Stanley Milgram has conducted several studies in the 1960s and 1970s in which he tested the obedience of people. His results created controversy among people at the time, but the question is still open. Current experiments have tried to replicate Milgram’s results but they have encountered several ethical problems.
One famous study conducted by Jerry M. Burger 2006 has come up with an ingenious idea to replicate experiment 5 without using the full 15 to 450 shock range. Burger’s experiment stopped at 150 volts, when the first verbal protest of the learner could be heard. Using the fact that 79% of Milgram’s participants who passed the point of 150 volts continued to the end of the generator’s shock range, estimates could be made about how current participants would have continued if they were not stopped. Burger hypothesized that the difference in obedience between Milgram’s 1961-1962 participants and the 2006 participants would be minimal. The methods used were similar to the first experiment.
Participants were found using advertisement and flyers with the message patterned after Milgram’s recruitment notices. A screening procedure was used to select the people that were not familiar with Milgram and did not suffer from any psychological disorders. The final sample consisted of 29 men and 41 women, unlike Milgram’s participants who were mostly men. Their ages ranged from 20 to 81 years. Burger’s experiments consisted of two conditions, the base condition and the modeled refusal condition. The base condition was a replication of Milgram’s experiment, whereas the modeled refusal condition was used to determine whether the level of disobedience would increase if the participants see another person refuse to obey. They examined the percentage of people who continued the procedure after pressing the 150-volt switch. In the case of the base condition, 70% of participants continued with the next item on the test, until stopped by the experimenter. This value is slightly lower than in Milgram’s experiment, where 82.5% of the participants continued. They did not find a statistical significance in the case of the modeled refusal condition either, where 63.3% of the participants carried on with the experiment.
Additionally, Burger looked for differences in obedience between men and women. He concluded that there was no significant difference in the level of obedience of the two genders, although women showed a slightly higher inclination than men to continue in both experimental conditions. Before the experiments, the participants had to take two personality trait tests, which measured their empathic concern and desire for control. Burger was also interested in the relation between these two traits and the participants’ rates of obedience. He found out that the empathic concern scores of both the participants who continued and stopped the experiment was the same in the two conditions. However, in the base condition, participants who had higher desire to control were more prone to stop the experiment. In the modeled refusal condition, the desire to control did not seem to influence whether participants continued or stopped.
These results support the assumption that if participants passed the 150-volt mark, they were likely to proceed until the 450-volt switch. Moreover, Burger argues that “I found no effect for education, age, or ethnicity on participants’ behavior in the study. In short, I am as confident as a psychology researcher can ever be that my findings can be legitimately compared with Milgram’s.” (Burger, 2006, p. 10)
The limitations of this replication experiment are that the screening procedures might have had an influence on the findings, the female to male ration was different, although it has been shown that the level of obedience in both genders are the same. Furthermore, the age interval of Milgrams’ participants differed.
Other experiments that test the rate of obedience have been conducted. Researchers had to come up with different methods because of ethical and moral reasons. The advances in technology have come to great help in what concerns these solutions. Research (Slater, et al., 2006) has discovered that people respond in the same way when interacting with a virtual character, as they do with a real person. The participants had the same subjective, behavioral and psychological responses when administering shocks to virtual humans as if the victims were real. This research opens the door for new methods of dealing with unethical experiments.
A recent experiment (Dambrun and Vantine, 2010) conducted using an immersive video environment studied the level of obedience in people. The researchers manipulated the victim’s ethnicity and degree of visibility; the results showed that when the victim was hidden, the level of obedience was similar to Milgram’s. They also discovered that there are two significant predictors of the level of obedience: state-anger and right-wing authoritarianism. The ethnicity of the victim had an effect on the level of anxiety and distress of the participants. When it was a North African origin, the participants showed a lower level of anxiety or distress than when they had the same racial background; they felt less anguish and conflicting emotions. However, the participants were not inclined to punish the North Africans more than the French victims; although North Africans are a prime target of prejudice and discrimination in France. The author suggests that: “This also seems to confirm that individuals are more ”sensitive” to in-group members’ emotions and feelings than to out-group’ ones” (Slater et al., 2006, p.772), as shown in a variety of studies. (e.g., Chambon et al., 2008; Mondillon et al., 2007). This research also supports the fact that the participants’ mental history was the same as those who obeyed fully in the original experiments of Milgram.
Another experiment (Navarick, 2009) studied the obedience paradigm in the era of informed consent. The researchers used an alternative approach in which they explored the paradigm by creating conceptual analogues in new situations. The participants had to choose between 25 seconds of cartoon followed by 5 seconds time-out or 5 seconds of video followed by 25 seconds time out. In the first 15 minutes, they had to choose the preferred length of the video, while in the next 15 minutes they were instructed to choose the opposite. The results supported Milgram’s (1965a) hypothesis that disobedience increases when it appears to be normative. They also concluded that participants were more likely to withdraw on early trials than on later ones.
Obedience in the context of a TV game show was studied by J.L. Beauvois, D. Courbet and D. Oberle in 2011. The researchers wondered whether the circumstances might not represent an authority that is able to lead the people in the studio to inflict cruel acts on each other. Their results were similar to Milgram’s classic situation. They discovered that the determining factor of obedience was the physical proximity of the television host. The level of obedience in the standard condition (81%) was not statistically different from Milgram’s voice-feedback condition (62.5%). Moreover, a notable result of the research was that in the host-withdrawal condition, their results (28% obedience) were close to Milgram’s “experimenter absence” condition (20.5% obedience). The researchers also found out that the intervention of an assistant who disobeyed did not cause the participants’ level of obedience to drop, a result also present in Burger’s experiment.
Another similarity to Burger’s 2009 experiment is the fact that they did not find a difference between the levels of obedience of men and women. Furthermore, no difference between age groups or socio-occupational categories has been found.