Teaching Machines and Programmed Learning

Print Friendly, PDF & Email

Charlotte Thomson IserbytDay 8: Skin­ner Hor­ror Files

Teach­ings Machines => Com­put­ers

Skinner quoteComput­ers have become the per­fect vehi­cle for those who want to imple­ment soci­etal oper­ant con­di­tion­ing meth­ods on the chil­dren and pop­u­lace. These change agents desire to break down human behav­ior into the low­est com­mon denom­i­na­tor, and mea­sure it bit by bit via con­di­tioned respons­es desired (or required) by the pow­ers-that-be. In this man­ner a new soci­ety will be formed.

B.F. Skin­ner authored a chap­ter in a book titled Teach­ing Machines and Pro­grammed Learn­ing: A Source Book, edit­ed by A.A. Lums­daine and Robert Glaser in 1960. Appen­dix II of my book gives many inter­est­ing excerpts from this book. Skinner’s chap­ter “The Sci­ence of Learn­ing and the Art of Teach­ing” pro­vide some clues to his view of edu­ca­tion. First note that the human is mere­ly an “organ­ism” to Skin­ner.

Baby on ComputerIn all this work, the species of the organ­ism has made sur­pris­ing­ly lit­tle dif­fer­ence. It is true that the organ­isms stud­ied have all been ver­te­brates, but they still cov­er a wide range. Com­pa­ra­ble results have been obtained with rats, pigeons, dogs, mon­keys, human chil­dren, and most recent­ly… with human psy­chot­ic sub­jects. In spite of great phy­lo­ge­net­ic dif­fer­ences, all these organ­isms show amaz­ing­ly sim­i­lar prop­er­ties of thelearn­ing process. It should be empha­sized that this has been achieved by ana­lyz­ing the effects of rein­force­ment and by design­ing tech­niques which manip­u­late rein­force­ment with con­sid­er­able pre­ci­sion. Only in this way can the behav­ior of the indi­vid­ual organ­ism be brought under such pre­cise con­trol. It is also impor­tant to note that through a grad­ual advance to com­plex inter­re­la­tions among responses, the same degree of rig­or is being extend­ed to behav­ior which would usu­al­ly be assigned to such fields as per­cep­tion, think­ing, and per­son­al­i­ty dynam­ics. (p. 103)

Watch DOSITEJ project - empowering education

Watch DOSITEJ project — empow­er­ing edu­ca­tion

Note the admis­sion that the goal is “con­trol­ling the organ­ism.” And the teacher’s “inad­e­qua­cy” with­out “mechan­i­cal devices.” And the sen­tence high­light­ed in bright red.

These require­ments are not exces­sive, but they are prob­a­bly incom­pat­i­ble with the cur­rent real­i­ties of theclass­room. In the exper­i­men­tal study oflearn­ing it has been found that thecon­tin­gen­cies of rein­force­ment which are most effi­cient incon­trol­ling the organ­ism can­not be arranged through the per­son­al medi­a­tion of the exper­i­menter. An organ­ism is affect­ed by sub­tle details of con­tin­gen­cies which are beyond the capac­i­ty of the human organ­ism to arrange. Mechan­i­cal and elec­tri­cal devices must be used. Mechan­i­cal help is also demand­ed by the sheer num­ber of con­tin­gen­cies which may be used effi­cient­ly in a sin­gle exper­i­men­tal ses­sion.

See here

See here

We have record­ed many mil­lions of respons­es from a sin­gle organ­ismdur­ing thou­sands of exper­i­men­tal hours. Per­son­al arrange­ment of the con­tin­gen­cies and per­son­al obser­va­tion of the results are quite unthink­able. Now, the human organ­ism is, if any­thing, more sen­si­tive to pre­cise con­tin­gen­cies than the oth­er organ­ismswe have stud­ied. We have every rea­son to expect, there­fore, that 896. The sim­ple fact is that, as a mere rein­forc­ing mech­a­nism, the teacher is out of date. This would be true even if a sin­gle teacher devot­ed all her time to a sin­gle child, but her inad­e­qua­cy is mul­ti­plied many-fold when she must serve as a rein­forc­ing device to many chil­dren at once. If the teacher is to take advan­tage of recent advances in the study of learn­ing, she must have the help of mechan­i­cal devices. (p. 109)

In case any­one doubts my stance that this is a delib­er­ate dumb­ing down of amer­i­ca, see this sen­tence high­light­ed in bright red:

The impor­tant fea­tures of the device are these: Rein­force­ment for the right answer is imme­di­ate. The mere manip­u­la­tion of the device will prob­a­bly be rein­forc­ing enough to keep the aver­age pupil at work for a suit­able peri­od each day, pro­vid­ed traces of ear­li­er aver­sive con­trol can be wiped out. A teacher may super­vise an entire class at work on such devices at the same time, yet each child may progress at his own rate, com­plet­ing as many prob­lems as pos­si­ble with­in the class peri­od. If forced to be away from school, he may return to pick up where he left off. The gift­ed child will advance rapid­ly but can be kept from get­ting too far ahead either by being excused from arith­metic for a time or by being giv­en spe­cial sets of prob­lems which take him into some of the inter­est­ing by-paths of math­e­mat­ics. (p. 110)

Skinner Rat2Skinner’s goal is the “manip­u­la­tion” of human behav­ior via the “manip­u­la­tion” of the device. And mech­a­nis­tic devices such as com­put­ers assist with that manip­u­la­tion.

Some objec­tions to the use of such devices in the class­room can eas­i­ly be fore­seen. The cry will be raised that the child is being treat­ed as a mere ani­mal and that an essen­tial­ly human intel­lec­tu­al achieve­ment is being ana­lyzed in undu­ly mech­a­nis­tic terms. Math­e­mat­i­cal behav­ior is usu­al­ly regard­ed not as a reper­toire of respons­es involv­ing num­bers and numer­i­cal oper­a­tions, but as evi­dence of math­e­mat­i­cal abil­i­ty or the exer­cise of the pow­er of rea­son. It is true that thetech­niques which are emerg­ing from the exper­i­men­tal study of learn­ing are not designed to “devel­op the mind” or to fur­ther some vague “under­stand­ing” of math­e­mat­i­cal rela­tion­ships. They are designed, on the con­trary, to estab­lish the very behav­iors which are tak­en to be the evi­dences of such men­tal states or process­es. This is only a spe­cial case of the gen­er­al change which is under way in the inter­pre­ta­tion of human affairs. An advanc­ing sci­ence con­tin­ues to offer more and more con­vinc­ing alter­na­tives to tra­di­tion­al for­mu­la­tions. The behav­ior in terms of which human thinking must even­tu­al­ly be defined is worth treat­ing in its own right as the sub­stan­tial goal for edu­ca­tion. (p. 111)

Repro­duced from pages A-8 and A-9 of Appen­dix of my book the delib­er­ate dumb­ing down of amer­i­ca.