Even physics proves the enemy of your enemy is your friend

EVANSTON, Ill. — In our hyperconnected digital age, online social networks have become an integral part of how we interact and relate to one another. From Facebook friendships to heated exchanges on X (formerly Twitter), these virtual connections mirror the complexities of real-world relationships. But what if the key to understanding the intricate dance of online interactions lies in a decades-old theory?

Enter structural balance theory, a concept from social psychology first proposed by Fritz Heider in the 1940s. The basic idea is simple: in social groups, people tend to form relationships that create a sense of balance or harmony. “The friend of my friend is my friend,” “the enemy of my friend is my enemy,” and “the enemy of my enemy is my friend” — these common sayings capture the essence of balance theory.

However, applying this elegant concept to the tangled web of online networks has proven tricky. Despite the widespread belief that virtual communities tend toward balance, the conclusions drawn from network analysis studies have been inconsistent and even contradictory. However, a new study published in Science Advances and led by researchers Bingjie Hao and István Kovács from Northwestern University may finally help clear the air.

The key, according to Hao and Kovács, lies in how we randomize network data to assess the statistical significance of balanced relationship patterns. Imagine you’re dealt a hand in a card game. To determine if your hand is especially good or bad, you’d want to compare it to many other possible hands that could have been dealt from the same deck. That’s essentially what network researchers do – they generate randomized versions of the actual network to see if the observed relationship patterns occur more frequently than mere chance would predict.

The problem is that the randomization techniques used in most studies neglect two critical features of real social networks. First, they fail to account for the fact that some people are inherently more “friendly” or “hostile” than others, with a higher proportion of positive or negative connections. Second, they often disrupt the underlying structure of the network in the randomization process.

To illustrate the importance of preserving both of these features, consider a hypothetical social network in a high school. The popular kids may have mostly positive connections, while the school bullies exhibit predominantly negative ones. Randomization methods that ignore these individual tendencies might conclude the network lacks balance, even if the popular kids all like each other and dislike the bullies and vice versa. Additionally, if the randomization breaks up tightly-knit social circles in the process, important structures contributing to balance can be lost.

Sad or jealous girl holding phone while friends talk behind her back
Is the enemy of my enemy really my friend? A new study finds this way of thinking still applies even in the digital realm. (© Photographee.eu – stock.adobe.com)

The solution proposed by Hao and Kovács is a new network randomization method called signed degree and topology preserving (STP) randomization. The STP approach maintains the exact web of connections while shuffling the positive and negative signs in a way that, on average, matches each person’s ratio of friendly to hostile relationships. The researchers put their method to the test on several large-scale online social network datasets, including Bitcoin trust networks and the Slashdot friend/foe network.

The results were striking. With the STP randomization, the online social networks consistently exhibited significant patterns of structural balance, even in more complex four-person relationship configurations. This contrasted with the mixed and often paradoxical conclusions drawn from previous randomization methods. The study even found balance in a model network specifically designed to be unbalanced, highlighting the flaws of earlier approaches.

So what does this all mean? On a basic level, it suggests that perhaps we’re not as different online as we sometimes imagine. Just like in face-to-face interactions, we still seek out harmonious groups of friends and like-minded allies while distancing ourselves from those we perceive as foes. The new study provides a rigorous mathematical foundation for these age-old social instincts.

However, these implications go deeper. By identifying consistent signatures of structural balance across diverse online communities, the findings hint at potential “wiring mechanisms” that guide the formation of digital connections. One intriguing possibility proposed by the researchers is an “edge-copying” mechanism, where people adopt the friendship or enmity patterns of their friends (copying positive connections directly and negative ones in reverse). Further research into these network-building processes could shed light on how echo chambers and polarization emerge in the online world.

Beyond the study of online behavior, the STP randomization method could have far-reaching applications in other fields that employ network analysis, from uncovering genetic interaction patterns to mapping brain connectivity. By faithfully preserving key structures, this approach may help reveal hidden architectures and dynamics in complex systems across the scientific spectrum.

The new study also underscores the importance of revisiting established concepts as our social landscapes evolve. While the core principles of balance theory still resonate in the digital realm, the tools we use to explore them must adapt to the unique qualities of online interaction. With more sensitive and nuanced approaches like STP randomization, we can begin to map the subtle topologies of our virtual ties.

So, the next time you navigate the sometimes turbulent waters of an online community, take heart in the knowledge that you’re part of an ancient human dance toward social equilibrium, now playing out on a digital stage. As researchers continue to refine their methods and uncover the governing principles of these digital networks, we may gain a richer understanding not only of our online selves but of the fundamental social fabric that binds us all.

StudyFinds Editor-in-Chief Steve Fink contributed to this report.

Follow on Google News

About the Author

StudyFinds Staff

StudyFinds sets out to find new research that speaks to mass audiences — without all the scientific jargon. The stories we publish are digestible, summarized versions of research that are intended to inform the reader as well as stir civil, educated debate.

The contents of this website do not constitute advice and are provided for informational purposes only. See our full disclaimer