Number 11 LOGAN

Number 11 LOGAN - FRANKKLOGAN Expenimenral Psycholoqy of...

Info iconThis preview shows pages 1–7. Sign up to view the full content.

View Full Document Right Arrow Icon
Background image of page 1

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full Document Right Arrow Icon
Background image of page 2
Background image of page 3

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full Document Right Arrow Icon
Background image of page 4
Background image of page 5

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full Document Right Arrow Icon
Background image of page 6
Background image of page 7
This is the end of the preview. Sign up to access the rest of the document.

Unformatted text preview: FRANKKLOGAN Expenimenral Psycholoqy of Animal Lmuiuq and Now FranltA. Logamfi'ofassorofl’sydioiowatflsa Universityr of New Mexico, Albuquerque, is a noted psychologist and I'asaarehar in animd learn- ing. Although the recent humanistic movement in psychology has aroused for lass emphasis on animal research, Dr. Loom Wants out in the following article the unlit-ability of animal learning models to a wile variety of human durations. Apparently this Iineofrasaarchwill haasralerant inthafu- tore as it hes been in the past. 0 What is the incentive theory of learning? 0 What is the true role of reinforcement in learning? 0 Whataresomaolthaathicd problemsinha- hariord cont-cl? The experimental psychology of learning hm never enjoyed the luxury of an ivory tower. Very ' much to the contrary, it typically has occupied dark basements. From such quarters, it has been striving for about 50 years on two fronts. One has been to establish its respectability not only among the social sciences but also as an experi- mental science alongside biology, chemistry, _ geology, and physics. This goal appears to have been largely accomplidted. The other front has been to prove its worth as a relevant enterprise. This article reflects on some of the ways in which From “Experimental Hydrology of Animal Learning and Now” by Frank A. Logan fromAmerimrr Prydrob tit: 27. no. 11 (1972): 1055-1062. Copyright © 1912 by the American Psychological Association. Reprinted by mission ofthe Association and Frank A. Logan. basic experimental research on the learning pro? cess. predominantly with rats as subjects, qualifies a relevant now. . . . The intent is to state a few principles of behavior, go briefly into the laboratory to show from where those principles came, and then go out into the world with some of their many podble implica- tions. . . . WHAT IS LEARNED? An early question in psychological theory was, Whatislearned? . . . Tltetraditionalview,.. .was this: when a person learns something, he naturally is able to do it faster and faster. Indeed, one can measure how well he has learned by how fast he canrespond. . . . [He] learnswhcttodonnd speed reflects this learning. itwasalreadyclearthatthere wasatleast one ' problem with this traditional View, namely, that - a person should not necessarily do everything as fast as possible. But there remained the intriguing ‘ possibility that speed is better conceived as part ' of what is learned.— This thesis is that organisms learn not only what to do, but also how fast do dOiL... lnthetypicalcaseofaratinamaze.... it becomes apparent that how fast a rat runs de- pends on the particular conditions of reinforce— ment. He is perfectly capable of running slowly ' if doing so yields a larger or more immediate re- ward; he runs fast to the extent that it pays him to do so. . . . Th'n concept can be applied to the typical ele- mentary course in arithmetic. . . . During a test, the teacher may walk around the room and, in 24 Reading About taming About Learning ' pleasing, warn a student to slow down and be care- ful not to make any mistakes. By such tactics, that student is learning to be a slow “arithmetick- er." The probability is that most students are (013*! to be slow, not only in arithmetic, but in reading (which of course is well known but not ' yet reconciled) and a host of other behaviors in- cluding thinking. Teachers who take students through a logical argument in methodical, ardu- ous, repetitive detail teach students to think slow- ly. People can think much faster than people can talk. . r . [T] he point can be extended as a result of another study with further implications for education. . This study . . . also involved humans learning paired associates. In this case, both the stimuli and the responses were simple meaningful words. For example, the subjects might have to learn that the stimulus word cow was associated with the re‘ sponse word table. The list contained 10 such pairs and was given only twice, the second time to see how many responses had been learned from the first exposure. Initially, college students typically can remember about 2 of the 10 asso- ciations after a single trial. The subjects were then given additional com- parable lists, using different words for both Stimuli and responses, and with practice they learned to learn such paired associates. By the sixteenth list, they were able, on the average, to remember about 6 of the 10 responses from a single triaL In this study, response speed was varied, . . . but input speed also was varied. For half of the subjects, the memory drum was rotated rather rapidly, al- lowing only two seconds between exposure of the response word and the nextstimulus presentation. while for the other half, the drum was rotated rather slowly, allowing five seconds between trials. Both