But how are social networks represented in the human brain?  With mental maps, according to new research led by Oriel FeldmanHall, an associate professor of cognitive and psychological sciences and affiliate of the Robert J. and Nancy D. Carney Institute of Brain Science at Brown University and Apoorva Bhandari, assistant professor of cognitive and psychological sciences at Brown University. 

For the first time, FeldmanHall and her team show that people create mental maps of the connections between people in order to navigate their social worlds. Social navigation, the team found, is similar to spatial navigation.  Nature Human Behaviour published the findings on Thursday, Sept. 19.

"The cognitive process people use to navigate social networks appears to be similar to the cognitive process used by mice navigating a maze,” FeldmanHall said. “We know from decades of research that mice, and humans, build mental maps to understand their physical worlds. People also seem to use maps to make sense of social environments, as well.”

To understand the mental mechanisms behind social map-making, FeldmanHall and researchers in her lab walked a total of 146 Brown University students through a series of brief laboratory exercises. 

Using a computer flashcard game, study participants learned about connections between 13 individuals in a network. They only learned connections between two people and never saw the entire network. They were then asked to choose the fastest way to get a message to a member of the network: Give it to Person A or Person B? 

Subjects were asked to complete this task both right away and the next day, after they slept. Overwhelmingly, the team found that subjects scored higher when tested the next day, even when asked to route a message across longer distances that spanned different communities.

To understand what might be happening in subjects’ minds, the team ran different computational models of learning to characterize the mental maps subjects were making and to determine whether a cognitive process called “replay” was at work.

Replay works as it sounds. During the day, a person may learn that a coworker is friends with a cousin, and that cousin is friends with a classmate – who also knows the coworker. FeldmanHall and her team found that people use this knowledge to begin building, in real time, mental maps that help them learn, process, and store information. These mental maps can be faithful to reality, just as Google Maps reflect the world outside our doorstep. At night, during sleep, FeldmanHall and her team discovered that people explore these maps and revisit them through a replay-like mechanism.

"This finding helps solve a long-standing puzzle in brain science,” said Jae-Young Son, lead author of the paper and a postdoctoral researcher at Brown’s Carney Institute for Brain Science. “How are we able to navigate such a vast and complex social world? This work shows that we create fuzzy mental maps of all the social connections that make up our network. During sleep, our minds are busy simulating how information might flow from one person to another--as if we're mentally taking a walk around the social network--which lets us make better navigation decisions over longer distances."

Their work suggests that during this process, the human brain reshapes those maps, building out new possible social connections. People not only replay the relationships they know exist between the people they’ve met that day, but also map out possible connections those new acquaintances might have with other people they know. 

By generating these new connections, these mental maps no longer are faithful to reality. 

Instead, these denser, fuzzy maps appear to make navigating social networks easier – largely because they reduce the degrees of separation between people.

The study contains all the hallmarks of FeldmanHall’s work – posing an essential question about social behavior then answering it using a diverse set of theories and tools taken from social psychology, cognitive neuroscience, and behavioral economics. 

"Despite decades of active interest, we still know virtually nothing about the cognitive mechanisms that allow people to solve social navigation problems,” FeldmanHall said. “We filled this gap by drawing inspiration from a parallel line of research on spatial navigation in humans and rodents.”

The study was funded by the National Science Foundation (2123469), while a National Institutes of Health award (S10OD025181) supported access to computational resources and services through Brown University’s Center for Computation and Visualization.