Animal Behavior

By Keller Breland, Marian Breland and Robert E Bailey

All effective modern animal training has a common basis: the discovery of "operant conditioning" by B.F. Skinner. Assisting in this research were Keller & Marian Breland, graduate students of Skinner who continued to develop the science into a systematic method for training animals. They documented their work in the seminal text, A

• Language: English
• Publisher: Storymakers, Inc.
• Release date: Oct 22, 2019
• ISBN: 9780988807969

Book preview

ANIMAL BEHAVIOR

First published in 1966 by

THE MACMILLAN COMPANY

Collier-Macmillan Canada Limited

Toronto, Ontario

Published in 2018 by

STORYMAKERS, INC.

P.O. Box 91338

Houston, TX 77291-1338

www.storymakersinc.com

© 2018 Robert E. Bailey

All rights reserved.

No part of this book may be reproduced, stored in a retrieval system or transmitted in any form or by any means without the prior written permission of the publisher, except by a reviewer who may quote brief passages in a review.

Although the author and publisher have made every effort to ensure that the information in this book was correct at press time, the author and publisher do not assume and hereby disclaim any liability to any party for any injury, loss, damage, or disruption caused by errors or omissions, whether such errors or omissions result from negligence, accident, or any other cause.

The information provided in this book is for educational purposes only. The author and publisher are not responsible for any injury, loss, damage, or disruption that may result from practicing the techniques outlined in this book. Please consult a professional before engaging in any such activities.

Cover design by Fred Stawitz.

Editing by Nancy M. Kelly, J. Derek Furstenwerth, Nedah Rose, and Fred Stawitz.

ISBN: 978-0-9888079-2-1

ISBN: 978-0-9888079-6-9 (e-book)

Printed in the United States of America.

TO OUR CHILDREN

Bradley, Frances, and Beth

New Foreword

50 YEARS LATER

More than half a century has passed since the publication of Keller and Marian’s book, Animal Behavior. Sadly, neither of the authors are alive, Keller dying in 1965, a year before the book’s publication, and Marian dying in 2001. Personally and professionally, I knew the Brelands very well. Keller and I were colleagues from 1962 until his death. I also worked closely with Marian, marrying her in 1976 and being at her bedside at her death. I have the memories of Marian, Keller, Grant Evans (Technical Director of ABE and brilliant in his own right), Kent Burgess (ABE’s Training Director and the finest all-around animal trainer I’ve ever known), philosophers, teachers, scientists all, and myself, sharing long days and evenings in the living room at the Brelands’ farm house, discussing everything from animal behavior to cosmology. Keller called these get-togethers Prayer Meetings. At these prayer meetings, I was a student, contributing bits and pieces, but mostly listening to the wisdom presented to me. I am tied to Animal Behavior in another significant way. I was privileged to be mentioned in the original book’s acknowledgements, even though most of my contributions on biology and biochemistry were made while I was the Director of Training of the U.S. Navy’s Marine Mammal Program, and not a member of Animal Behavior Enterprises (ABE), the Brelands’ company.

The Brelands told me their purpose for writing Animal Behavior as a supplemental text was to promote closer understanding and cooperation between those in the fields of psychology and biology, especially college and high school level teachers. It was the Brelands’ contention that many in contemporary psychology were describing as totally separate respondent and operant behavior, while the results of the Brelands’ own work suggested that the respondent and the operant coexisted along a continuum. Animal Behavior was not well received by the behavior analysts of the day, nor by psychologists in general, as judged by the sale of the book in psychology departments. The book fared better in biology departments and most of the mail Marian received was from biology teachers and students. In particular, those biologists with ethological interests thought Animal Behavior held special significance.

In the late 1970s Marian and I considered a rewrite, a complete updating, of the Animal Behavior book. By that time, Marian, I, and our Animal Behavior Enterprises (ABE) team of about 50 technicians had gathered information on the training of many thousands more animals and a hundred more species. Perhaps more importantly, we had completed several open or free environment programs where we could study the responses of many species repeatedly and for long durations exposed to a wide range of stimuli, some quite severe. We believed these programs were unique for that time.