groups learned comparable lists, learned ap- proximame equally welL and learned to learn. But one group was accustomed to fast input and the other group to slow input. At that stage of the study, the speed of the memory drum was reversed for half of each group. These subjects now were presented with additional lists to learn, but at a speed of input different from that to which they had become accustomed. The result was that the control groups continued at their high level. while the reversed groups, al- though instructed that the speed had been changed. reverted to their original level. Most dramatically, the group that had learned to learn paired associ- ates when the material was presented rapidly did not profit from their experience with the task when the additional material was presented at a slow pace. In spite of the fact that they now had more time for rehearsal, they could not learn as well. These results indicate that not only do people learn to remand at particular speeds, but they also learn to learn at particular speeds. As a practical example, a child from the North whose family moves to the South may do poorly in school in part because he had become accustomed to a teacher who speaks rapidly and not to one who _ drawls slowly. The child who initially learns from slow-speaking parents may have difficulty with fast-speaking teachers, and this may be especially true in bilingual families where the language of the teacher is the secondary language of the home. There are many implications of this realization. First, of course, many problems can be solved by appropriate remedial training and certainly need to be. But in a more futuristic image, techniques are available to compress tape-recorded speech, not by simply speeding up the tape but by deleting parts of each sound. A teacher could record a lecture, have it reduced to halfits length, and stu- dents could learn to learn perfectly well from that faster input. Indeed, they should learn better from two fast exposures in the same length of time. Just as the well-trained speedreader can as- similate considerably more information than those who only learned to read slowly, so can people learn to listen rapidly. Probably more refinements will be needed, such as “setting” a psychological dial for the appropriate speed of input, but these are largely engineering problems stimulated by basic research in the animal laboratory. ROLE OF REINFORCEMENT Another conceptual problem also has been around for a long time: What is the role of reinforcement I. in learning? The prevailing View. . . was that leam- ing required some kind of affective effect in order to be “stamped in." And to many, this conception still prevails. In an effort to analyze this question experimentally, we studied decision making in rats. . . . Specifically, rats were given a choice be- tween two alternatives, one of which offered a larger reward than the other, but also entailed a longer time of delay. The rats had to decide be- tween getting a smaller reward now or waiting for a larger reward. This approach was extended to probability of reward—the rat could get a larger reward in one alternative, but was less certain of receiving it. The finding was that rats are contin- uously sensitive to the conditions of reinforce- ment, preferring the larger reward unless the delay , becomes too long. These results, as well as those of related studies, make it clear that reward does not affect learning direcrbz, at least in the claSSic sense of performing a response. . . . [T] his ap- proach is referred to as “incentive theory." Incentive theory has twc major practical implica- tions. The first is that people learn simply from- practice, which, when combined with the micro- molar point, implies that people learn what they practice the way they practice it. The second im- plication is that reward affects performance in a motivational manner. This conceptualization gen- erates important reinterpretations of the deter- minants of behavior. Consider first the implication that learning re- sults from practice. if one is interested in in- . creasing the probability of a behavior, of oneself - or another, then nothing more is required than that the environment be arranged so that that be- -_ .havior occurs appropriately. An example . . . con- cerns toilet training a child (or a pet, for that matter). It is not necessary to bestow praise for “right” behavior or blame for “wrong” behavior; it is only necessary to insure that, for a period of time, the desired behavior occurs in the desired place. So reared, the child does not think there is anything especially good about one way or es» pecially bad about another; he simply behaves ac- . cording to the way he practiced behaving. There is no jusfification for the neuroses frequently traced to traumatic toilet training. The principle that learning results only from practice becomes ominous with respect to unde- sirable responses. It is not the case that one can simply ignore behavior in the belief that it will not become habitual unless it is rewarded. Much to the contrary. . . . [A] person who practices a re- sponse in a sloppy or inefficient manner learns to behave in a sloppy and inefficient manner. As a very minor but true personal illustration, I once thought it would be fun to see if I could type fast without spaces between words. So I spent time practicing the familiar line, “Nowisthetimeforall- goodmentocornetotheaidoftheircountry,” and learned that I could type without a space or a pause in doing