22.7: Intelligent Tutoring Systems

Last updated
Save as PDF

Page ID: 88292

\( \newcommand{\vecs}[1]{\overset { \scriptstyle \rightharpoonup} {\mathbf{#1}} } \) \( \newcommand{\vecd}[1]{\overset{-\!-\!\rightharpoonup}{\vphantom{a}\smash {#1}}} \)\(\newcommand{\id}{\mathrm{id}}\) \( \newcommand{\Span}{\mathrm{span}}\) \( \newcommand{\kernel}{\mathrm{null}\,}\) \( \newcommand{\range}{\mathrm{range}\,}\) \( \newcommand{\RealPart}{\mathrm{Re}}\) \( \newcommand{\ImaginaryPart}{\mathrm{Im}}\) \( \newcommand{\Argument}{\mathrm{Arg}}\) \( \newcommand{\norm}[1]{\| #1 \|}\) \( \newcommand{\inner}[2]{\langle #1, #2 \rangle}\) \( \newcommand{\Span}{\mathrm{span}}\) \(\newcommand{\id}{\mathrm{id}}\) \( \newcommand{\Span}{\mathrm{span}}\) \( \newcommand{\kernel}{\mathrm{null}\,}\) \( \newcommand{\range}{\mathrm{range}\,}\) \( \newcommand{\RealPart}{\mathrm{Re}}\) \( \newcommand{\ImaginaryPart}{\mathrm{Im}}\) \( \newcommand{\Argument}{\mathrm{Arg}}\) \( \newcommand{\norm}[1]{\| #1 \|}\) \( \newcommand{\inner}[2]{\langle #1, #2 \rangle}\) \( \newcommand{\Span}{\mathrm{span}}\)\(\newcommand{\AA}{\unicode[.8,0]{x212B}}\)

Intelligent tutoring systems attempt to mimic the “perfect instructor”. The basic requirements of an intelligent tutoring system include the ability to:

model the learner
track misunderstandings
generate appropriate responses.

None of these basic requirements have been perfectly resolved.

Although it is possible to incorporate a model or two of student learning into a computer-based training application, a fixed model does not represent intelligence. How can a “typical” student be modelled when students and their learning preferences are so diverse? It is not sufficient to simply categorize students into one of two types and then create two ways for students to learn the material. This has been the premise in some “intelligent” tutoring systems. A compounding factor is that learner preferences vary depending on the situation and material being taught. It is impractical to create a different teaching strategy for every individual. See Chapter 20, Learning Strategies, for more information on learning styles.

Although intelligent tutoring systems should be adaptable, based on the learner’s previous successes and failures, it is a challenging goal. It is simple to record where students make mistakes, but a challenge to know when there is a misunderstanding, what caused it, and what to do about it. In a sense, the computer would have to be able to read each student’s mind.

Generating the appropriate response would be difficult even if the first two needs were met. How can a designer determine all of the response possibilities? Every possibility must be based on a known rule. Intelligent tutoring systems can and should have responses for expected misunderstandings but this is, at best, limited to the finite expressed problems.

There are some excellent intelligent tutoring systems available. However, these tend to be labour-intensive and expensive to develop. Although the potential of intelligent tutoring systems is exciting, the reality is that much research still needs to be done. In other words, instructors need not worry about being replaced by an intelligent tutoring system. Given the present state of the technology, it can be argued that well-designed instructional multimedia applications are essentially the same from a student’s perspective.