19.2 PRECURSORS OF ITS
19.2.1 Early, Mechanical Systems
Charles Babbage (early 1800s), is typically credited with being the first to envision a multi-purpose computer. He dreamed of creating an all-purpose machine which he called the "analytic engine." However, because of the technological constraints of the time, he was never able to build his dream, although he did succeed in building a difference engine, an automatic (mechanical) means of calculating logarithm tables.
The notion of using "intelligent machines" for teaching purposes can be traced back to 1926 when Pressey built an instructional machine teeming with multiple-choice questions and answers submitted by the teacher(see 2.3.4.2). It delivered questions, then provided immediate feedback to each learner:
The somewhat astounding way in which the functioning of the apparatus seems to fit in with the so-called 'laws of learning' deserves mention in this connection. The 'law of recency' operates to establish the correct answer in the mind of the subject, since it is always the last answer which is the right one. The 'law of frequency' also cooperates; by chance the right response tends to be made most often, since it is the only response by which the subject can go on to the next question. Further, with the addition of a simple attachment the apparatus will present the subject with a piece of candy or other reward upon his making any given score for which the experimenter may have set the device; that is the 'law of effect' also can be made, automatically(see 2.2.1.3), to aid in the establishing of the right answer (Pressey, 1926, p. 375).
While the above system was definitely clever for its time, it could not be construed as intelligent as it was mechanically set with pre-specified questions and answers. So, although it was inflexible, this system did incorporate contemporary learning theories and pedagogical strategies into its design (e.g., giving out candy for correct responses).
General-purpose digital computers arose in the mid-1900s, paving the way for truly (artificially) intelligent machines. Basically, these computers consisted of a numerical central processor whose mechanism was electronic, not mechanical, and based on a binary, not decimal, system. They were also characterized by having a built-in ability to make logical decisions, and a built-in device for easy storage and manipulation of data.
During this period of computer infancy, Alan Turing (1912-1954, British mathematician and logician) provided a major link between these modern, digital computing systems and thinking. He described a computing system capable of not only "number crunching" but symbolic manipulation as well. He also developed what is now known as the "Turing test," a means of determining a machine's intelligence. The test consists of an individual asking questions, in real-time, of both a human and computer. The interrogator attempts, in any way possible, to figure out which is which via conversations over the communication links. The Turing test has particular relevance to intelligent tutoring systems. The core concept behind the test is whether a reasonable person can distinguish between a computer and a person based solely on their respective responses to whatever questions or statements the interrogator renders. Thus, for a computer to pass the test, it would need to communicate like a human, which is a nontrivial goal. This line of inquiry has challenged and occupied researchers for the past 20+ years, and continues to play a prominent role in the development of ITS (see Merrill, Reiser, Ranney, and Trafton, 1992). Other communication-related research includes devising knowledge structuring and hypertext techniques within ITS to provide answers to the many possible questions that students could pose to the system. So, the success of this ITS enterprise really can be measured in a way that is similar to the Turing test: How well can the ITS communicate? We should point out, however, that the goal of ITS is to communicate its embedded knowledge effectively, not necessarily in an identical manner as human teachers. In fact, some teachers have great difficulty achieving the effective communication goal themselves.