in the UK [ www. cccnet. com ]. Investors in these companies are confident in a bright future for ILS.
To their credit, most of the ILS on the market have been redesigned in recent years to take advantage of multimedia capabilities and advances in instructional design. Unfortunately, the new versions of ILS have not been subjected to rigorous research and evaluation studies. The WWW sites associated with ILS vendors contain both testimonials and anecdotal evidence, but there is a complete lack of large-scale, externally conducted, rigorous research studies reported in the sites or obtainable through public information resources such as ERIC( Educational Resources Information Clearinghouse) [ http:// www. askeric. org ]. In addition, there is evidence that vendors underestimate the training required for teachers to make effective use of ILS or other forms of software( Robinson, 1992; Wiburg, 1995).
Intelligent Tutoring Systems
The basic components of intelligent tutoring systems( ITS) were conceptualized 25 years ago( Hartley & Sleeman, 1973) as 1) knowledge of the domain, 2) knowledge of the learner, and 3) knowledge of teaching strategies. In ITS language, these are often referred to as the expert model, the student model, and the tutor( Larkin, 1991). Others trace the history of ITS all the way back to 1926 when Sidney L. Pressey built an“ instructional machine” that presented a student with multiple-choice questions( Shute & Psotka, 1996), a device which could even dispense candy for correct answers. Advocates of ITS promote these systems as“ the most promising approach to delivering individualized instruction”( Shute & Psotka, 1996, p. 571) because the“ artificial intelligence” aspects of the program can allegedly diagnose and remedy student misconceptions with the precision of a human tutor.
Although much of the development of ITS has been done in the context of military and industrial training, there have been significant efforts to develop ITS for education, especially in challenging subjects such as algebra, calculus, and programming. For example, John Anderson( 1993) is well known for his work building a geometry ITS. An evaluation of Anderson’ s geometry tutor in an urban school setting indicated that the system had both positive learning outcomes and encouraged more cooperative problem-solving among students( Shute & Psotka, 1996). Unfortunately, despite a few positive evaluations in loosely controlled studies, few ITS have demonstrated the significant results promised by their developers.
ITS attracted much more attention, funding, and research a few years ago than they do today. One telling sign is that the Journal of Artificial Intelligence in Education recently changed its name to the Journal of Interactive Learning Research. Even those who have been most involved in research and development targeted at producing " intelligent tutors " have begun to acknowledge the lack of impact they have had on mainstream education( Lajoie & Derry, 1993). A major factor contributing to the lack of success of ITS is that the technical difficulties
14