How Overlapping Social Interactions Can Trigger Sudden and Explosive Spread of Trends

Researchers from the London team at Northeastern University have uncovered the hidden mechanics behind why certain trends, behaviors, or even illnesses can spread with surprising speed across communities. Their new work focuses on how overlapping group interactions and one-to-one connections combine to create the conditions for what scientists call explosive contagion—a rapid, almost sudden jump from minimal adoption to widespread uptake.

The research was led by Professor Istvan Kiss from Northeastern’s Network Science Institute in London, with contributions from Iacopo Iacopini and colleagues. Their findings are presented in a paper titled Disentangling the Role of Heterogeneity and Hyperedge Overlap in Explosive Contagion on Higher-Order Networks, published in Physical Review Letters.

This study provides a detailed mathematical explanation for a phenomenon many of us recognize in everyday life: a friend group suddenly switching to iPhones, a classroom adopting the same popular water bottle, or a social media trend appearing everywhere overnight. According to the researchers, these aren’t just cultural quirks—they are examples of how network dynamics behave under certain structural conditions.

Understanding How Group and Pair Interactions Work Together

A major insight from the study is that people don’t interact solely in pairs. They are also part of multiple overlapping groups: classmates, colleagues, friend circles, sports teams, online communities, and more. These groups frequently share members, and these overlapping connections can dramatically accelerate the spread of behaviors or infections.

The team analyzed datasets tracking how individuals interact both in groups and in pairs. They found that when a contagious behavior—like adopting a new app or catching a seasonal illness—passes between two people, it can then spread even further as those individuals enter various group settings. The more groups they belong to, and the more these groups overlap in membership, the faster and more widespread the contagion becomes.

According to the researchers, this overlapping structure creates a scenario similar to a compressed spring. As repeated exposures build up across pairs and groups, the system reaches a point where a small change can trigger a rapid and aggressive spreading event. This is what they describe as explosive contagion.

The Mathematical Framework Behind the Findings

To break down this phenomenon, the researchers used a modeling approach known as Group-Based Compartmental Modeling, which is specifically designed to analyze contagious processes in higher-order networks. In this context, “higher-order” refers to interactions that involve groups of three or more individuals—not just simple pair connections.

The key mathematical elements they examined include:

• Heterogeneity, meaning differences in how many groups each person is part of
• Hyperedges, which represent group interactions rather than pair interactions
• Hyperedge overlap, referring to how much groups share the same members
• Inter-order correlations, which show how pair interactions relate to group interactions

Their calculations revealed that when there is a high degree of overlap between groups—and when people vary significantly in the number of group interactions they have—the system becomes extremely susceptible to explosive contagion. In other words, the structure itself allows for a situation where nothing appears to be spreading at first, but suddenly the entire network experiences a rapid shift.

This helps explain why some trends erupt so suddenly even if early adoption seems slow or minimal.

How This Connects to Previous Research

Professor Kiss and Associate Professor Iacopini have been exploring social contagion for years. A key inspiration for this study was Iacopini’s 2019 paper, Simplicial Models of Social Contagion, which introduced the idea that group interactions can cause sudden jumps from no spreading to widespread adoption. That earlier work showed that the structure of group connections can create thresholds: below a certain level of exposure, nothing spreads, but above that threshold, adoption becomes widespread rapidly.

However, the new study goes much deeper. It identifies the exact mechanics behind these jumps, including how overlapping groups and heterogeneous participation trigger the conditions for explosive contagion. The research essentially provides the detailed blueprint for how and why the phenomenon occurs.

Why This Matters for Understanding Real-World Trends

While the paper focuses on mathematical modeling, its implications extend far beyond theoretical analysis. The findings shed light on how ideas, behaviors, and even diseases spread through real populations. They help explain:

• Why some fashions or tech choices suddenly become universal
• Why certain memes or trends go viral instantly
• How illnesses spread quickly through connected communities
• Why small shifts in social behavior can lead to large-scale change

By understanding the overlapping structures of our social environments, researchers and policymakers can better predict and respond to rapid changes—whether they are related to public health, consumer behavior, or online social dynamics.

Professor Kiss explains that this research contributes to a larger understanding of how people cluster together and how these clusters overlap. When individuals appear in several groups of varying sizes, it becomes crucial to understand how new behaviors or infections can ripple through these structures.

Additional Background: What Is Explosive Contagion?

To fully appreciate the significance of the study, it’s helpful to explore the concept of explosive contagion more broadly.

Explosive contagion refers to a process where a spreading event—such as adoption of a technology, a rumor, or a disease—jumps suddenly from low levels to widespread dissemination. It is not gradual or linear. Instead, the spread accelerates abruptly once the network crosses a certain threshold of overlapping interactions.

This phenomenon is different from classical contagion models, where the spread increases steadily over time. In social systems, explosive contagion often appears as “overnight adoption” or “viral moments,” where a trend seems to come out of nowhere.

The study provides deeper mathematical confirmation of how higher-order interactions and overlapping groups create the perfect conditions for these sudden shifts. Understanding this allows researchers to better model and predict social dynamics in increasingly connected environments.

Additional Background: Higher-Order Networks in Simple Terms

Most traditional network models focus only on pair connections—A interacts with B. But real-world social systems involve multi-person interactions: small groups, teams, families, online communities, and so on.

Higher-order networks capture these multi-member interactions using hyperedges, which connect more than two nodes at once. Modeling these correctly is essential because group interactions can behave very differently from pair interactions. For example:

• Influence within a group can be stronger than between individuals
• Adoption in groups can pressure members more strongly toward conformity
• Overlapping memberships amplify exposure

The research team’s modeling framework brings mathematical clarity to these effects.

Research Paper Reference

Disentangling the Role of Heterogeneity and Hyperedge Overlap in Explosive Contagion on Higher-Order Networks
https://link.aps.org/doi/10.1103/z3d5-94zb