I heard some claims that most participants of Milgram's studies were happy they had done the experiment and were fine years latter, other times I've been told that the participants experienced psychological distress and needed counseling.
Summary: The electric shocks given to "victims" in the experiment were fake. A "scientist" ordered "assistant" participants to apply higher and higher voltages; measuring these participants' willingness to obey authority was the true objective. Some of these participants suffered psychological effects such as emotional discomfort, uncontrolled laughter, and anxiety including anxiety over future life choices such as participating in the military. There are no credible, widely-reported incidents of anyone being institutionalized or committing suicide as a result of the experiment. Nevertheless, the experiment is infamous and helped others advocate for stronger ethical standards in human subject research.
Wikipedia uses the term "inflicted insight" (notice also the mention of deception):
Inflicted insight is a possible consequence for subjects participating in certain kinds of research. It occurs when the subject is given insight into his flaws through his participation in an experiment, often unexpectedly or causing emotional pain. It is especially likely in social and psychological research and especially when that research involves deception of the subject by the researcher.
The Milgram experiment is a well-known example of an experiment with a very high potential for inflicted insight. Through their participation in the experiment, many subjects realized that they were capable of committing acts of extreme violence on other human beings. After having this realization, many subjects experienced prolonged symptoms of anxiety. (However, 84 percent of former participants surveyed later said they were "glad" or "very glad" to have participated.)
I have performed a number of human subject experiments in Economics for the purpose of testing various auction designs (e.g. BICAP 1996), and am familiar with navigating the Human Subject Committee/ Institutional Review Board apparatus that came about as a response to unethical research. Generally, experiments in Economics and Game Theory receive an expedited review because of the limited risk of harm. However, use of deception in an experiment can be an issue.
Deception was an essential element of the Milgram psychology experiment.
For those who are unaware, this infamous experiment involved an experimenter who hires an assistant (but the assistant is actually the experimental subject and does not know that his behavior is the subject of the study) who is ordered to apply electric shock repeatedly and at higher voltages to a supposed subject (who is not really the subject, but actually a research assistant who is not being harmed, but screams and pretends to pass out).
BoingBoing and thesocietypages.org provide similar copies of a "Milgram recruitment flyer". This flyer advertises a payment of $4/hour to participate in a study on memory. When these participants arrived, they were told they would participate in the role of the experimenter's assistant.
The goal was to understand the nature and limits of obedience to authority.
describes the experiment as follows.
During the experiment, a scientist — the “authority figure” — ordered participants to ask another individual a series of questions and administer increasingly painful electric shocks for every wrong answer. The intensity of the shocks started at a level of mild pain when the experiment began but could be built up to lethal doses of electricity as the experiment continued. Unbeknownst to the participant, the setup was fake — there was no real electricity shocking anyone, all other people in the experiment were actors, and the actual purpose of the study was to observe how much pain the participant would inflict under orders. Milgram found that 65 percent of participants administered the final, lethal shock.
Gina Perry, guest blogging at Discover Magazine, disputes Milgram's numbers:
By examining records of the experiment held at Yale, I found that in over half of the 24 variations, 60% of people disobeyed the instructions of the authority and refused to continue.
The negative effects on the subject pool are mostly described as psychological effects:
In 1963, Milgram told the News that the experiment, which used 43 Yalies as participants and took place in Linsly-Chittenden Hall, reduced several “naturally poised” undergraduates to “twitching, stuttering wrecks, on the verge of nervous collapse.” In the process, Milgram said they proved themselves willing to obey people in positions of higher authority, even suggesting that they would agree to drop a bomb or push a button launching an atomic missile.
Milgram tested over 1,000 men from the Yale and New Haven community, some of whom he said fell into fits of “bizarre” laughter and flashed “unnatural smiles” as they pressed buttons marked “Danger: Severe Shock.”
Published in the Journal of Abnormal and Social Psychology, Milgram's article (paywall, abstract visible) also describes in the abstract similar psychological effects on subjects:
26 Ss obeyed the experimental commands fully, and administered the highest shock on the generator. 14 Ss broke off the experiment at some point after the victim protested and refused to provide further answers. The procedure created extreme levels of nervous tension in some Ss. Profuse sweating, trembling, and stuttering were typical expressions of this emotional disturbance. One unexpected sign of tension—yet to be explained—was the regular occurrence of nervous laughter, which in some Ss developed into uncontrollable seizures.
Wikipedia's article on the Milgram experiment mentions a subject who became a conscientious objector when called for military service:
Six years later (at the height of the Vietnam War), one of the participants in the experiment sent correspondence to Milgram, explaining why he was glad to have participated despite the stress: While I was a subject in 1964, though I believed that I was hurting someone, I was totally unaware of why I was doing so. Few people ever realize when they are acting according to their own beliefs and when they are meekly submitting to authority… To permit myself to be drafted with the understanding that I am submitting to authority's demand to do something very wrong would make me frightened of myself… I am fully prepared to go to jail if I am not granted Conscientious Objector status. Indeed, it is the only course I could take to be faithful to what I believe. My only hope is that members of my board act equally according to their conscience…
The experiment incited reform. Reformers advocated the establishment of review boards and human subject committees to provide oversight and peer review of human subject research (as well as help limit liability to universities for poorly thought out projects by denying permission at the review stage).
From the Yale Daily News article:
Milgram’s study incited national controversy and led in part to major human testing regulation reform from Yale administrators and the federal government.
“At the time, we didn’t have ethics committees or even consent forms for these tests,” Dovidio said. “Milgram’s study made people think more seriously about the ethics of research.”
By 1980, Yale had instituted reforms mandating that any experiment using paid subjects receive approval by a six-member Committee for the Protection of Human Subjects, and much tighter rules were put in place limiting the degree of deception that could be used in an experiment, a 1980 article in the News stated.
Milgram's argument (see footnote 3, p. 3 of Milgram, pdf below) that only 1.3% of subjects were sorry that they had participated and that the rest enjoyed or were neutral to the experiment conflicts with the nature of his own study and raises disturbing questions. Would a subject, who felt troubled, complain to the authority about feeling bad in an experiment where the stakes for disobedience/non-compliance during the experiment seemed much higher and demonstrated a degree of callousness towards human life and emotional state? Did Milgram report complaints accurately? If Milgram was willing to deceive his subjects in their participation, was he willing to deceive academia or others to be able to continue his research? Even if he has a reputation for being an honest researcher, does this research strategy still hint of impropriety and unethical behavior? Was the knowledge gained worth doing this research if it also caused emotional pain and discomfort to some non-zero number of people? Is it ethical for the researcher to decide and answer this latter question without any sort of review or should it be peer reviewed or left to the consent of the subject?
It is now common for all human subject research at a university to require pre-approval by a human subjects committee or an institutional review board. These boards operate on the principles of subject consent and protection from harm. Subjects must volunteer, and must know they are in an experiment and the nature of any risks of the experiment. Usually subjects must also be told that they may refer complaints directly to the board. Furthermore, deception is now thought to be a form of research pool pollution, in that there is a perceived danger that subjects who were deceived in previous experiments might not take factual instructions at the word of the experimenters.
Additional reading - Milgram's experiments
Additional reading, re: treatment of human subjects