Research results that 'connection with other people' may be a hindrance when escaping from crisis such as disaster

Human beings are known to be social animals, and it is important for many humans to communicate with each other and communicate information in order to improve the safety of groups and work efficiency. However, a new psychological study showed that 'connections with others may be a hindrance to escape disasters and other crises.'

Collective communication and behavior in response to uncertain'Danger' in network experiments | Proceedings of the Royal Society A: Mathematical, Physical and Engineering Sciences

Carnegie Mellon and Yale University collaborative teams conducted game-like experiments to see how social groups impact coping with crisis. First, the research team recruited more than 2000 subjects online, 'group of 10 people' 'group of 20 people' 'group of 40 people' 'group of 60 people' We divided each into a total of 108 groups. In addition to these groups, 168 subjects were recruited as 'isolated subjects that did not belong to the population,' and the research team conducted experiments on each group and individual.

The experiment was conducted in a game format of 'deciding whether to evacuate in the situation where a disaster may occur in the near future'. In this game, you will be asked to choose between two actions, 'evacuate' and 'do not evacuate,' based on the information 'safety' and 'danger' sent by other subjects and your own judgment. Subjects will receive $2 (about 215 yen) before the start of the session, but if they choose to evacuate, they will have to pay $1 (about 108 yen). On the other hand, if you choose 'Do not evacuate', you do not have to pay anything.

Each subject is connected to four 'neighbors' as nodes (collection points), and neighbors can see what color each node has. When the subject presses the 'safety' button at hand, the self-identifying node lights up in blue for 5 seconds to inform the surroundings that 'no disaster will occur in this session', and when the 'danger' button is pressed The node lights red and can convey the information that 'a disaster will occur in this session'. Subjects who paid $1 and chose to evacuate would not be able to press any further buttons to convey the information, nor could the other subjects know 'whether the neighbor evacuated.'

In each session, only one of the subjects in the group was informed of the correct 'whether this session was a disaster', and the only four neighbors connecting to this one were who He knew if he had the information. The 'safe' and 'danger' buttons can be pushed not only by subjects with correct information but also by anyone belonging to the group freely, even if the neighbor turns on the 'safe' lamp. Even if they did, it was possible to ignore the lamp and turn on the 'danger' lamp, or to give no information to the neighbor without pressing a button. In other words, in these sessions, someone in the group could intentionally spread fake news.

The game lasted 75 seconds per session, and in the event of a disaster, at the end of the session 75 seconds after the game started, the disaster was informed. The subject could complete the evacuation by selecting the 'evacuate' action by the end of the session, but if the 'evacuate' action was not selected within the time limit, it is determined that the subject remained on the spot without evacuation. It was. The chance of a disaster was adjusted to just 50%.

In this experiment, information such as 'one session lasts 75 seconds,' 'a disaster occurs when the session ends,' and 'the probability of a disaster occurring is just 50%' is not known to the subject. It was. It can be said that the subjects, like the actual disaster, were unaware of the situation such as 'when the disaster will reach you, and how likely is it to occur'.

If 'no disaster occurred during the session', the subjects who did not evacuate earned $2 in addition to 'the number of people who did not evacuate in the same group x $0.1 (about 11 yen)' , The evacuated subjects were able to obtain '$1 of people who did not evacuate in the same group x $0.1' in addition to the $1 left at hand. On the other hand, in the case of 'a disaster occurred during the session', the subjects who did not evacuate lost $2 in hand and earned nothing, and the evacuated subjects replied that they were in the same group as the remaining $1. The number of people evacuated at $0.1' was earned.

In other words, the more people who make the right decisions in the same group, the more money they will earn, which means that subjects who have the correct information will 'have the advantage of pressing the lamp to convey accurate information to their neighbors.' .. On the other hand, the subject does not know that the session will end after 75 seconds, so the session ends while pressing the lamp to teach the neighbors the correct information, and as a result there is also a pattern that it may be delayed to escape. Get

As a result of the experiment, it was found that the number of people who evacuated in both the 'session without a disaster' and the 'session with a disaster' was significantly smaller in the subjects belonging to the group than in the subjects not belonging to the group. In other words, while belonging to a group reduces the chance of unnecessary evacuation, even if some people know the fact that 'a disaster will occur in this session', they will not evacuate as necessary. Have been shown to be more likely to be affected by disasters.

In the experiment, the research team also found that some subjects in the group produced 'fake news'. He also found that the fake news that 'disaster will not occur' was overwhelmingly easier to spread and believable than the correct information that some subjects sent 'disaster will occur'. It seems that the fake news spread was unreasonable because the subjects understood the rules of the game, but the signal of 'safety' was spread over the signal of 'danger' throughout the study. ..

Even in a fictitious situation, people in the group liked to maintain the status quo and tended to avoid active behavior, so the research team said, 'In a sense, interpersonal communication is actually a trade-off for collective security. Could compromise the security of.' 'Social networks may not work well as a way for people to tell the truth that they want to ignore,' as the group is less likely to disseminate inconvenient truths such as 'disasters occur,' the research team said.