Back

The Dilemma Of Disobedience: An ERP study

Fabre, E. F.; Causse, M.; Othon, M.; Vanderhenst, J.-B.

2021-07-05 neuroscience
10.1101/2021.07.05.451127 bioRxiv
Show abstract

The present experiment aimed at investigating the decision-making and the associated event-related potentials (ERPs) of subordinates under hierarchical pressure. Participants (N = 33) acted as UAV operators and had to decide to crash their defective drone either on a civilian site killing all civilians present on the site or on a military site destroying military material but preventing any human losses. While in the no-command condition, participants decided according to their own preferences, in the command condition they were ordered to protect the military material at the expense of civilians for undisclosed strategic reasons. The results revealed that in the no-command condition participants almost always crashed the drone on the military site (96%), whereas in the command condition they chose to obey orders and sacrifice civilians to protect the military material 33% of the time. In the command condition, participants were longer to make their decisions, mobilizing greater attentional and cognitive resources (i.e., greater P300 responses) to resolve the conflict between their internal moral values and the orders they were given (i.e., greater N200 responses) than in the no-command condition, where they automatically applied the "you shall not kill" rule. Participants also showed a greater negative affective response (i.e., greater P260 amplitudes) after choosing to disobey than to obey orders. This result suggests that disobeying authority could be perceived as a greater moral violation than obeying and sacrificing civilians, suggesting that individuals may sometimes choose to obey malevolent authority to avoid the negative affective reaction triggered by disobedience.

Matching journals

The top 7 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.