• Home
• Software
• Project
• Contact

Tutor Types

CTAT supports the creation of two types of tutors: example-tracing tutors, which can be created without programming but require problem-specific authoring, and cognitive tutors, which require AI programming to build a cognitive model of student problem solving but support tutoring across a range of problems.

If you are new to the process of creating a tutor, we recommend that you start with an example-tracing tutor.

Example-Tracing Tutors

Example-tracing tutors, formerly called 'pseudo tutors' as a shorthand for pseudo-intelligent tutors, resemble cognitive tutors. The example-tracing tutor's knowledge, however, is not generalizable; it is unable to reason on its own across multiple problem scenarios. You must therefore 'teach' the tutor various behavior for each of the problems you create. Depending on the complexity and number of problems, this trait may be a consideration. If, in the future, you decide to construct a full cognitive tutor, your example-tracing tutor can serve as a building block for that cognitive tutor.

Example-tracing tutors can be built quickly, and can serve as a tool for "rapid prototyping", or creating iterative prototypes over a short amount of time.

Cognitive Tutors

Cognitive tutors are the more robust of the two tutor types. They are based on cognitive psychology theory, particularly the ACT-R theory. Developing a cognitive tutor involves creating a cognitive model of student problem solving by writing production rules that characterize the variety of strategies and misconceptions students may acquire. Cognitive Tutors are sometimes are also called "Production Rule Tutors", a reference to the production rules you will create.

Your decision may also depend on:

• the complexity of the tutor problems (i.e., the problems that students solve as they work with the tutor);
• whether many problems of the same type need to be developed;
• how many alternative solutions paths exist for each problem;
• whether subtle ordering constraints exist on the steps within the tutor problems
• whether later steps in a problem depend on earlier steps; and
• whether in-house expertise in AI programming is available to develop a cognitive model.

Choosing between Flash and Java for Authoring and Delivery

Before developing a tutor, you must choose between creating the user interface in Java or Flash. For each of these languages, CTAT provides a set of specialized tutoring widgets. These widgets can be used to create interfaces in drag-and-drop manner, without requiring any programming. (If you're familier with programming in either Java or ActionScript, you can customize the interfaces and add non-standard behavior, but often, this is not necessary.)

Other alternatives

In addition to the options explained above, you may want to use CTAT to facilitate a number of other goals. You may want to:

• develop a student interface without tutoring but with logging;
• log from an existing tutoring system or problem-solving environment using the CTAT logging library;
• use a home-grown student GUI with CTAT for behavior recording, cognitive tutoring and/or logging (for example, using an existing simulator such as the Chemistry VLab or problem-solving environment, or a GUI that has been developed previously, in Java, Flash, or a different language. Typically, such simulators or environments do not provide tutoring on what scenarios to explore with the simulator, how to interpret the results, etc. A CTAT-based tutor could be developed as a plug-in tutor agent for the simulator.); or
• develop a Cognitive Tutor using TDK as the cognitive modeling language.

If you would like to accomplish any of those goals, please contact us for assistance.

Continue to Download CTAT.