-
-
Story board of the Social Connectome
Inspiration
In neuroscience, structural connectivity refers to the physical wiring of the brain, while functional connectivity describes how brain regions interact and activate together. These networks often differ, and more importantly, they evolve over time through brain plasticity.
We found this concept strikingly similar to human relationships.
In our social lives, structural connections can represent people who are geographically close to us, such as those in the same city, school, or environment. Functional connections, however, reflect how we actually interact. These include who we communicate with frequently, who we maintain strong relationships with, and how those relationships change over time.
Just like the brain, each person's social connectivity pattern is unique. Some people maintain a few deep connections across long distances, while others build broader networks with many people nearby. These patterns also evolve over time as we move through new environments and life stages. This reflects a form of social plasticity.
While brain connectivity can be visualized and studied, our social connections often remain abstract and invisible. We wanted to explore what it would look like to visualize a person's social brain, mapping how relationships form, evolve, and reshape over time.
This led us to create Social Connectome, a visual representation of human relationships inspired by structural and functional connectivity, and the idea that our social networks, like our brains, are dynamic and evolving.
What it does
Social Connectome visualizes a person’s relationships as a dynamic social brain inspired by structural and functional connectivity.
We model structural connectivity using location-based relationships, mapping where people are geographically situated. This allows us to visualize a user's social network across cities, countries, or regions.
We then model functional connectivity using communication patterns such as interaction frequency, duration of interaction, response time, and timing of communication. These factors determine the strength of connections between individuals.
Together, these two layers form a Social Connectome, a visual network showing both geographical distribution and relationship strength.
To further explore social plasticity, our system visualizes how connections evolve over time. Users can observe how their social networks reorganize as they move through different environments, build new relationships, and maintain existing ones.
We also generate a representative social neuron using AI, transforming each person’s connectivity pattern into a unique visual identity.
In addition, we create relationship theme music using ElevenLabs. By analyzing connectivity patterns and interaction dynamics between individuals, we generate a theme song that reflects the tone of their relationships. Strong emotional connections may produce calm and layered compositions, while broader social networks may generate more dynamic and energetic soundscapes.
This creates a multi-sensory experience that allows users to explore their social brain visually and through sound.
How we built it
We built Social Connectome by modeling human relationships using concepts inspired by structural and functional connectivity in neuroscience.
First, we modeled structural connectivity using location-based data. Each individual is represented as a node positioned according to geographic location. This allows us to visualize social relationships across cities and countries on a world map. This structural layer reflects the geographical distribution of a person’s social network.
Next, we developed a functional connectivity scoring system based on communication patterns. Connection strength is determined using multiple factors such as interaction frequency, duration of interactions, response time, communication timing, and holiday interactions.
To guide this design, we drew inspiration from commonly used proxies in social and communication research. These include communication frequency, responsiveness, time spent communicating, network size, and perceived relationship closeness. These factors helped us build a more balanced representation of social connectivity beyond simple interaction counts.
We then visualized these networks using dynamic graph theory-based visualizations. This allowed us to overlay geographical proximity and functional connectivity, creating a Social Connectome that reflects both regional context and relationship strength.
To capture social plasticity, we incorporated time-based changes, allowing users to observe how their social connections evolve across different environments and life stages.
Finally, we used ElevenLabs AI to generate a representative social neuron from each user’s connectivity map. By transforming connectivity patterns into neuron-like visual structures, we created a unique visual identity that reflects an individual’s evolving social brain.
We also integrated ElevenLabs music generation to create relationship-based theme songs. By analyzing connectivity strength, clustering patterns, and communication dynamics, we generate music that reflects the emotional tone and structure of a user's social neuron.
Together, these components create a multi-sensory interactive experience that maps how human relationships form, evolve, and reorganize over time.
Challenges we ran into
One of the biggest challenges was translating complex human relationships into meaningful connectivity. Unlike brain networks, social relationships are nuanced and context dependent. This made it difficult to define structural and functional connectivity in a meaningful way.
Designing the functional connectivity scoring system was particularly challenging. Frequent communication does not always indicate strong emotional or social connection. For example, people may communicate often with coworkers or classmates due to shared responsibilities rather than close relationships. This made it difficult to balance communication frequency with meaningful social connections.
Another challenge was selecting appropriate factors and weights to represent relationship strength while keeping the system interpretable and intuitive.
We also faced challenges when generating the representative social neuron. We wanted the visualization to reflect meaningful aspects of a person’s social identity while maintaining biological inspiration from neuron structures. Mapping functional connectivity patterns to neuron-like structures required careful selection of features and iterative refinement.
These challenges pushed us to carefully think about how to model human relationships and translate dynamic social behaviors into meaningful visual representations.
Accomplishments that we're proud of
We are proud of translating neuroscience concepts like structural and functional connectivity into a meaningful representation of human relationships.
One of our biggest accomplishments is building a dynamic Social Connectome that visualizes both geographical distribution and relationship strength. This allows users to observe how their social networks changes in various environment.
We are also proud of creating a time-evolving social identity visualization, capturing how relationships reorganize and develop across environments and life stages.
Another accomplishment is the development of a representative social neuron generated using AI. This creates a unique and intuitive visual identity for each user's social brain.
We are also proud of building AI-generated relationship theme music using ElevenLabs. By translating connectivity patterns into sound, we extended the visualization into a multi-sensory experience. Users can both see and hear their evolving social identity.
Finally, we are proud of turning abstract social relationships into a visual and interpretable representation that helps users better understand their own relationship patterns and social development.
What we learned
Through this project, we realized that relationships are one of the most important factors shaping personal identity. While aspects such as nationality or background are inherited, our social connections continuously shape who we are.
We also learned that structural connectivity does not always equal functional connectivity. Not having many people nearby does not necessarily mean someone is socially isolated. Strong connections can exist across long distances.
This led us to think about social plasticity, inspired by brain plasticity. Just as neural connections reorganize over time, social connections also shift as we move through new environments and life stages.
We also learned that visualizing social connections **makes these dynamics easier to understand. By observing how **connections evolve, we could better reflect on personal growth and social identity.
Ultimately, this project helped us see identity as dynamic, evolving, and shaped by relationships, much like the continuously adapting networks of the brain.
What's next for Social Connectome
Moving forward, we see Social Connectome evolving into both a personal reflection tool and a research-driven visualization platform.
One potential direction is connectome-based matching, where users with similar social connectivity patterns or complementary social structures can be recommended to each other.
We also see strong potential for research applications. Many social relationship studies rely on bar graphs or tables, but Social Connectome provides a more intuitive and dynamic visualization of social connectivity. This could be used to study how factors such as academic stress, environmental changes, or life transitions influence social relationships.
Another future direction involves expanding AI-generated social experiences using ElevenLabs, including AI narratives that describe evolving social identity.
We also explore generative social music, where a user's preferred music style is blended with the preferences of highly connected individuals to create relationship-based soundscapes.
These future directions aim to deepen the connection between neuroscience-inspired visualization, social identity, and AI-generated creative experiences.
Log in or sign up for Devpost to join the conversation.