The initiation ceremony experiments

06 October 2015

The subject of highly unsavoury initiation ceremonies has been unceremoniously thrust into polite dinner table discussion around the country, and indeed around the world, by the claim from Lord Ashcroft that UK Prime Minister David Cameron ‘put a private part of his anatomy’ into a dead pig’s head as part of an initiation ceremony for the Piers Gaveston Society at Oxford University.

Regardless of the veracity of this claim, it is clear initiation ceremonies are still widespread in modern society. Data on precisely how common initiation ceremonies are in British Universities has not been reliably collected, but a 1999 study of randomly selected athletes from the National Collegiate Athletic Association in the USA found 98 per cent had been involved in initiations. The harms are clear: at the extreme end there have been countless deaths attributed to initiation ceremonies, though there is no official list, and a list on Wikipedia of “hazing” deaths in the USA shows the variety of different ways that people have died during 'hazings' over the decades. Only last month, 37 people were charged over the death of a student who died after being blindfolded and repeatedly slammed into frozen ground wearing a backpack full of sand. It is clear that initiation ceremonies are not the preserve of any one class or culture. From indigenous tribes to biker gangs, from sports teams to elite student societies, from mainstream religious groups to cults, from fraternities to sororities; throughout humanity initiation ceremonies are everywhere you look, once you scratch beneath the surface.

Much has been written about the ubiquity of unpleasant initiation ceremonies throughout different parts of societies and cultures all around the world. There is plenty of speculation about the different purposes and justifications, far too much and far too varied to begin to cover in any depth here. But what can experimental psychology tell us about the reason people seem to be so drawn to the often disgusting and barbaric activities that are so typical in initiation ceremonies?

Experimental research

One of the earliest psychology experiments to examine the psychological impacts of initiation ceremonies was conducted by Elliot Aronson and Judson Mills in 1959. The experiment was based on the hypothesis that 'individuals who undergo an unpleasant initiation to become members of a group increase their liking for the group; that is, they find the group more attractive than do persons who become members without going through a severe initiation'.

The researchers took 63 female college students and had them engage in what they were led to believe was a group discussion of sex, in which the participants were given privacy through an intercom system. In reality, this was an illusion and the discussion was pre-recorded to ensure every participant in each condition experienced the same things. The participants were first subjected to an 'embarrassment test' designed to emulate an initiation ceremony, which they were told the other participants hadn’t been through and that had just been introduced because the experimenter needed to make sure the participants could talk about sex freely. If the students didn’t pass the embarrassment test, they didn’t get to be part of the discussion group.

During the 'embarrassment test', the students were told they were being monitored by the experimenter for signs of embarrassment such as hesitation and blushing, while they read out a list of obscene words before reading aloud vivid descriptions of sexual activity from contemporary novels. In a 'mild' condition, participants read words that were related to sex but not considered obscene and therefore not meant to be embarrassing.

The students were then told that they would participate in a group discussion on a book titled 'Sexual Behavior in Animals' which the experimenters gave the students the impression they were expected to have read. Of course, none of the participants had read the book in question, so the experimenters told the participants that because of this, they could only listen in on the discussion, but that they were barred from participating.

The discussion itself was designed to be as dull and as banal as possible to make the group members as unlikable as possible: 'The participants spoke dryly and haltingly... "inadvertently" contradicted themselves and one another, mumbled several non sequitiurs, started sentences that they never finished, hemmed, hawed, and in general conducted one of the most worthless and uninteresting discussions imaginable.'

After this was over, the students were asked whether they liked the group members that they’d heard speaking. Aronson and Mills found that those students who went through the researchers’ very own embarrassing initiation ceremony were far more likely to decide that they liked the group members.

Aronson and Mills came to the conclusion that this effect occurred because of cognitive dissonance. According to their theory, a dissonance occurred between experiencing the embarrassing task in order to gain access to the group and the realisation that the group turned out to be a bunch of rather awful people. Rather than conclude correctly that the group members were indeed awful people, the participants in a bid to justify to themselves their participation in the embarrassing task that they just went through, concluded the opposite.

