Course Description and Goals: This course is an introduction to connectionist models in cognitive science. The course is organized around a progressive increase in the architectural complexity of the models, rather than around topics of application (see the schedule of topics on a separate page). Applications to perception, memory, language, reasoning, neuro-psychology, etc., will be studied as concrete illustrations of general computational properties of the architectures.

One goal of the course is for you to understand the generic properties of connectionist models, independent of particular applications. Applications to particular topics will illustrate the general computational properties. As a consequence, the behaviors of these applications will be explicated.

A second goal of the course is for you to better understand the role of (connectionist) models in explaining the cognitive phenomena being modeled. In the best models, there is a clear correspondence between (1) the informal explanatory principles for the cognitive phenomena and (2) the formal computational mechanisms in the model. (This is explored more deeply in the more advanced course, Q750.)

Prerequisites:

• Mathematics: Lectures will emphasize mathematical properties, but homework assignments will emphasize hands-on experimentation with models and will not require much mathematical manipulation. Nevertheless, students with a good working knowledge of calculus and linear algebra will get the most out of the course.

• Cognitive Science: Some background in cognitive science (e.g., Q500) is helpful but not required.

• Computer Programming: No actual programming is required. We will be using pre-packaged programs written in C (and possibly others written in Pascal). You will, however, need to edit and manipulate computer files.

Required Textbooks:

• Anderson, J. A. (1995). An introduction to neural networks. Cambridge, MA: MIT Press.

• McClelland, J. L. & Rumelhart, D.E. (1988). Explorations in parallel distributed processing. Cambridge, MA: MIT Press.

  This book comes with software for either Macintosh or PC, but not both; get the version you prefer. Along with two companion volumes published in 1986, this book is often referred to as the PDP book.

Computers: We will be making extensive use of the various campus computers.

• Network ID: If you have not already done so, please activate your IU computer accounts immediately. To do so, take a photo ID to 084 IMU. Even if you already have your accounts activated, it can help to make sure you have a network ID.

• World Wide Web: We will use the World Wide Web for posting announcements, homework assignments, and discussion. Our class home page is
  

   
  http://www.indiana.edu/~jkkteach/Q550/
                                   ^
                                   upper case "Q"
  

• PDP Software: We will definitely be using the programs supplied with the PDP book. Availability on shared Unix computers will be discussed in class.

• J. A. Anderson Software: We might be using software referred to in the Anderson book. This will be announced in class.

Homework: There are eight homework assignments corresponding to the eight main topical divisions of the course (see the schedule of topics). Different assignments will be of different length, and they are worth different numbers of points toward your course grade. The points for each homework will be explicitly marked on each exercise.

Course Grading Method: Course grades are based on performance on the homework assignments. There are no exams and no papers. Grading is "on a curve" insofar as individual grades are assigned relative to the central tendency of all students in the course, but there is no preset shape of the grade distribution. As this is a graduate course, most grades will probably be A's or B's, with C's (or lower) assigned only rarely.

Lecture Notes: Neither lecture notes, nor copies of overhead projector slides, are available. If you must miss a lecture, get notes from a classmate, and then please see Prof. Kruschke or the assistant if you have questions.

Schedule: A detailed schedule is shown on a separate page.

Instructor and Assistant Contact Information is listed on a separate page.

Disclaimer: This syllabus is meant to be suggestive, not absolute. Any and all of the information on this syllabus is subject to change at any time, including exam or homework dates, grading policies, office hours, etc. Changes will be announced in class.