Winter Term - 2017
Prerequisites: COGS 100/3.0 or COGS 201/3.0, PSYC 100 or PSYC 221/3.0, basic linear algebra, statistics and software programming. Recommended prerequisite CISC 101 (Matlab).
Farhana H. Zulkernine, PhD, PEng
756 Goodwin Hall, School of Computing, Queen's University
Kingston, Ontario, Canada K7L
E-mail: Farhana at cs dot queensu dot ca (firstname.lastname@example.org)
Computational cognitive models provide effective and useful representations of the complex and detailed theories of cognition and enable better understanding of these theories by computer simulation of cognitive processes. Models can be used to predict a cognitive behaviour given some known parameters. Consequently, they may provide detailed interpretations and insights of cognitive processes that no other experimental or theoretical approach can provide. As research continues to unfold the unknown, existing models are refined continuously and new models are defined. Different modeling techniques are used to create the models, and a single process can be represented using multiple models some of which may perform better than the others depending on the data domain and distribution.
General outline of the course includes the following: Importance and challenges of building cognitive models; steps of model building, programming simple models using some of the existing computational and statistical techniques and tools such as Matlab, and learn some of the important cognitive models and some recent models from research publications.
In this course, the students will get practical knowledge of model building starting with very simple models using easy-to-use tools and techniques (such as Matlab) and leave with the knowledge of some of the well-known and important cognitive models and some new more recent models. The course would lay the necessary foundation for developing and understanding complex cognitive models.
Course Learning Outcomes: http://www.cs.queensu.ca/students/undergraduate/CLO.php#COGS300
COGS 300 W/3.0: Programming Cognitive Models
Importance and challenges of building cognitive models; steps of model building, choice of parameters given the goal, performance evaluation by measuring fitness of the model to the actual data and refinement of the models.
|Unit 2:||Program simple cognitive models using computational and statistical techniques and tools such as Matlab, and other programming languages, extend existing models, and observe the effect of parameter manipulation and optimization techniques on the performance of the model.|
|Unit 3:||Read and present some of the important cognitive models, both old and recent models, from the literature.|
Class time and location
|Days & Times||Room||Dates|
|Mon 11:30pm - 12:30pm||JEFFERY RM 102||Jan 09 - Apr 07|
|Tue 1:30pm - 2:30pm||JEFFERY RM 102||Jan 09 - Apr 07|
|Thu 12:30pm - 1:30pm||JEFFERY RM 102||Jan 09 - Apr 07|
Thursday 1:30pm - 2:30pm or request a meeting at another time if you cannot make it at this time.
Paper Presentation and discussion 20%
Details about the course content can be found at the Moodle website (https://moodle.queensu.ca/community/).
Computational Modeling in Cognition
by Stephan Lewandowsky and Simon Farrell
Cognitive Modeling (Bradford Books) Paperback
by Thad Polk (Editor), Colleen Seifert (Editor)
Bayesian Cognitive Modeling - A Practical Course
by Michael Lee and Eric-Jan Wagenmakers