Springer Nature
Browse
13688_2019_200_MOESM9_ESM.pdf (816.34 kB)

MOESM9 of The dynamics of collective social behavior in a crowd controlled game

Download (816.34 kB)
journal contribution
posted on 2019-06-07, 05:00 authored by Alberto Aleta, Yamir Moreno
Tug of war commitment, 2 votes. Meter position of the political tug of war if only votes from committed players (those who sent at least 2 votes throughout the whole game) are taken into account (blue) and if only votes from visitors (those players who only participated once in the voting) are taken into account (pink). The introduction of the voting system was mainly motivated by a puzzle where the crowd had been stuck for over 20 hours with no progress, but it was not the only reason. It was known that later in the game the character would need to get through a special area, the safari zone, where the number of steps one can take is limited to 500. If it goes over that limit the character would then be automatically teleported outside of the area. Voting seemed the best way to go through this area because even though progress can be slow, it is only necessary half of the crowd, at most, to be coordinated. The democracy system also added a new element, the possibility of sending compound commands. Indeed, under democracy mode it was possible to concatenate up to 9 commands, either by writing one after the other or by adding a number which would repeat the previous command that number of times. For example, aleftright would be equivalent to execute a, then left and lastly right. Similarly, start9 was equivalent to pressing start nine times in a row. However, the relevance of these commands in the results was limited. First, because most of the time the game was played under anarchy mode. And second, because the total amount of messages containing those commands is much lower than the ones for the simple ones. As briefly discussed in the main text, in contrast with the ledge event, in this case the behavior of users who sent few commands differs from the ones with several commands. In Additional file 9 we show the hypothetical position of the meter if only users who sent just 1 command are taken into account and if only users with 2 or more commands are taken into account. The position of the meter matches perfectly with this second set of users. If we increase the threshold and consider only users with less than 10 commands as visitors and with 10 or more as committed players, the position of the meter still resembles the results of this second group, even though differences are now noticeable. Interestingly, though, the visitors were clearly in favor of democracy as it can be noticed from the position of their votes and the slight shift downwards of the position of the committed players votes, Additional file 10. (PDF 816 kB)

Funding

MINECO

History