If by a miracle of mechanical ingenuity . . .

Have you ever thought where did all this computer enabled learning come from?

Miracle in the making

In 1912, an American education researcher Dr. Edward L Thorndike said “If by a miracle of mechanical ingenuity, a book could be arranged that only to him who had done what was directed on page one would page two become visible and so on, much that now requires personal instruction could be managed by print.”

Then, in the early 1920s, Dr. Sidney L Pressey of Ohio State University realised Thorndike’s “miracle” when he developed a mechanical device for testing and scoring simple attainment tests.  He discovered that the device also had instructional qualities;  not only could it deliver a test and score, it could teach.  His device consisted of a predetermined sequence of questions which were shown, one at a time, through an aperture.  The student selected an appropriate answer from a card and pressed a key representing that answer; the selection of the correct answer resulted in a new question appearing in front of the student.

From there, progress was made during the 1930s and 1940s in the development of teaching machines and the theory of automated self-paced learning and assessment, but an efficient means of developing and delivering the concepts was lacking.

Teaching Machines

The use of paper media represented constraint, particularly in the assessment of the material and subsequent decisions.  This was recognised, and researchers began to turn to automating self-paced learning through some form of mechanical means.

Perhaps the earliest of these was Pressey’s 1926 device which when introduced in the late 1920s, included a chute which produced a sweet if the student got the answer/s correct.  It presented multiple choice questions one at a time by rotating a cylindrical drum on which the questions were printed and could be seen through a glass window.  The student would then press one of four levers to correspond with their choice of answer.  This machine automatically recorded all responses, which Pressey used to adjust his lecture plans for his classes, to allow more time to be spent on concepts that were consistently missed and less on those easily grasped.  But it was not able to indicate correctness to students of the answers they chose.

He subsequently and gradually improved the machine with additional functions such as scoring, rudimentary feedback and the ability to bypass questions when a student had previously answered correctly. Then in 1932, although confident an industrial revolution in education was coming, he disbanded his work in due to lack of funds.  He inserted the following epilogue in an article written the same year. “The writer has found from bitter experience that one person alone can accomplish very little, and he is dropping further work on these problems.  But he hopes that enough may have been done to stimulate other workers and that this fascinating field may be developed”.

Further progress

In 1934, James Little commenced some studies into Pressey’s early teaching machines.  He found a significant difference between the final examination grades of students who were immediately informed of their results and those of students who did not have this feedback.  He found that immediate feedback was of greatest benefit to students who had struggled with the subject.

A number of learning devices were developed in the late 1930s, in fact it was reported in 1936 that over 600 inventions had been patented as educational aids.  But it would not be until the end of World War 2 that more widespread acceptance of mechanised teaching and testing took place.

Programmed learning inaugurated

It was in 1954 that Professor B F Skinner published an historic paper “The Science of Learning and the Art of Teaching”, in which he foresaw the possibility of applying concepts deduced in the experimental laboratory, directly to classroom-teaching.  He suggested that his extensive studies on experimental analysis of behaviour could be applied in the construction of a teaching machine which could present a carefully sequenced set of material to a student and reinforce the student’s responses to what Skinner called “desired behavioural capabilities”.  The paper attracted wide attention and a new era for Programmed Learning (PL) was inaugurated.

Skinner was particularly concerned about ensuring students obtained immediate feedback on responses to questions, and also, that they could progress in a step-by-step manner to mould complex behaviour.  Skinner’s first machine enabled students to construct responses to questions rather than simply selecting the correct statement in a multiple- choice fashion.  Students decided on their responses by placing slides on the machine’s front panel corresponding to appropriate letters or numbers.  The student then turned a crank; if the response was correct the next question was displayed, if not correct the crank would not turn.  In this way the knowledge of results and reinforcement (the positive movement of the crank) were both immediate.

Where was the feedback?

Some on-going problems remained because, although students were rewarded for correct responses, no feedback was provided for incorrect responses.

This led to advances in the field of individualised self-paced Programmed Learning.  This differed from the more conventional methods of teaching, common in the early 1900s.  It represented an individual learning process in which the student accepts a far wider measure of responsibility for his/her own learning while proceeding at his/her own rate.  The learning was not passive in nature as progress required a response from students to questions and the immediate confirmation of results.  It also ensured the student is successful more often, and this in turn enhanced motivation and enabled a desired outcome of the learning.

So, this is where it all started!  But at that time, if there was non-mechanical device that could provide individualised learning and testing, with appropriate feedback, it would be extremely useful.

I can’t find anything documented where researchers of the day re-lived Thorndike’s “miracle” (openeing paragraph to this post) so many years after he saw that early vision, but I can imagine their excitement as computers started to be built and researched.

Computer Based-Training

But these were early days and, while some exciting developments took place at the University of Illinois around 1960 with the development of Donald Bitzer’s PLATO CBT System, it wasn’t until the mid-1970s that the notion of using computers to teach grabbed the attention of significant numbers of academics.

From there, we have seen the commercial use of mainframe computers that ran Computer Based Training and in the early late 1970s and early 1980s the micro-computer was born.

When I started in this field (1983) we were using 8inch floppy disks in disk drives larger than a typical tower unit now and 512×512 resolution touch sensitive screens in any colour you wanted as long as it was green!  We progressed to 51/2 inch floppy disks, 3 inch diskettes, video disk and CD ROM before e-learning made its mark with the advent of the Web in about 1994.  Since then of course the options to deliver quality interactive learning material have expanded significantly.

Thorndike’s miracle has happened.  And wouldn’t it be great to show him what is possible now.


Do you have a vision?  Can you envisage a “miracle” as we ponder where learning, knowledge and technology will take us in say, the next 20 years?