influenced by a myriad of factors, including cultural norms, individual experiences, and emotional states that cannot always be measured in numbers. Reducing human behavior to statistics may not adequately capture the complexities of human behavior and social dynamics, and insights from social physics may risk being overly simplified. 

In a similar vein, since social physics focuses on collective behavior, this also minimizes the importance we assign to individual action and agency. While we are certainly influenced by our social networks and contexts, many people believe that humans are independent agents that have the ability to make decisions. 

For example, social physics may collect quantitative data from medical records to understand what population (such as age, gender, and geographic location) is most likely to get vaccinated for the flu season. However, this doesn’t take into account people’s individual experience that influences their behavior. If someone has had a negative experience with vaccines, they may choose not to vaccinate even if they fit the demographic profile.

Reality Mining of Social Networks

One method used in social physics is tracking cell phone usage to better understand human interactions. What you may not realize is that every time you call someone, it pings a cell phone tower, showing your location. Mobile phone service providers also record the numbers you dial and the length of your phone calls. Companies do this to get a better understanding of how their consumers use their devices in order to deliver the best service possible.

A team of researchers at MIT collected and analyzed phone data from approximately 100 MIT students and professors to map out cell phone users’ proximity to one another and create a map of social relationships.¹⁷ This data made it possible to predict where individuals would meet with other people on any day of the week, allowing them to draw correlations between someone’s social network, academic performance, and health outcomes.

While this may seem like a scary reality, the information gathered through social physics can be really valuable in informing interventions. By identifying individuals who are less connected, MIT could design targeted interventions, such as mentorship programs. Knowing where people network on campus can also inform campus planning, adding more study lounges in popular communal spaces.¹⁷

Predicting Crime through Social Physics

In 2011, the Los Angeles Police Department began Operation LASER to analyze the profile and behavioral patterns of former felons in order to predict where crime might happen again.¹⁸ By analyzing location data and social networks of past offenders, the LAPD could assess crime “hotspots” and ensure that they targeted resources effectively, which could help them reduce the crime rate. 

In Operation LASER, the LAPD also analyzed past property crimes by exploring which areas, times, and locations property crimes occurred, creating a map that can be updated daily to show 500-by-500-foot hotspots where property crimes were likely to occur. That way, the department was able to send police on patrol in those specific hotspots as a preventive measure.

While the operation aimed to reduce crime through data-driven insights, it raised concerns about biases, such as racial profiling. Using historical crime data can reinforce existing discrimination, as algorithms may focus on communities that have been heavily policed in the past. This can create a feedback loop, leading to over-policing of marginalized groups without new criminal activity. To address these issues, transparency, regular audits of data and algorithms, and involving community stakeholders are crucial to ensure that predictive policing does not unfairly target vulnerable populations.

Harnessing Social Norms for Social Good

Social physics can provide valuable insights on how to promote prosocial behaviors. In this article, our writer Paul Deutchman explores how behavioral science provided useful knowledge on how to promote social distancing during the COVID-19 pandemic.

Social Media and Moral Outrage

At some point, we’ve all likely seen a Facebook post or a Tweet that sparked anger and outrage. In this article, our writer Paridhi Kothari explores the Meaningful Social Interaction model that Facebook uses in its feed algorithm and takes a deep dive to understand why information is more likely to spread if it references ideas, objects, or events typically construed in terms of the interests or good of a unit larger than the individual. 

1. Pentland, A. (2014). Social physics: How good ideas spread – The lessons from a new science. Penguin Press.
2. Pentland, A. (2019, June 14). Data for a new enlightenment. MIT Technology Review. https://www.technologyreview.com/2019/06/14/65678/data-for-a-new-enlightenment/
3. Perc, M. (2019). The social physics collective. Scientific Reports, 9(1), 16549. https://doi.org/10.1038/s41598-019-53300-4
4. Lieb, L. (n.d.). Social physics. Hmolpedia. Retrieved September 23, 2024, from https://www.eoht.info/page/Social%20physics
5. Barnes, T. J., & Wilson, M. W. (2014). Big data, social physics, and spatial analysis: The early years. Big Data & Society, 1(1). https://doi.org/10.1177/2053951714535365
6. Rousseliere, D. (2010). Saint-Simon, Claude-Henri de Rouvroy (Comte de). In H. K. Anheier & S. Toepler (Eds.), International encyclopedia of civil society (pp. 1345-1349). Springer. https://doi.org/10.1007/978-0-387-93996-4_811
7. Britannica, T. Editors of Encyclopaedia (2024, May 15). Henri de Saint-Simon. Encyclopedia Britannica. https://www.britannica.com/biography/Henri-de-Saint-Simon
8. Biography.com Editors. (2023, August 9). Auguste Comte. Biography. Retrieved September 23, 2024, from https://www.biography.com/scholars-educators/auguste-comte
9. Poot, J., Alimi, O., Cameron, M. P., & Maré, D. C. (2016). The gravity model of migration: The successful comeback of an aging superstar in regional science (IZA Discussion Paper No. 10329). Institute of Labor Economics (IZA).
10. Social Science Bites. (2017, January 3). Sandy Pentland on social physics. Social Science Space. Retrieved September 23, 2024, from https://www.socialsciencespace.com/2017/01/sandy-pentland-social-physics/
11. Data Pop Alliance. (n.d.). Alex “Sandy” Pentland. Data Pop Alliance. Retrieved September 23, 2024, from https://datapopalliance.org/bio/alex-sandy-pentland/
12. Carr, D. (2024, February 28). The impact of social science research on public policy: Understand your own impact. Center for Interdisciplinary Studies and Social Sciences. Retrieved September 23, 2024, from https://www.bu.edu/ciss/2024/02/28/the-impact-of-social-science-research-on-public-policy-understand-your-own-impact/
13. Weir, K. (2024, January 1). This election year, fighting misinformation is messier and more important than ever: Psychologists are using science communication to set the record straight. But it’s ugly out there. American Psychological Association, 55(1), 40.
14. Baker, S. (2020, August 24). How do tech innovations really spread? New evidence. Institute for New Economic Thinking. https://www.ineteconomics.org/perspectives/blog/how-do-tech-innovations-really-spread-new-evidence
15. Franklin, J. (2020, March 24). Case studies in social medicine: A review. Somatosphere. Retrieved September 23, 2024, from https://somatosphere.com/2020/case-studies-in-social-medicine-a-review.html/
16. Public Health Agency of Canada. (2022). Insights for public health's use of mobility data. National Collaborating Centre for Infectious Diseases. https://nccid.ca/publications/insights-for-public-healths-use-of-mobility-data/
17. Pentland, A. (2009, February 1). Reality mining. MIT Technology Review. https://www.technologyreview.com/technology/tr10-reality-mining/
18. Lapowsky, I. (2018, May 22). How the LAPD uses data to predict crime. Wired. Retrieved September 23, 2024, fromhttps://www.wired.com/story/los-angeles-police-department-predictive-policing/