CMSC 477/677 - Spring 2005

Discussion Questions for Class #10, March 3

Reading: Sloman, "Beyond shallow models of emotion"; Sandra Clara Gadanho, "Learning behavior-selection by emotions and cognition..."

What is an emotion? What's the difference between an emotion, an attitude, and a mood (according to Sloman, at least)?
What are primary, secondary, and tertiary emotions? Is this distinction crisp or soft?
What does Sloman mean by shallow vs. deep models of emotion?
What function do you think emotions serve in human thought and cognition? What are the possible purposes of modeling human emotion? For what purposes/applications are shallow vs. deep models appropriate?
How could you evaluate a model of emotion? (That is, how would you know whether you had a "good" model of emotion?)
Similar to the "intentional stance," we can take the "emotional stance" to ascribe emotions to agents. When and why might it be appropriate to take such a stance?
Can an agent have emotion without having consciousness? (We can leave this discussion for Wednesday, when we talk about consciousness...)
What is the purpose of the horizontal and vertical layers in the CogAff architectures? Where are emotions in this architecture? Do they emerge from the organization of the system or are they imposed onto the organization?

What is a "homoostatic" variable? How does EBII control these variables? What happens if two different homeostatic variables result in opposite pressures on the agent -- i.e., to bring them into balance requires conflicting actions?
This approach seems like it could potentially be a good model for learning within a hybrid architeture. What do you think?
What is the difference between the emotional model in the EB architecture and the "emotion-inspired" goal system in EBII?
The ALEC learning architecture is built on top of EBII, not EB. Why do you think Gadanho made this decision?
Would you characterize the learned rules as reactive? Do you think that the description of the rule extraction process in this paper would be sufficient to allow you to re-implement the approach? If not, what questions would you need to ask before you could re-implement it?
In your opinion, is the model of emotion used in this paper a realistic model of human emotions? Why or why not?
In your opinion, is the concept of "emotion" essential to the design decisions that the author has made, or could she just as well have dispensed with the idea of emotions, and called that part of the system "meta-reasoning"?

Do you think that including emotions in intelligent agents is a good idea? Why or why not?
Do you think that including emotions is necessary if we want to reach human-level intelligence? Why or why not?