MW 4 PM- 5:20 PM, 200 Deschutes
CRN: 16781, 4 Credits
Web Pages: http://www.cs.uoregon.edu/classes/09F/cis677/
Prof. Anthony
Hornof, 356 Deschutes
Office Hours: T 1-2 PM, or by appointment
hornof@cs.uoregon.edu
Computational cognitive modeling provides a framework for synthesizing and integrating psychological theory into computer programs. Cognitive modeling is an advanced research area that integrates artificial intelligence, human-computer interaction, and cognitive psychology, and that advances towards a "unified theory of cognition." This class will introduce the concepts and history of mental chronometry (the use of reaction-time studies to discern the stages of human-information processing) and introduce students to a computational "cognitive architecture" called EPIC (Executive Process-Interactive Control) that instantiates a current understanding of human perceptual, cognitive, and motor processing capabilities. The approach to cognitive modeling taught in this class advances psychological theory and aids in the design of more useful and usable human-computer interfaces.
The class will involve reading and writing papers, collecting and analyzing human reaction time data, substantial computer work using the EPIC cognitive architecture, and comparison of model predictions to human reaction time and eye movement data. This course takes a hands on approach to learning how to use state-of-the-art computational cognitive architectures to simulate and predict human behavior. Students will learn (a) the basic cognitive psychology of the sensory, memory, motor control, and executive control processes needed to model complex cognitive tasks, and (b) through a series of hands on modeling assignments, how to use the EPIC cognitive architecture to simulate and predict human performance for a range of human-computer tasks.
The course will require students to build and run some cognitive models using the EPIC cognitive architecture (which will be provided) which runs on Macintosh computers such as those in 100 Deschutes. The EPIC cognitive architecture will be taught in class but it would be best if students start with some computer programming experience.
There is no course textbook but readings will be made available.
The prerequisite of CIS 671 Artificial Intelligence can be waived for students with appropriate background and interest. Please email the instructor if you have any questions.
You are expected to read the assigned papers before class, take notes when you read, and come to class ready to discuss those readings, with questions about what you read. You are expected to take notes on what you read, understand the material, and to look up words that you don't know in a reputable dictionary or in the index of the textbook.
All students are expected to attend all lectures, presentations, and group meetings with the professor, and to take written notes at all classes and meetings. Students are also expected to participate in class presentations, group meetings, and discussions. Your in-class participation may be supplemented by emailing the instructor with questions or comments, which he will try to incorporate into the next class. All students will also be expected to participate in a few in-class group presentations.
Come to class ready to engage the material and to interact face-to-face with the other human beings in the classroom. Please leave your 21st-Century distractions behind. Turn off your cell phones and pagers (do not just put them into "vibrate" mode). Cell phones ringing in class are inappropriate, discourteous and disruptive. Activities such as surfing the web or checking email during class is inappropriate and discourteous. Please practice your skill of focusing on the intellectual activity occurring in a physical room with real live people.
There will be three or four modeling projects. The last project will be due in the final week of class.
There will be one midterm exam and a final exam. The final exam will be in the time slot allocated on the UO final exam schedule. There will also be quizzes and in-class exercises to further supplement learning and assessment.
You are expected to schedule other events such as trips or job interviews to avoid conflicts with the exam dates. The only acceptable excuses for missing an exam on the scheduled date and time are documented medical problems, religious holidays (if cleared in advance), presenting papers at leading academic conferences (if cleared in advance), and documented personal emergencies. An unexcused absence at an exam will result in a grade of zero on the exam.
Students have the responsibility to communicate with the instructor about any questions, concerns, or problems that they have during the course. These might relate to any aspect of the class including lectures, group dynamics, communication breakdowns, or anything. Please visit the instructor during his office hours or set up an appointment to discuss any problem or concerns.
Please contact the instructor with any questions and concerns at all regarding the class, following these guidelines:
Grades will be determined based on your performance on a midterm, a final exam, and in-class quizzes; and based on your group's performance on group projects. Your final grade will be weighted as follows:
Projects: 50%
Midterm: 20%
Final Exam: 20%
Class Participation and Quizzes: 10%
Grades for the course will be determined on the following scale:
90-100% = A
80-89% = B
70-79% = C
60-69% = D
59 and below = F
Assignments submitted late will be subjected to a full letter grade penalty. Assignments will not be accepted more than two calendar days past the due date. Exceptions will only be considered if the request is submitted before the due date.
Any grading discrepancies (such as a miscounting of points on an assignment) must be resolved within a week after the assignment is returned to the student.
Projects will be assessed against a set of evaluation criteria that will be made available with each project. Read the criteria carefully because they reflect the aspects of the projects that are important given the pedagogical goals of the class.
While much in the arts and sciences can be evaluated objectively (such as whether a poem is recited accurately, or whether a computer program will compile and produce a specified answer), much of the material covered in this class will be concepts, ideas, terminology, conventions, and practices that cannot be defined in pure objective language such as that of a string of text, or a computer program. Exams, quizzes, and projects are graded based on the instructor's subjective assessment of the accuracy and completeness of the answers and materials provided by the students. The instructor will apply an understanding of the material that he has established based on first-hand experience working on cognitive modeling research projects for fifteen years, learning from and collaborating with the leaders of the field, publishing papers on cognitive modeling projects in top peer-reviewed journals and conferences, and having won over a million dollars in funding from federal agencies to pursue EPIC-based cognitive modeling research.
Projects will be evaluated in part based on the instructor's subjective assessment of the quality of the written materials submitted. A modern skilled technology expert must be able to communicate his or her ideas clearly and concisely. Good writing occurs on three levels:
If you have any doubts about the quality of your writing, work with your project collaborators to critique each other's drafts. Also, take drafts to the drop-in writing lab at the Teaching and Learning Center in the basement of the PLC, open Monday - Friday from 9:00am to 4:00pm.
Modern writing standards advocate the use of inclusive language, and you should follow this standard. For example, if you are referring to a single anonymous person, you should write "he or she," "the user," or "the programmer," but not just "he." Also, please refer to women as "women" and not "girls" or "ladies".
People with disabilities should be first recognized as people, and then by their distinguishing characteristic. For example, "people with disabilities" is preferred over "disabled people".
The modern technology workplace is diverse, international, and intercultural. This course welcomes and values these differences as an opportunity to increase our awareness of the contemporary global society, how to work better in groups, and how to build better computer systems.
If you have a documented disability and anticipate needing accommodations in this course, during the first two weeks of class please (a) ask the counselor for students with disabilities send the instructor a letter verifying your disability and (b) arrange to meet with the instructor to discuss your needs.
You may not make audio or visual recordings of the class without explicit permission from the instructor.
Students who are found to have committed an academically dishonest act in this course will receive an F for the course.
Anemically honesty includes the following. You should do all of the following:
Academic dishonesty includes the following. You should not do any of the following:
All evidence of academic dishonesty will be rigorously pursued consistent with the University of Oregon Faculty Guide for Addressing Academic Misconduct.
This course is based on a Human Performance Modeling class developed by Dr. Travis Seymour at UC Santa Cruz. Some of the course materials used here may be directly copied, with permission, from Dr. Seymour's materials. We are grateful to Dr. Seymour for sharing his course materials with us.
A.Hornof 9/30/09