so. But to this day, I occasionally - miss the space bar when typing fast. . . . ROLE OF PUNISHMENT Since incentive theory assumes that learning re- sults from practice and that rewards act in a mo- tivational manner, a quite natural further con- sideration involves the role of punishment. The prevailing view, unfortunately largely a result of pronouncements by both qualified and self-ap- pointed psychologists, is' that punishment is an ineffective means of control, and worse, is detri- mental to well-being. No one could deny that be- havior could be suppressed by making an aversive event contingent on that behavior, but the effect was said to be transitory and the side effects un- desirable. Armed with a richer view of the role of rein- forcement, it seemed that punishment could be integrated equally into a general incentive theory of behavior. . . . Again we used a choice pro- cedure with rats. In this case, they could get a larger reward, or a more immediate reward, or a more certain reward in one alternative than in the other, but the “price” was to cross a grid floor that gave a brief but punishing electric shock. To make that story as short as possible, the rats be- haved intelligently. If the shock was weak, they would take it in order to get the preferred reward, but if it was strong enough, they would settle for the lesser reward. And insofar as I can identify with rats, 1 would say that they did so happily, at least in the sense that they ran even faster than expected to the lesser reward and showed no signs of aggression against the experimenter. This point is that punishment is a perfectly ap— propriate means of control, although it is impor- tant that the principles be understood and used . properly. Punishment does not automatically “stamp out” behaviors, anymore than reinforce- ment stamps responses in, but it does contribute to the total motivational complex. Those who - have preached against the use of punishment may well have contributed to the decadence that is be- coming increasingly apparent in our society. Con— sider, for example, the mother who has discovered that her 5-year-old son has stolen a dollar from her purse. She might go to one of the several most popular books on child rearing and development and read that many 5-year-old boys sometimes steal. She then reads about the undesirable effects of punishment, its transitory nature, and consoles herself that this is merely a phase of development that will pass with time. It is difficult not to get emotional when reflecting'on this type of scene. it is easy to accept the fact that many 5-year— old boys (and girls) steal. I did. But the best guarantee that they will not steal at age 6, or 16, or 60 is to provide proper treatment at age 5. Cer- tainly, this is not intended to imply rampant, ill- considered punishment, but it might include vigor- ous punishment that suits the crime. The child should be required to do more than enough extra work to repay the money. In addition, the parent needs to be aware that an act of thievery is moi tivated, and also help the child learn to deal with _ those needs in some acceptable fashion. If the child really needs a dollar, he should learn how to earn it or at least ask for it. if the child actually is seeking attention, albeit painful, then he should be provided with better means of gaining atten- tion. Ptuu'shment is undesirable only if used un- wisely and as the only means of behavioral con- trol. ' THE PAIRED-AVOIDANCE PARADOX —To this point, reference has been to principles that are more or less already convincing on intuitive grounds before going into the laboratory. . . . It was necessary to analyze these principles in the laboratory, but the general outcome was reason- ably certain. However, the laboratory is not always so gracious, and there is another line of research in which the results have been surprising. The situation is what is technically called a dis- criminated avoidance task. Rats were run in an apparatus in which a row of lights occasionally came on as a warning signal that an unpleasant electric shock was about to be applied to the grid floor. . . . This apparatus contained a small wheel protruding from one wall, and the program was such that if the rat turned the wheel after the warning lights and before the shock, the lights - would go out and the shock was precluded. Fail- ing that, however, the shock came on and stayed on until the rat turned the wheel to escape from it. This is called "discriminated" because there is a warning signal preceding the shock, and “avoid— anCe" because there is something the rat can do to' prevent getting shocked. Most tats learn to behave reasonably adaptively, in the sense of usually ' turning the wheel when the lights come on and hence not getting shocked. intrigued by reports from Aarin and his col- leagues . . . , we put together two rats that had been trained individually in this situation and that had learned to avoid most shocks. The obvious expec- tation was that they would do at least as well to- gether as one or the other had been doing alone. indeed, one might reasonably expect the pair to do even better than that, on the grounds that if one rat were for some reason distracted or otherwise preoccupied, the second rat might fill the breach. But this is rarely what happens. More commonly, avoidance behavior is totally disrupted. The rats I cringe at the sight of the warning lights; in anthro- pornorphic language, they “know” that a painful . shock is impending, but neither rat will turn