Psychology 575: Computational Modeling

Spring 2009, T/R 11:00 — 12:15, LSII 285D

This course will survey various connectionist modeling techniques and their application to the field of psychology. Connectionist modeling is a computational method that strives to create models of psychological processes. Thus, this course will be of interest to anyone involved in psychology, artificial intelligence, economics, statistics, cognitive neuroscience, or one of the social sciences.

This course does not involve any programming or calculus — you only need a background in basic statistics and an understanding of simple equations (e.g., knowing what a "sigma" tells you to do). Any other necessary expertise (e.g., linear algebra) will be covered in class. A basic knowledge of psychology is assumed.

Throughout the semester, we will examine a range of modeling techniques including:

• Hebbian learning
• Linear associators
• Backpropagation
• Competitive learning
• Radial basis functions
• Reinforcement learning

An integral part of the course will be the hands-on creation of models. You will build small models and test their behavior. The final project in the class requires the development and presentation of a model (using any tool you wish to use) of a specific problem of interest to you. Teamwork will be allowed and encouraged.