In addition, Marian and I decided to address in detail issues brought up by some concerning the Brelands’ landmark paper, The Misbehavior of Organisms (Breland & Breland, 1961.) We decided to present much more data from ABE animal training records (no actual training data, just training results, were included in the original paper), including some field observations collected by myself over the previous 20 years.

Marian contacted the MacMillan company, the publishers of the original work, to ascertain their interest. MacMillan officials expressed no interest in an Animal Behavior rewrite, nor for a second edition of the original book. Kindly, the company gave to us the rights for republication and sent us the original photo plates to aid in the printing. We began a chapter by chapter rewrite. Marian and I were not using word processors in the 1970s. Further, in the case of the rewrite manuscript, we mostly wrote as we had for decades, either on a typewriter, or more often on legal-sized writing pads. As an historical aside, Marian had developed her own version of Gregg shorthand and could record dictation at a blazing speed, allowing us to document our thoughts at home, in the office, or even while traveling in an auto.

In 1989, Marian and I suffered a devastating fire that completely destroyed our home. Stored in our home were most of ABE’s records and Keller and Marian’s personal histories, gathered from 1916 through 1988. The ABE data records lost include the original training information used when writing The Misbehavior of Organisms (Breland & Breland, 1961) from 1957 to 1960. It was this information plus later training records we intended to use in the new edition to more fully develop the old Breland idea of the operant-respondent continuum. Of course, also lost was the entire manuscript for the new Animal Behavior book. We made a halfhearted attempt to rewrite what was lost, but we were overwhelmed by the enormity of losing the more than 10 years’ work on that book plus three others on which we had been working.

In 1990, Marian and I closed ABE’s doors as we began our retirement from production animal training. Marian continued teaching psychology and statistics at Henderson State University and I began doing video production work. In 1994-95, Marian and I began teaching animal trainers using chickens as teaching models. These training classes were similar to classes taught to ABE employees and clients of ABE since 1947.

Fast-forward to 2003, two years after Marian’s death: I resurrected the Animal Behavior book rewrite. After several years of work, I had to admit I had little heart to finish the book. Further, I had little success trying to further the other books Marian and I had started. But there was another dilemma not tied to emotion: the loss of 50 years of data to the household fire in 1989—I could no longer support my words with the hard data gathered so meticulously by ABE personnel. The flames in 1989 had changed my hard data (trials, dimensions, subjects, results) to hearsay (these are my recollections or the recollections of others with whom I worked.) As a scientist, and an animal trainer teacher-practitioner, publishing anecdotal evidence without severe qualification is a no-no.

Recently, another thought struck me: Animal Behavior, the original 1966 book by the Brelands, captured the experiences and contemporaneous thoughts of two people who likely had more experience changing the behavior of more animals, representing more species, than anyone else up to that time; the Brelands’ experiences were unique inside or outside academia. In my 2005 publication Operant Psychology Goes to the Fair: Marian and Keller Breland in the Popular Press, 1947–1966, (Bailey and Gillaspy, Jr., 2005) I suggest that contemporary psychologists were often not exposed to the Brelands’ scientific and applied work, and when exposed, were not receptive to the significance of the Brelands’ findings as spread through public media or even through their more limited number of academic publications. Perhaps today there are those who might like to look through a window at the vision of two very bright individuals outside of academia who persevered to develop a modern technology – A New Field of Applied Animal Psychology (Breland & Breland, 1951.)

My thinking was maybe a complete rewrite of Animal Behavior is not the best contribution I could make now; perhaps it would be best to simply reprint what the Brelands wrote, exposing an entire new generation of scientists and practitioners to what the Brelands wrote over a half-century ago, as well presenting selected classic work by those who preceded them. Thus, here is the book, Animal Behavior, exactly as the Brelands wrote it in 1964-65 except for a short introduction, written by myself to give more background to their work and a correction of some misspellings not caught during the original proofing. Left intact are the Brelands’ words, grammar, writing style, and syntax.

Bob Bailey, Sc. D.

