The Impact of Social Media on Voting
This article discusses how social networks influence decision-making. It discussed a study of online games where groups of 24 players were split into two teams of 12. After talking to only 5 other players, each player had to vote on which team would win. If a team got 60% of the votes, it won, and its members got $2 and opponents got $0.50. If neither team got 60% of the votes, neither team won money. They found players voted quite differently depending on who they were allowed to talk to, resulting in either compromises or deadlocks. This correlates to how the structure of social networks influence the way people vote in a democracy because one’s social network dictates who they can communicate with.
In the game used to model decision making, a member of team A has two strategies: A1 (vote for itself) and B1 (vote for team B). Similarly a member of team B has two strategies: A2 (vote for team A) and B2 (vote for itself). There are two Nash Equilibria: (A1, A2) and (B1, B2). Since there is no dominant strategy, for either side to choose a best response, it must first predict the other side’s strategy. To study how a side would predict the other side’s move, the experiment paired members of each side with various combinations of members from their own side and the opposing side. They found that when members were grouped in a majority of one side, the dominating side voted for itself, and the minority compromised, polarizing the population. Using this fact, the structure of a network could skew the game to a specific side or a deadlock. As we discussed in class, the network of online information networks shows significantly more edges between sources of a single party than between parties. This isolates the two parties, and the game study implies that isolated teams are less likely to compromise, resulting in deadlocks.