Background:

Based on the characterization of harmful behaviors, understood as preventing others the fulfillment of their behavioral criterion.

Objectives:

This study aimed to evaluate the effect of different harm distributions and amounts regarding the choice of harming others or not.

Method:

An A-B design was used with a baseline and an experimental phase. Thirty-five (35) higher education students participated in the study. They were selected through a non-probability convenience sampling and randomly assigned to one of the seven study groups or a control group. Such study groups differed from each other by the type of harm distribution -i.e., received or observed- and the number of harm notifications -i.e., 1, 5 and 15-. The task was to solve arithmetic operations wherein the participation of three students was simulated. Said simulated students’ role could be to harm or be harmed, according to the group they belonged to.

Results:

The results indicate a significant differential effect among the distributions. Additionally, a tendency towards reciprocity was shown when someone was harmed, which was directly proportional to the harm amount.

Conclusions:

It is concluded that there are multifactorial emergency conditions of harmful behaviors, including the identification of the person who harms others, the score, the moment and the harmed person.