the wheel to prevent it. ' When the programan shock arrives, one of the rats turns the wheel, and it is predictable which rat it is most likely to be. It is the rat that has been judged independently to be the submissive member of the pair. But don't jump to conclu- sions. It is not the case that the dominant rat “forces” the submissive rat to do the work; if anything, it is quite the opposite. The dominant rat actually may attempt to prevent the submissive rat from turning the shock off. A large dominant male rat may hold a small submissive male by the head and keep him from getting to the wheel. The dominant rat is standing on the hot grid himself; he “know” the wheel has to be turned to get the shock off, but he will not do it himself and inter- feres with the submissive rat’s doing it. And then, when the submissive rat finally turns the shock off, the dominant rat frequently aggresses against him. This line of research suggests an almost in— credible role of dominance in social behavior. Or- ' ganisms will subject themseh'es to painful stimuli to display their dominance. It is reminiscent of a marine sergeant who wakes the recruits in the mid- dle of a cold, rainy night, and marches them through a swamp, concurrently subjecting himself to the urinary. We are far from understanding these observa- tions, but at the risk of offending some, recall the biblical quote, “The meek shall inherit the earth." Perhaps that is nature’s principle, and dominant types will push the self-destruct button. But we can study in a rat laboratory questions such as the extent to which dominance is heritable, what en- vironmental circumstances foster its appearance, what immediate circumstances might mitigate against dominant stupidity. It would be so easy for a dominant rat to push a submissive rat out of the way and say, in effect, “Don't worry about it, [’11 take charge." It is both fascinating and ter- rifying that he does not. Were this article properly to make the case for the experimental psychology of learning and now, many more dramatic examples might be given. We ' 'now know that the range of learnable responses in- cludes the heart, the liver, and other visceral or- gans. . . . Dare to imagine training a patient to control his blood pressure. We have long known that principles of conditioning comribute to many maladaptive behaviors, but we now know that those same principles can be used effectively to modify them. . . . We now know that schedules of reinforcement can be used to increase the emission of socially acceptable behavior among people judged to be .“insane.” . . . The described work is a very small piece of a potentially very large pie. Consider one more example. Research involving both animals and humans indicates that learning is bidirectional. That is to say, any two things that occur together get asociated with each other. Suppose a person is for some reason feeling de- pressed, guilty, or just sorry for himself. Suppose further that that person then engages in some form ' of self-reinforcement—perhaps sexual relief, per- haps alcohol, perhaps any way to get a “kick.” Re- gardless. there are two clear messages. The first is that, since reinforcement increases the probabil- ity of any behavior, the person is more‘likely to feel depressed, guilty, or sorry for himself in the future. At the same time, the person is equally likely to resurrect those feelings when he engages in such behaviors. A person who daydreams when feeling lonely will learn to feel lonely when day- dreaming. A person who drinks when mad at the world will learn to be mad at the world when drink- ing. A person who engages in sex when feeling sorry for himself will tend to feel sorry for him- self when indulging in sex. And all of the effects of self-reinforcement apply equally to reinforcement provided by the environment. It is tempting to continue with this theme. For example, most parents know they can help put a child to sleep by repeating a bedside story, but few realize that they can do the same for them- selves by drinking about a story upon retiring. Contrariwise, tired students who go to sleep while reading wonder why they cannot stay awake when they try to study. The practical implications of even the few principles included here are consider- able. But there remains what, in the language of an earlier era, would be called the “364 Question.” Why is the experimental psychology of learning still confined largely to dark basements? If it has been shown that the experimental analysis of ani- mal learning can lead to the discovery of principles of value in understanding human behavior,_why is ' not a larger share of our national resources being devoted to this enterprise? . . . And most distress ing personally, why, when we assert basic princi- ples of behavior that have relevance now to the in- dividual behavior of everyone, and most especially to parents and teachers, are these not eagerly em- braced? It seems that the typical person may hear a lecture, read a book, or even take a course in the experimental psychology of learning, and come away with feelings that it was very interesting, intriguing, even exciting, but then return to the everyday world where it’s business as usual. In general, what is the plague that haunts this field? . . . ['1‘] here is always an inertia, laziness, and general resistance to change. But these are not enough. We complain about the quality of our educational systems, we lament the growing abuse of drugs. we . get adamant at the rising rate of crime, we