There was, however, a major problem with Aronson and Mills’ experiment and many questions unresolved by their explanation. As you will have noticed, their mock initiation ceremony simply does not even come close to capturing the unpleasantness of the typical initiation ceremony. Furthermore, there are plenty of alternate explanation for their result. Could it be that the explicit tales aroused the women sexually causing them to react differently to the group, than those women who read less sexually explicit words? Or could it be that the explicit stories intrigued the women enough to make them want to join the group more? Perhaps the sexual stories made the women anxious about the group discussion and the discovery that the discussion was completely banal proved a huge relief. Another interpretation is simply that the experience of the group discussion seemed more pleasant following the 'severe' initiation than after the 'mild' initiation, because of a greater contrast between the two experiences.

In order to eliminate these alternative explanations and create a far more realistic simulated initiation ceremony, Gerard and Mathewson (1966) replicated Aronson and Mills’ experiment, upping the ante somewhat. Instead of reading aloud sexual stories to make the participants uncomfortable, the participants received powerful electric shocks. In order to ensure the participants really wanted to join the group and so replicate a genuine group initiation ceremony, half the participants were recruited for 'a discussion club' while half were recruited as normal for a 'psychology experiment'. Half of the participants in each group received electric shocks that were so powerful that the majority of them rated them as 'extremely unpleasant' while the other half received only weak shocks.

The procedure of the experiment says a lot about the era that the experiment was conducted in – it bears more likeness to a sequence from A Clockwork Orange than it does to a modern psychology experiment. Indeed, there have been few psychology experiments that so pushed the boundaries of research ethics, except possibly the Zimbardo Prison Experiment which occurred five years later. The electric shocks were delivered to the participants in the “initiation” group under the guise that they were part of a test to make sure the participants would be objective members of the discussion group (see box A for the justification given to the participants in the “initiation” group).

 Box A

“In the past we have had considerable difficulty with some of the girls who have joined these discussion clubs. The problem is that some people cannot maintain an attitude of objectivity during the discussion. When this happens, naturally the discussion tends to deteriorate and emotions runs very high. In order to avoid this difficulty in the future we have just instituted a screening test to weed out those girls who would tend to let their emotions run away during the discussion. You are the first person to whom we will be administering the test which is a very good one that has been used by psychologists for many years. It consists of determining your physiological reaction to a series of stimuli. We do this by hooking you up to these electrodes [the experimenter shows the subject a pair of dummy GSR electrodes] that detect changes in your skin resistance during the test which is done with the aid of this recorder [the experimenter shows the subject a small strip-chart recorder]. By your response on this chart we can tell how objective you are likely to be under conditions represented by the morals discussion.”

The participants were then subjected to a bizarre sequence of events consisting of being sprayed in perfume from an atomiser located on the ceiling, before being shown a series of paintings by Picasso, Matisse, Klee, Vollard and Roualt, projected on a wall. Participants were then played an audio recording of a shooting sequence from Billy The Kid, before being given a series of three electric shocks each 15 seconds apart. This bizarre sequence of events is probably the closest a laboratory psychology experiment has ever come, and indeed ever will come, to replicating the combination of bizarre and painful activities that can be present in a real initiation ceremony.

After the initiation was over, the participants were informed they had either passed or failed the 'screening test'. In Aronson and Mills’ experiment, all of the participants were told they passed the screening test, making it impossible to say for sure that it was the experience of the initiation and not the satisfaction of being told that they passed successfully that was the true cause for the effects found.

Finally, Gerard and Matthewson’s participants listened to a recorded group discussion from the group they were told that they were going to join, on the topic of cheating in college. Like Aronson and Mills’ experiment, the discussion was intended to be entirely worthless consisting mainly of 'hemming, hawing, clearing of throats and pauses'.

