Information Gerrymandering and Political Decision-Making
Source: https://www.nature.com/articles/s41586-019-1507-6
Today, people have huge volumes of information on their fingertips. In particular, people can read the news based on their special interests through their smartphones and other technologies. Contrary to popular belief, information does not travel freely; instead, it travels through a social network, which can be altered by zealots and automated bots. People generally vote on their party’s overall interests rather than individually review the stances of each member of their preferred party with multiple, reliable sources, to conform to the group.
This brings us to a phenomenon, information gerrymandering: “the structure of the network can sway the vote outcome towards one party”, regardless of the sizes and influences of the parties (Stewart 1). When a small number of zealots are strategically placed on an influence network, information gerrymandering and biased voting outcomes occur. Fake news disrupts public discourse and decision-making. Social media platforms, such as Facebook and Twitter, are especially at risk because they allow users to silence opposing opinions while remaining anonymous. Distorted information then carries over to “traditional news media and voter behavior” (Stewart 2). For instance, people wonder whether the Russian automated bots and the rise of false information on the Web contributed to Trump’s win in the 2016 presidential election. While Democrats argue that Trump won due to these two factors along with winning the majority of the votes of the electoral college, Republicans assert that Trump won fair and square. All in all, fake news and automated bots, as well as the rise of the polarizing political climate, threaten how people navigate through politics.
Researchers developed a game to study the overall decision given incomplete information. Initially, the researchers assumed that the losing team would rather compromise than to end in deadlock, which is when both parties do not end up making a decision. While most Americans generally have moderate views on issues, people with a zero-sum view (i.e. act like zealots) always vote on their intended party regardless. Researchers first studied how someone updates their vote over time as they gather their political party preference along with information and the wish to avoid deadlock. In the model, a player conveys their purpose to vote for their preferred party “according to a probability that depends on the poll they currently see and the time remaining in the game” (Stewart 5). In this experiment, the influence network’s structure has an equal amount of players from each party from the polls they see, so that the polls represent the samples of the group. The decision process on this kind of network is unbiased because “the expected vote share for each party equals the frequency of its membership in the entire group.” (Stewart 5). This leaves me wondering how biased the players themselves were in this experiment. How effective was the design of the study? Were all the participants willing to vote for their assigned party? Based on the voter game, information gerrymandering still exists even when the bias is minimized as much as possible.
The polarized political climate and echo chambers are essential in public discussion surrounding politics. When interacting with people of various backgrounds and opposing views on controversial issues like immigration, people would hopefully be more open to understanding how other people navigate through the world, which may prevent them from adopting a zero-sum view. A YouTube channel called Jubliee has a series named Middle Ground in which people from opposing viewpoints discuss their perspective on a hot-button topic like the death penalty or abortion while finding common ground with each other.
The intention of the research study was well done; the researchers of this study provided ample background to how information gerrymandering can affect people’s voting behaviors. With the rise of fake news and automated bots that alter our democratic decision-making, we have to remain diligent and check the reliability of each source before forming opinions about the upcoming presidential election or hot-button topics like climate change. While the design of the experiments leaves me wondering whether the researchers should have programmed a simulation of the experiment, they tried their hardest to negate people’s pre-existing political biases. While reading this research paper, I kept on thinking about the 2016 presidential election and questioned whether people will learn from this event. How will social media platforms remove automated bots or detect fake news? Will people learn how to distinguish fake news from real news? Only time can tell until the 2020 presidential election comes around.