be- came concerned about pollution, we even some- tirms admit to the problem of mental illness. Yet, while all of these are basically psychological problems, and most basically learning problems, the known fundamental principles generally are ignored. I believe that the underlying reason is . . . that psychology, as a science, proclaims as its cardinal principle that behavior is determined. It is further that the experimental psychology of learning . . . places a strong emphasis on the environmental control of behavior. The typical person feels threatened with the loss of what may seem his most cherished possessions, namely, the feeling of individual free will, self-determination, and personal and social conscience. This issue ha been brought more clearly to the fore in a recent book by B. F. Skinner (1971) en- titled Beyond Freedom and Dignity. Skinner‘s basic premise is by no means new. it has been ever one half-century since Watson . . . pro- claimed the thesis of behaviorism, and recall his claim to the effect that he could take any normal, healthy baby and control his environment to pro- duce a doctor, a lawyer, a garbage collector, or a thief. And the syndrome to which i am referring is characterized by the typical response: “Bully for you, Dr. Watson, but keep your cotton-pickin’ psychological hands off my baby!" What was clearly implicit in Watson’s proclama- tion, and what Skinner has now made dramatically explicit, is that the environment inevitably is going to help determine what happens. Return a mo- ment to the earlier discussion of speed of doing arithmetic. The contention was not simply that a teacher could arrange an environment in which speed of responding was taught; it was the much stronger point that however the environment is arranged will determine how rapidly the student learns to reapond. Hence theconclusion: Why leave to “chance” the probability that the student will learn an inefficient speed? Why not arrange the environment deliberately with response speed in mind? _ . . . .To take Watson‘s illustration, why leave to a willy-nilly environment whether a baby grows up to be a doctor, lawyer, garbage collector, or a thief? What he incomes is going to be importantly determined by his environment; why-not program the environment most adaptively? Now, parents may say, “Well, I’d be happy enough to arrange things so that he did not become a thief, but I’d rather let him decide for himself between being a doctor, a lawyer, or even a garbage collector.” To say that misses the point, but even so, think a moment longer. We now know that the upcoming generation is overloaded with doctors of philoso- phy, and undermanned with doctors of medicine. Most of those PhDs have the native intelligence to be successful MDs, and for many their environ- ments could have been arranged so that they would have been perfectly happy (and richer) as medical doctors. . . . in any event, the thesis is that behavior is de- termined and that much of it is learned through interaction with the environment. It is possible, however, that Skinner did not make that thesis palatable to most people. He may be interpreted as implying that a group of psychologists should sit in an ivory tower and direct the destiny of everyone. I do not believe that Skinner had any such intentions, but the title of his book, . . . and the way it is written may lead the public to use it as Confirming their worst fears of behavioral con— trol. The message that needs to be accepted is this: Man is not free in any absolute sense, but he can learn to be motivated to control his own behavior and can learn effective habits of self-control. Man is not born dignified, but he can learn the concept of dignity and learn to be motivated to behave accordingly. Man d0es not inherently feel a per- sonal or social conscience, but he can learn the niles that history has shown to be best for indi- vidual and cultural survival. . The point is that many of the ills of our society can be traced to the false doctrine that man is by nature “good” and has a natural free will to ex- press this goodness. This is not to say that man is by nature “bad.” It is to say that one's sense of personal freedom is in no way lost by the knowl- edge that this was learned in interaction with an appropriate environment. Dignity is no less dig- nified by the fact that the concept itself was learned and the requisite behaviors learned. Con- science is-no less personal and meaningful from the fact that our understanding of what is “right” and what is “wrong” reflects learning to behave on the basis of society’s collective drive for survival. The goal is not to get the experimental psychol- ogy of learning out ofthe basement into an ivory tower. It should be in the home. in the street, in the school; it also should be in the individual. We must conquer the traditional belief in what Skinner calls “autonomous man,” but not by threatening to deprive him of meaning and purpose. This view makes learning even more important. It is more than learning reading, writing, and arithmetic. It is more even than learning a trade or profession. Surely principles of learning can be used to im- _ prove on those, but now is none too soon to pro— mote the role of learning human values. ...
View Full Document

{[ snackBarMessage ]}

Page1 / 7

Number 11 LOGAN - FRANKKLOGAN Expenimenral Psycholoqy of...

This preview shows document pages 1 - 7. Sign up to view the full document.

View Full Document Right Arrow Icon
Ask a homework question - tutors are online