Society of Mind

Assignment 3: Levels of Mental Activities + Project Proposal

Note: This week's assignment has two parts with separate due dates. In the first part, you are asked to submit responses to Chapter 5 of The Emotion Machine (due by 6am Wednesday). In the second part, you are asked to submit some preliminary ideas for your final project (due by 11:59pm Friday).

Part 1: Levels of Mental Activities

In Society of Mind, Prof. Minsky proposes that minds are produced by interconnected networks of unintelligent processes, and in Chapter 5 of The Emotion Machine, he suggests a hierarchical scheme for organizing some of those processes.

After you have read Chapter 5 of The Emotion Machine, please send an e-mail to the staff (society-of-mind@mit.edu) containing at least two of your questions or criticisms about the content of the chapter that you would like to be addressed in lecture on Wednesday. Your responses should be thoughtful, clear, and important. Your e-mail is due by 6am Wednesday morning (20 Mar), although you may submit it earlier. We'll use these responses to direct and shape the course of the lecture.

(Note: If you have remarks about other chapters of The Emotion Machine or Society of Mind which you would like to be discussed in lecture, please feel free to include them along with your remarks on this chapter.)

Resources:
You may find the following resource and prompts helpful for this assigment; they are not required.

• In this paper on cognitive architectures, Aaron Sloman presents many ideas that overlap with those discussed in Chapter 5. How do the cognitive architectures of Minsky and Sloman differ from each other? What useful insights do you gain from each?
• What are Critics, and how do they work? How might the Critics paradigm help to explain the Joan example from the previous chapter? What about how a low-level stimulus (like pain) can override a high-level process (like formulating a strategy)? What about how a high-level process overriding a low-level one (as when one's principles override one's instinctive drives)?
• From an evolutionary standpoint, is it more reasonable to suppose that brains are highly structured and arranged in neat layers, or that brains are a mostly unorganized amalgam of patches and hacks?
• What do you think of Prof. Minsky's six-layer model of the mind? Is it a useful conceptual tool? Does it make the wrong kinds of distinctions? Too many distinctions? Too few? How could it be improved or extended?

Part 2: Project proposal

• Our minds are built of many different resources and strategies for using them.
• We learn more from interacting with others .
• Emotions are different Ways to Think.
• Suitcase words inhibit clear understanding.
• We learn to think about our recent thoughts.
• We learn to think on multiple levels.

• The main requirement for your final project is that it embody the ideas of the class, such as the theories in Society of Mind and The Emotion Machine or the strategies for studying the mind as an architecture built of many interoperating processes.
• In satisfying this requirement, you have many options; for example, in past years students have used their final project to:
  • Develop a theory about how some aspect of the mind works (such as the fear response, or the process of creating music.)
  • Design a computer program that exhibits some feature of intelligent behavior, and analyze the results.
  • Compare theories of a particular phenomenon (such as altruism or addiction) given by different people or in different fields (such as psychology, sociology, neuroscience, artificial intelligence, evolutionary biology, philosophy, or everyday life), and explain how the ideas in 6.868 help to explain the phenomenon.
  • Criticize or build upon a theory from Society of Mind that could be usefully extended, adjusted, or implemented concretely.
• You may submit more than one project idea in your proposal. For each idea, please include a clear and concise description of a problem you want to solve, and what general strategy you'll use to solve it (e.g. experimentation, research, programming).
• You may collaborate with others on the project, provided the amount of work involved is proportional to the number of people in the group.
• This initial proposal is not final or binding— you may change topics later. Our aim is to help you to identify which aspects of intelligence interest you and to incorporate them into a valuable final project.
• If you have any questions about the final project, or would like assistance with thinking of or evaluating ideas, feel free to e-mail the TAs (dxh@mit, rlm@mit); we are happy to help.