When designing in a new genre, it is important to learn from the very best. So, as my team started to create Rumors, a social deduction game in which you have to deduce other players’ relationships through rumors, I wanted to study Among Us. Among Us likely needs no introduction, but it is a mobile game for people of all ages in which crewmates, the majority of players, try to detect the imposters, a secret minority, before the imposters kill all the crewmates. I played with five total players, one of whom would be an imposter each round. While the mobile game only accommodates text chat, we were all together in person, so we deliberated out loud. Our game Rumors shares with Among Us a goal of learning information through a combination of information and inference; Among Us provides an excellent example of achieving fellowship- and discovery-based fun by striking this balance.
The structure of Among Us as a cycle between distinct discovery and discussion periods forces players to employ both their deduction skills and evidence-gathering skills. A player who is really good at reading people can’t use their social skills outside of an emergency meeting, so they are forced to use observation in addition to manipulation; on the other hand, a player who really knows the maps and is very sneaky about killing crewmates still has to stand the test of her peer’s questioning in a meeting. Furthermore, the discussion phase promotes crowd-sourced analysis of observations from the action phase. One of the most fun occurrences in Among Us is having your accusation be backed up by someone else’s evidence; in this way, the discussion period is also about gathering and reinforcing information. While previously open discussion has been permitted in Rumors, we are considering adding an explicit discussion phase to force this discussion because, as we learn from Among Us, it provides a lot of information and a lot of fun.
By providing myriad opportunities to get ground truth information but never evidence of that information, Among Us preserves the challenge of social deduction while making that deduction feasible. In Among Us, you can directly watch someone get killed or directly watch someone enter or exit a vent, or you can watch either of these events on the security cameras, or you can even see someone running away from a dead body (which isn’t absolute certainty, but this degree of evidence is also interesting); however, there is no way to prove to the rest of the players that you observed these things. This mechanic brilliantly ensures that deduction is important and, in fact, that convincing people of the truth is important. At least one person knows absolutely a fact that they must try to convince others of. Rumors, as of right now, has limited ways to get absolute information; each player can confirm one relationship with a moderator in the whole game. A problem arises where people have so many possible theories that they don’t know what to try to convince others of or ask others. Among Us presents one solution to overcoming this: giving certain players secret information.
Social deduction is a two way street. There has to be a party who has information and a party who wants that information. Rumors is struggling with balancing these two ingredients and ensuring that players have information to share, but Among Us suggests we maybe have to look no further. Among Us uses its phase-based and observational mechanics to perfectly plant the seeds for a fun collision of certainty and disbelief.
