Intermediate level
Prerequisite: Build Intelligent Tutors with CTAT
Rule-Based Cognitive Modeling for Intelligent Tutors
Flexible schedule
4 weeks, 6 to 8 hours per week
Instructor feedback
Get guidance on your work
Verified certificate
Share on LinkedIn
*Proof of full-time student enrollment required. Acceptable forms of ID include a letter from your university’s registrar office or an unofficial transcript. Email your documents to learnlab-help@lists.andrew.cmu.edu.
What you will learn
- Explain how rule-based cognitive models differ from simpler tutor representations.
- Represent problem-solving knowledge and steps as rules within CTAT.
- Connect model structure to tutor feedback, hinting, and error handling.
- Evaluate the strengths and limitations of rule-based cognitive modeling for intelligent tutors.
Course description
Building more advanced intelligent tutors requires a model of how learners solve problems, not just a record of correct answers. Rule-based cognitive models make it possible to represent the steps, knowledge, and decision points behind learner performance so tutors can respond more intelligently.
In this course, you will learn how to develop rule-based cognitive models in CTAT for cognitive tutors. You will examine how to represent expert problem solving step by step, connect model structure to tutor behavior, and build more sophisticated tutoring systems for domains where process matters as much as outcome.
Syllabus
Module 1: ACT-R and Cognitive Modeling
- Define the main claims of ACT-R.
- Distinguish between declarative and procedural knowledge.
- Explain evidence that production rules are units of procedural knowledge acquisition.
- Describe key relationships between ACT-R and Cognitive Tutor technology.
Module 2: Foundations of Production Systems
- Define production rules.
- Explain how a basic production rule interpreter works.
- Given a set of rules and task context, interpret production rules.
- Write production rules for a given task.
Module 3: Model Tracing – Nools
- Describe how working memory is represented in Nools.
- Implement a fact template with a define statement.
- Create a fact in working memory.
- Create a Nools rule.
- Populate working memory with a bootstrap rule.
- Implement tutor flow with backtracking, halt, and checkSAI.
- Explain model tracing in the context of intelligent tutoring systems.
- Implement a tutor-performed action.
- Create a hint in Nools.
- Create a Done rule in Nools.
- Set up a Nools Development Environment.
Module 4: Model Tracing – Debugging
- Interpret the output of debugging commands.
- Given an error, select which debugging command to choose.
- Diagnose problems with a cognitive model using the Conflict Tree.
- Given a set of production rules and a state of the interface, predict the output of debugging commands.
Module 5: Model Tracing – Assignments
- Assignment 1: Create a Cognitive Tutor for Squaring Numbers Ending in 5.
- Assignment 2: Create a Cognitive Tutor for proportional reasoning.
- Each assignment will be graded by the instructor and you will receive personalized feedback along with a sample solution.
Module 6: Course Project or Final Exam
At the end of the course, you’ll have an opportunity to build a rule-based tutor for a task for which building an example-tracing tutor is not practical, namely, an abductive reasoning task in Mendelian genetics. That will provide you with a nice experience to apply the fundamentals you will learn in the modules to a larger, more authentic context. It will be graded by the instructor and you will receive personalized feedback along with a sample solution.
The project offers an opportunity to learn about doing an analytical cognitive task analysis, designing your own working memory representation, modeling steps with multiple rules, linking the interface representation dynamically to the semantic representation of the problem, creating a dynamic interface with a Nools tutor, and using additional Nools constructs such as not, exists, salience, and arrays as slot values.
You will also have the option to take a final exam with 20 questions. The exam can be taken multiple times, and each attempt draws new questions randomly from a pool of questions.
You may also complete both the course project and the final exam. The higher of the two scores will count toward the certificate.
Meet the instructor
Dr. Vincent Aleven
Vincent Aleven’s research aims to advance the science of how people interact and learn with adaptive, AI-based learning technologies, and to advance the design and engineering of these technologies. Practically, he aims to help realize the smart classroom through strong synergy among learners, those who facilitate learning such as teachers, instructors, peers, tutors, and parents, and novel AI applications. In this context, he is excited to help a new generation of scientists and professionals develop interest and skill in research and development.
Where to go next
Build Intelligent Tutors with CTAT, No Programming Required
Create example-tracing tutors with CTAT without programming and develop intelligent tutoring experiences through demonstration and behavior design.
Predictive Modeling and Knowledge Tracing for Adaptive Learning
Build predictive models and use knowledge tracing to model learner progress and support adaptive learning systems.
Adaptive Learning and Intelligent Tutoring Systems
Design adaptive learning experiences and intelligent tutoring systems that personalize support based on learner knowledge, behavior, and affect.