A use of PageRank on social thinking studies: Improving the methods of the structural approach to social thinking
This paper proposes adapting the PageRank algorithm to Social Representations Theory as a more robust and accurate alternative to traditional associative degree metrics for identifying the central nucleus of social thoughts, a method validated through an empirical study on students' perceptions of universities.
Original paper licensed under CC BY 4.0 (https://creativecommons.org/licenses/by/4.0/). This is an AI-generated explanation of the paper below. It is not written or endorsed by the authors. For technical accuracy, refer to the original paper. Read full disclaimer
Human beings do not think in isolation; we share a collective mind, a vast, invisible web of ideas that binds a community together. When a group of people discusses a familiar concept, like a university or a job, they do not simply list random facts. Instead, their thoughts form a structured landscape where some ideas are heavy, central, and unshakeable, while others are light, flexible, and changeable depending on the situation. This concept, known as the structural approach to social thinking, suggests that every shared belief has a core, a solid center that defines what the group truly believes, surrounded by a periphery of details that allow the belief to adapt to different people and circumstances. For decades, psychologists have tried to map this mental landscape, using mathematical tools to figure out which ideas sit at the very heart of a group's consciousness and which ones float on the edges. The challenge has always been finding a way to measure the true importance of an idea without being misled by the noise of the data.
In a recent study, researchers Rafael Wolter and João Paulo Costalonga from the Federal University of Espírito Santo in Brazil proposed a new way to solve this puzzle. They turned to a tool originally designed for a completely different purpose: the internet. The algorithm they chose is called PageRank, the same mathematical engine that Google uses to decide which websites appear at the top of a search list. Just as a website is considered important if it is linked to by many other important websites, the researchers argued that an idea in a group's mind is important if it is connected to many other important ideas. By applying this logic to the study of social thoughts, they hoped to create a more accurate map of how groups organize their knowledge, one that could see the true strength of an idea even when it did not have the single strongest connection to another.
To test this new method, the team looked at how 234 high school students in Brazil thought about the concept of a university. The students were split evenly between public and private schools, providing two distinct groups to compare. The researchers asked the students to sort eighteen different words related to university life, such as "effort," "money," "freedom," and "dedication," into categories of how well they fit the idea of a university. From these answers, the team built a network where each word was a point, and the lines connecting them represented how closely the students felt those words were related. Traditionally, researchers would analyze this network by building a "maximum spanning tree," a simplified structure that keeps only the strongest connections and cuts away everything else to avoid loops. This method calculates an "associative degree," which is simply a count of how many strong lines a word has in this simplified map.
The problem with this traditional method, the researchers found, is that it is extremely fragile. It acts like a sieve that lets the most important details slip through. If a word has many good connections but none of them are the absolute strongest, the traditional method might cut them all away, leaving the word looking weak and unimportant. The new PageRank method, however, looks at the entire network at once. It does not cut away connections; instead, it calculates the probability of moving from one idea to another across the whole system. It recognizes that an idea can be powerful because it is consistently linked to other powerful ideas, even if no single link is the strongest of all.
When the researchers applied both methods to the students' data, the results revealed a startling difference in how the two groups viewed their education. In the private school group, the word "effort" appeared to be a minor player in the traditional map, with a low score because its connections were not the single strongest ones. However, the PageRank method showed that "effort" was actually the most central idea in the entire network, holding the highest score of all. It was connected to so many other important concepts that it formed the true backbone of their thinking, a fact the old method had completely missed. Similarly, in the public school group, the word "studying" was relegated to the very edge of the map by the traditional method, appearing as a weak, isolated leaf. Yet, the PageRank analysis showed it was the second most important idea in their minds, deeply woven into the fabric of their collective thought.
These findings suggest that the old way of measuring social thoughts was often blind to the true structure of a group's mind. The traditional method tended to favor ideas that had one or two massive connections, while ignoring ideas that were consistently important across the board. The new approach, by contrast, offers a much more resilient and accurate picture. It showed that while both public and private school students agreed on the importance of "dedication" and "effort," they differed significantly in how they valued other concepts. For instance, the idea of "work" was central to the private school students but peripheral to the public school students, while "future" was a driving force for the public school group but less so for the private one.
This study does not claim to have solved the mystery of human thought, but it provides a sharper lens through which to view it. By borrowing a tool from the digital world to map the human mind, the researchers have shown that the way we measure importance matters deeply. The PageRank method reveals that the core of a group's belief system is often more complex and interconnected than previously thought, hidden beneath the surface of simplified maps. It allows scientists to see the true weight of an idea, not just by how loudly it shouts, but by how deeply it resonates with everything else the group believes. This shift in perspective opens the door to a more precise understanding of how different social groups, with their unique experiences and backgrounds, construct the shared realities that guide their lives.
Drowning in papers in your field?
Get daily digests of the most novel papers matching your research keywords — with technical summaries, in your language.