As an example of the sorts of social sophistication that agents are beginning to manifest, consider this sample dialog [inset] from the Affective Reasoning project at DePaul University, which took place between a researcher, Elliott, and an Affective Reasoning agent, Sam (figure 1). The agent was embodied as a morphing schematic face (with a wide range of emotion expression), used speech recognition for input, text-to-speech for dialog output, and played musical selections to enhance the expression of emotion. While not interesting from a natural language understanding point of view (the agent has no understanding about anything except emotion states), it does illustrate the elements of history maintenance, some sophistication in social awareness, and the rudiments of understanding about emotion states. (Note that the dialog has been cleaned up somewhat to filter out SR miscues.)
[Begin INSET]
Elliott: I am really furious at Joe.
Sam: Boy what a blameworthy thing he must have done! What do you call such an action?
Elliott: Joe was stealing.
Sam: When Joe was stealing, he must have caused you distress. What is the name of the thing you wanted, or wanted to preserve?
Elliott: My sandwich.
Sam: How important was your sandwich to you, Clark?
Elliott: Not very important.
Sam: How surprising was it that Joe was stealing and caused you distress about your sandwich?
Elliott: It was not surprising.
Sam: How bad was it that Joe was stealing?
Elliott: It was not very bad, actually.
Sam: How important is your principle about stealing?
Elliott: Quite important.
Sam: Hmmm. I would think that you are only somewhat annoyed at Joe. I do not understand that you are furious at Joe. After all, what Joe did was not that bad. Your sandwich is not that important to you. It is not very surprising that Joe was stealing. Perhaps your principle about stealing is very important to you. Is this correct?
Elliott: Yes.
Sam: I will remember that about you.
... later ...
Sam: Clark, if I was stealing and it affected your sandwich, would you still be my friend?
[END INSET]
In this sequence, Sam ``believes'' that anger is caused when some entity performs a blameworthy act that affects one's goals. He uses dialog to discover what the blameworthy act was, and what the goal was. He then asks for information about variables that can affect the intensity of anger. Since these do not match a general pattern for fury, he reasons that one explanation for this might be the greater weight of the principle about stealing, and asks Elliott about this. After confirmation, he updates his internal representation of what is important to Elliott. Later, he draws on this information to ask Elliott an intelligent question about what might happen if their friendship relationship were at odds with a strongly held principle.