The blind obedience of others: a better than average effect in a Milgram-like experiment

ABSTRACT

In two highly powered studies (total N = 1617), we showed that individuals estimated that they would stop earlier than others in a Milgram-like biomedical task leading to the death of an animal, confirming the relevance of the Better than Average Effect (BTAE) in a new research setting. However, this effect was not magnified among participants displaying high self-esteem. We also showed that participants who already knew obedience studies expected that others would be more obedient and would administer more damaging treatment to the target. However, knowledge of Milgram’s studies was unrelated to a higher estimate of their own behavior (study 1), and was even linked to the prediction that they would stop earlier (study 2, preregistered). Despite the wide educational use of Milgram’s studies to increase people’s awareness of the risks inherent to blind obedience, it may be that this knowledge only serves to evaluate other’s behaviors, and not oneself.

KEYWORDS:

• Better than average effect
• obedience
• social comparison
• self-processes

Disclosure statement

No potential conflict of interest was reported by the author(s).

Data availability statement

Datasets of study 1 and 2 are available here: https://figshare.com/account/articles/21171331

Notes

¹ This was not actually a real fish, but a biomimetic robot (Airo 9) which was covered in very realistic silicone rubber which had been modeled and painted by a designer.

² The observed number of doses injected across 5 behavioral experiments was M = 7.88 (Bègue & Vezirian, 2022a).

³ We did not include participants who used a device other than a computer, as it was a prerequisite explicitly mentioned in the advertisement of our study, as well as excluding participants who failed the attention check.

⁴ Two other measures were also included in the survey: Need for cognition and Cognitive Reflection Test. Basically, they were unrelated to BTAE. They are presented in the preregistration document available at the following web address: https://osf.io/y5fa4?view_only=b3995259a3e943ad9d70ae1d6d49015c.

⁵ We further explored the role of presentation by examining the range of the BTAE among participants who answered first for themselves before guessing what others would do. The results showed that the lower the participant’s self-esteem, the higher the BTAE, r (488) = −.15, p < .001. However, when the participants had to evaluate others first, there was no relationship at all with self-esteem, r(458) = −.01, p = .73).

⁶ We thank an anonymous reviewer for the suggestion of this promising line of research.

Additional information

Funding

Preparation of this manuscript was supported by Grant [IUF2017] from the University Institute of France to Laurent Bègue, and Grant [MENRT 2018] to Kevin Vezirian.