Whenever an agent acts on its target by producing a value, estimate, evaluation, praise, or criticism, it checks to see if there is the possibility of a conflict.
For an agent to indicate a same entity conflict, all of the three conditions given below have to be satisfied:
If all these conditions are true, the agent notices the possibility of a conflict and indicates a conflict. The indication of a conflict does not mean that there is necessarily going to be a conflict. Once there is indication of a conflict, the conflict detection task will determine if there is actually a conflict or not.
For an agent to indicate a conflict involving two entities within a parameter block or across parameter blocks, the agent has to know which entities are related to its target, and check the contents of those entities. The agent cannot simply look for identity between its target and the related entities. It should use a simple metric of incompatibility between the contents of the entities in order to decide if there is the possibility of a conflict and whether the more costly conflict detection task should be carried out.