Carlsmith, J. M, & Gross, A. E., Some effects of guilt on compliance, JPSP, 1969, 11, 232-239.

These authors extended the Milgram obedience studies to look at the effect of guilt on compliance.

They set up a Milgram like experiment to ostensibly measure the effect of shock and non-shocks (buzzers) on learning. After the test was over, the confederate/learner/victim asked the subject if he would make some phone calls for an environmental group. The rate of compliance and number of calls the subject was willing to make was recorded.

There were three conditions tested in a matrix -- a) experimenter present/not present during request b)shock or nonshock in study and c)status of the requester high/low.

The results were that experimenter presence and requester status had no effect on compliance, but previously shocking the requester had a dramatic effect on compliance.

The researchers concluded there could be various explanations for this behavior. They could have felt guilty, or possibly sorry for the victim, or possibly that the person wanted to make restitution to the victim.

A second experiment was carried out to determine if any or all of these factors were involved. They added a witness role to the activity and created the following conditions
1. Control - subject is teacher and does not shock the victim. Victim makes request to subject in teather role..
2. Subject is teacher and gives electric shock. Victim asks teacher (restitution condiiion)
3. Subject is teacher and gives electric shock. The witness asks the teacher (generalized guilt)
4. Subject is witness and watches shocks. Victim askes witness (sympathy).

The results were 1=13.0 calls, 2=6.5 calls, 3=23.5 calls, 4=39.5 calls

The guilt condition resulted in the most compliance, while the sympathy condition produces the least compiance. It shows that guilt can lead to compliance even when there is no way to make amends to the vitcim. Another study confirms that people are more willing to comply even if the requester has no knowledge of the "crimes", ostensibly to bolster a person's self-image after the "crime".