Social Networks and Information Cascades
Link: https://www.pnas.org/doi/10.1073/pnas.2102147118
Commentary:
This research paper, titled “Polarized information ecosystems can reorganize social networks via information cascades”, regards how computational modeling can show how unfollowing can produce homogeneous networks. We usually regard unfollowing as something as breaking edges on a network, however, in this case, research observes institutional polarization causing informational cascades resulting in groups forming that to reject exposure to challenging information and encourage outgroup hostility. By participating in these decisions based off of information, individuals in a social network commonly rewire and participate in information cascades after re-evaluation of internet behavior. This also results in individuals, or as we can observe as nodes, being more likely to adjust their existing connections within the network to avoid information.
How the resource connects to the course:
What is observed here is, in essence, two separate concepts that we’ve observed in class. One was taught earlier in the year and one towards the latter half. We usually regard networks as something that, while possibly flexible, is rigid in formation after establishing existing connections. However, we can see how social networks, something that can be mapped very simply in a diagram are as fluid as any other biological network. We also are familiar with information cascades altering an individual decision but have yet to observe the actual effect of that within a network. In this case, however, we see an example of this spurred through information cascades that affect user behavior and subsequent decisions as well as networks. Polarization eventually leads to homogenous networks of political identities and ideology, and we associate the formation of sorted networks through deliberate actions by nodes that create ties or avoid ties, however, this research shows that polarization in an information ecosystem isn’t completely necessary; sorting can be the result of individual reactions through an informational cascade by rejecting the cascade and unwittingly “self-sort” to create similarly homogenous networks.