The result of the experiment was that the more severe the electric shock, the higher the rating given by the initiates to both the group members and the quality of discussion, supporting Aronson and Mills’ original finding. Intriguingly, telling participants that they had 'passed the test' actually reduced the initiate’s opinion of the discussion and the group members. According to Gerard and Matthewson, this debunks the 'afterglow hypothesis' – that the pleasure experienced from passing the initiation generalised to the group discussion. The fact that the severe shocks only lead to liking in the initiates, and not those who just thought that they were taking part in a psychology experiment, strongly challenges the relief hypothesis that the participants were simply relieved the initiation was over.

The replication provided powerful support for the 'suffering-leading-to-liking' hypothesis and, the authors argued, ruled out the main possible alternative interpretations of Aronson and Mills’ classic experiment. Furthermore, the electric shock experiment provided far stronger findings than the original experiment, presumably because the experience in this experiment was so much more realistic and unpleasant.

In the years since Gerard and Matthewson’s electric shock experiment, there has been little research that comes so close to imitating a real life initiation ceremony, for obvious ethical reasons. The research that has been done hasn’t been comparable. In 2012 for example, a study (Kamou, 2012) tried and failed to replicate Gerard and Matthewson’s and Aronson and Mills’ findings, by using an initiation ceremony consisting of mathematical subtraction tasks. It is clear this initiation task is simply not representative of a real life initiation ceremony.

Due to the ethical problem of experimentally recreating an initiation ceremony as unpleasant and realistic as those used in the real world, we are therefore heavily reliant on this early research to come to conclusions about the practice of initiation ceremonies.

Will initiation ceremonies ever go away?

One popular answer to the question of initiation ceremonies is to ban them. There may be benefits beyond the avoidance of harm: according to a report by the British Universities Sports Association, Southampon University saw a 27 per cent increase in participation in sport following a ban of initiation ceremonies (and incidentally subsequently reached an all time high score in the university league tables).

Banning them outright however, runs the risk of creating a vacuum and driving them underground, as has occurred in cases in the USA where 'hazing' is still widespread and often extremely severe despite anti-hazing laws in 44 out of 50 US states. In one survey of over 11,000 American college students, of the 55 per cent that had been 'hazed', only 10 per cent actually believed they had been hazed and only 5 per cent reported it. The reluctance to report hazing was driven by social factors, such as not wanting to get others in trouble or be excluded from the group. They also saw participation as a choice, minimised the harms, rationalised benefits and justified it as a 'rite of passage' and a tradition they wanted to be part of. Perhaps they have a point: total bans may eliminate the opportunity for group bonding that it seems we are somehow driven to recreate.

Safe alternatives exist which can fulfil the same aim of drawing groups together without resorting to ritual humiliation. In the book: Making the Team: Inside the World of Sport Initiations and Hazing, Jay Johnson and Patricia Miller suggest a number of alternatives that allow for some of the benefits of initiation ceremonies in a safe environment without causing pain, sexual degradation or harm. The suggestions include games involving problem solving, a little imagination, ropes, blindfolds and team work. Team building games however, don’t solve the problem of the draw of illicit activities, or provide the feeling of belonging that seems to result from enduring an extremely horrible experience. We can therefore probably expect initiation ceremonies to continue well into the future, regardless of efforts to prevent them.

- Simon Oxenham is a science writer based in Bristol: follow @neurobonkers and find more of his work at He is keen to hear of more psychological research on initiation ceremonies.


Allan, E. J. (2009). Hazing in view: college students at risk: initial findings from the National Study of Student Hazing. DIANE Publishing.

Aronson, E., & Mills, J. (1959). The effect of severity of initiation on liking for a group. The Journal of Abnormal and Social Psychology59(2), 177.

Gerard, H. B., & Mathewson, G. C. (1966). The effects of severity of initiation on liking for a group: A replication. Journal of Experimental Social Psychology,2(3), 278-287.

Hoover, N.C. (1999) National survey: Initiation rites and athletics for NCAA sports teams.

Johnson, J., & Miller, P. (2004). Changing the initiation ceremony. Making the Team: Inside the World of Sport Initiations and Hazing. Canadian Scholars’ Press.

Kamau, C. (2013). What does being initiated severely into a group do? The role of rewards. International Journal of Psychology48(3), 399-406.