Type of Credit: Elective
Credit(s)
Number of Students
Cognitive modeling is an important research skill, which cannot only be an embodiment of a theory, but also be used to verify and compare quantitatively theories. Normally, researchers develop a cognitive model according to a particular theory, which addresses the issues about mental representations and mental processes for particular cognitive functions. The issues relevant to cognitive modeling include how to translate written contents of theories to something computable (i.e., model implementation), how to optimze a model's performance, and how to make a fair comparison between models. For model implementation, one needs to know how to use a computer language (e.g., Python or R) to compile the script specifically created for a theory. For model optimzation, we need to know how to choose suitable parameter values to make model predictions as much similar to observed data as possible. To this end, students will be introduced several algorithms for optimizing model performance. The methods of cognitive modeling can also be extended to modeling with the theories in addition to cognitive theories.
This course is specifically designed for Ph.D. students. Anyone who is interested in taking on this course is required to discuss with the lecturer of this course in advance. The prospective students are expected to be able to use at least one computer language, such as Python, R, Matlab, or C/C++.
能力項目說明
After learning this course, students are expected to be able to (1) implement a theory as a computational model, (2) optimize a model performanace via choosing suitable parameter values, and (3) compare different models performance with different quantitative indices. With these skills, students are expected to be able to do modeling to verify theories.
| 教學週次Course Week | 彈性補充教學週次Flexible Supplemental Instruction Week | 彈性補充教學類別Flexible Supplemental Instruction Type |
|---|---|---|
Week Theme Content Activity and Homework Estimated time devoted to coursework per week (hr)
1 Introduction Introduction Examples of cognitive modeling 12
2 Theory I Choosing theories Discussion with lecturer 12
for each student to choose a theory for implementing
3 Theory II Choosing theories Discussion with lecturer 12
for each student to choose a theory for implementing 12
4 National Day
5 Optimization I Parameter parameters Hill-Climbing algorithm 12
6 Optimization II Parameter parameters Simplex algorithm 12
7 Optimization III Parameter parameters Bayeisan inference 12
8 Optimization IV Parameter parameters Bayeisan inference 12
9 Midterm examination
10 Optimization V Parameter parameters Regulization 12
11 Optimization VI Parameter parameters Cross validation 12
12 Model comparison I Goodness of fit RMSD, AIC, BIC 12
13 Model comparison II Goodness of fit G^2, likelihood ratio, etc. 12
14 Model comparison III Bayesian framework Bayes factor and so on 12
15 Model comparison IV Bayesian framework Super model 12
16
17 Oral presentation I Oral presentation of Students finish a project of
each student's project cognitive modeling 24
18 Oral presentation II Oral presetnation of Students finish a project of
each student's project cognitive modeling 24
Students are required to finish a project by the end of semeter, which must be an implementation of a psychology model. The degree of completeness, the adequancy of modeling methods used in the project, and the explannation to the modeling results will be largely emphasized.
Teacher's lecture notes.