Teaching: An Apple for the Computer

Computerized teaching, long tested and talked about, goes into operation this week. More than 100 first-graders at East Palo Alto’s predominantly Negro Brentwood School will begin learning math and reading from a good-humored, patient, quick-minded machine.

What every good human teacher yearns to do is give each student tutoring that matches his unique needs. Confronted with 20 or 30 kids at once, the teacher simply cannot achieve that goal. “Computer technology,” says Dr. Patrick C. Suppes, 44, director of Stanford’s Institute for Mathematical Studies in the Social Sciences and developer of the East Palo Alto system, “provides the only serious hope for accommodating individual differences in subject-matter learning.”

Suppes has been working on his robot teacher for the past five years, using $2,500,000 in grants from the Carnegie Corporation, the National Science Foundation and the U.S. Office of Education. At first, he tested a primitive drill and practice system consisting of Teletypes hooked into a Stanford computer by telephone wires. The new IBM computerized teacher is housed in a windowless, thick-carpeted new building at the Brentwood School, and connects to 16 student-instruction “terminals” that have Teletypes, TV screens and speaker systems.

That’s Correct. The first thing that the student does is peck his name out on the Teletype (to kids who write “Batman,” the computer politely responds, “Please file again”). This enables the computer brain to run through the student’s record of instruction and achievement and pick his next drill. One reading drill, for instance, consists of teaching the student to combine the initial sounds r, p and b with the endings an, at and ag, to make ban, pan, ran, bat, pat, rat, bag and rag. As each word flashes on the screen, the taped voice pronounces it. Then, for example, the computer’s taped voice asks the student to touch the word ran on the screen with a “light pen.” A correct response brings an encouraging “Yes. That’s correct.”

An incorrect answer brings remedial exercises. If the student answers rag instead of ran, he evidently does not understand the basic concept being taught, so the computer goes back over previous drills. On the other hand, if he touches ban, he gets remedial exercises in initial sounds. Unlike a human teacher, the computer keeps abreast of the student, holds his attention, never gives up, pushes him to perform at his best. At any moment, the computer is giving its whole attention to only one student, but it works instantaneously on a “shared-time” basis and easily covers all 16 students at once.