The First Experiment

In the early 1960s, a number of experiments were conducted by Lerner and his team at the University of Kentucky.

Female students were asked to participate in a study evaluating team play in group situations. The women were divided into groups of five on one side of a one-way mirror. On the other side of the one-way mirror, they could see a table with two chairs. On the table was a microphone and some cars with letters on them. Once they had settled in and had time to take in the scene, the mirror was blocked by a curtain. The women were told

The experiment was…a project to develop “some reliable and meaningful scales” which could be used to assess “the contributions of individual members to the solution of problems which face the group.” As the subjects were informed at the outset, this had become a serious problem in all sectors of our society, since most of the decision making and productivity, at least in managerial and professional levels, occurred in “teams,” or task forces. With the give-and-take interaction involved in these efforts, it was difficult to assess, as every organization must, the relative talents and contributions of each member. These must be known in order to make decisions about who should be promoted, who should be retained, and who should be let go to do something else for which he is better suited. Presumably there were no techniques presently available to make these assessments, so we were employing a “grass roots” approach, in which a number of people would observe a variety of others working in various intellectual group tasks. The observers during the first phase were merely to “develop impressions and hunches concerning what they observe, then by a process of pooling these various individual impressions I can begin to develop some categories to be used in more specific judgments.” (Lerner, 1980, pp. 31)

When the curtain was drawn back, the women witnessed two people – “Tom” and “Dave” – attempting to form words from the cards together; the more words, the better. “Tom” and “Dave” tried to create the words using different approaches but there was no clearly superior approach: both “Tom” and “Dave” attempted to create as many words as possible.

Once they were finished, “Dave” was paid.

But for half of the women watching, it was “Tom” who was paid.

The payment was entirely arbitrary: “Tom” was paid exactly 50% of the time and “Dave” was paid exactly 50% of the time.

The independent variable in this experiment was the “fate” of the these two workers in terms of receiving payment. The assignment of the important outcome – large payment – to one of the workers had to be clearly arbitrary, but also legitimate in the sense of violating no moral standards…it was expected that our subjects would try to persuade themselves that the worker favored “by chance” actually performed better, actually deserved the pay after all.  (Lerner, 1980, pp. 32)

This is exactly what happened: the female subjects retroactively assigned worth to whoever was arbitrarily awarded the money.


The Second Experiment


Witnesses of an innocent victim’s suffering will attempt to reestablish justice in the situation by compensating the victim. If they are unable to provide compensation, they will attempt to reestablish justice by finding the victim blameworthy, as a function of his/her actions or personal characteristics. (Lerner, 1980, p. 40)

If can’t help a victim, you will blame the victim.

The experiment:

Female undergraduates taking psych 101 participated in a study of “cues of emotional arousal” to fulfill the program’s requirement of participating in a psychological study.

The women again sat in rooms with one-way mirrors, this time in groups of 4 to 10. A woman dressed in a white lab coat “introduced herself as the Research Associate” conducted the experiment.

The Research Associate told the women that the experiment was trying to catalogue a set of “behavioral cues” that would indicate in a reliable way how upset another human being was. They were to look for to look for these non-verbal cues so they could be provided to supervisors, healthcare professionals, teachers and parents so that they, in turn, could “determine whether psychological first aid was necessary.” The hope was to avert psychological crises in all walks of life.

The women were to watch someone undergo a “known degree of stress” and they would try to see signs of these cues in that person’s behaviour – looking for physical movements and the like. The women were told that, by combining all of their observations together, the common behaviour cues would become known, which would serve as the basis for future study.

After these preliminary instructions, the curtains covering the one-way mirror were opened to show the test room, where a memory drum was seen on a table with two chairs drawn up to it. Dr. Stewart was observed “adjusting” the shock equipment and electrode leading to the memory drum. A technician was also adjusting the television camera. Carolyn then explained to the observers that the curtain would be closed once more, and they would observe the learning experiment over the black and white TV monitor in front of them. (Lerner, p. 41)

The subjects were told one of a few different scenarios:

  1. The experimenter only needed one session to decide what works as “negative reinforcement” and the first group of subjects would vote on what the next session would be but regardless of the vote it would be a “positive reinforcement session.”
  2. The second group of subjects were told the same thing but were not given the (false) outcome of the vote.
  3. Other groups of subjects were not even allowed to vote and were told either that the punishment was over or that what they had witnessed on the TV monitor was not actually live.
  4. Finally, one group heard the experimenter speaking to the victim, where the victim in the experiment would only suffer punishment so that the subjects could get credit for their class.



The Third Experiment

  • people get frustrated by witnessing suffering
  • reaction plays a role in judgment
  • if the observer thought they had a chance of suffering, it creates empathy.
The experiment:

In the third experiment, the video tape showed the victim entering the room receiving her instructions and performing “rather well” on the task.

Just as with the second experiment, different groups of subjects were exposed to different situations, for example:

  • the victim again said she was taking the shocks for the benefit of the subjects OR
  • the subjects learned that the victim wasn’t actually betting shocked.

The results were pretty much the same as the second experiment:

When people in our culture believe that there is a situation of direct or indirect competition for a scarce resource, and if everyone has an equal claim to the desired outcome, then there are fair ways to decide who is to “lose” and who is to “win”. And the winners need not feel guilty, or see the losers as “victims.” Of course, one of these ways is to leave it to “chance,” or the variant we know of each person drawing straws or picking a number.” (Lerner, 1980, pp. 60-61)

Three more experiments were conducted to verify the results of experiments 2 and 3.