New Introduction

TOWARDS A SCIENCE OF BEHAVIOR

Many readers of this book may be new to the field of psychology or perhaps students of other disciplines. Hence, they may be unfamiliar with the roots of behavioral psychology. For those readers, I thought it best to add a short tutorial on the rise of behavior—what organisms do—as a field of study. Today, we take it for granted that psychology studies how animals, including humans, respond to the world around them. This has not always been so and when behavior was studied, the studies and conclusions based on those studies did not always depend on what we might call good science.

As most sciences mature, they evolve into or spin off an applied science that we may call technology which has the goal to control their subject matter. Physics exhibits its control in various branches of mechanical, civil, and electrical engineering, as well being fundamental to aviation and the exploration of space. Chemistry has its chemical engineering and a huge realm of economically vital chemical manufacturing. More recently, biology has diversified into such fields as microbiology and physiology. Microbiology, including modern genetics, aids in producing pharmaceuticals and disease resistant plants. Physiology is crucial for developing new products in plant and animal husbandry and, of course, improving the quality of human and veterinary medicine.

Psychology as a discipline began in Europe in the mid-1800s. Most in the early field of psychology were not particularly interested in either studying or controlling behavior; these psychologists were proud of their pure science. The changing of behavior or the notion of usefulness held little importance. When psychology first became a science in the mid-1800s, the great German scholar Wilhelm Wundt thought that learning was not a proper subject for scientific study. Human learning, claimed Wundt and others, could not be directly studied because the subject matter was contaminated by the material that the subject had already learned; however, other psychologists disagreed. One of the most famous of these psychologists was another German named Hermann Ebbinghaus, who devised the use of nonsense syllables for studying learning. Words like GOK, DYL, ZUX, KIV, and FUB were indeed fresh material. They had no associations in the minds of the experimental learners; they were uncontaminated. And so Ebbinghaus established the first laws of human learning. Ebbinghaus and others who worked with human verbal learning established many principles which were later applied to education and experimental psychology; however, these early European psychologists contributed little to the push to study learning in animals. It was probably Darwin and Wallace publishing their concept of evolution and natural selection that inspired interest in what came to be comparative psychology, including animal learning and intelligence.

Much of early psychology was laced with anecdotal evidence of animal intelligence, reasoning, feelings (or sensibilities,) and animal learning. Countering this was a contemporary, the naturalist C. Lloyd Morgan. Morgan explained the reported human-like attributes as humor, pity, grieving, sense of aesthetics, understanding, thinking, reasoning, and judgment based on much simpler principles, and was moved to put forth a principle which came to be known as the law of parsimony, or Lloyd Morgan's Canon, a modern version of Occam's razor. This principle stated that it was poor scientific practice to attribute an animal's behavior to the operation of some higher faculty if it could be explained as the result of some simpler process. To his credit, Morgan did the first large-scale, though imprecise, experimental studies of animal psychology and anticipated Edward L. Thorndike's trial-and-error learning, thus becoming one of the first true animal psychologists.

PAVLOV AND CLASSICAL CONDITIONING

Around the turn of the century, a famous Russian physiologist, Ivan Pavlov, was studying digestive reflexes. Pavlov’s conditioning procedures were found to apply to many bodily processes, including emotional responses, and hence are important for psychology. Today these processes constitute what is called Pavlovian, classical, or respondent conditioning. Below is a brief explanation of what Pavlov learned:

Conditioning: The reinforcement process consists of pairing a neutral stimulus (NS) with an unconditioned stimulus (UCS.) By doing so, the NS would acquire the power to bring out the response, thus becoming a conditioned stimulus (CS) and creating a conditioned reflex (CR.)

Extinction: If the CS and UCS pairing stopped, the CR would gradually decline and disappear.

Generalization: If the initial CS was a 1000 cycles per second (cps) tone, and then a 1500 cps tone was presented, the CR would occur to the 1500 cps tone as well.

Discrimination: If you now reinforce (by pairing CS and UCS) only the 1000 cps tone, the CR to the 1500 cps tone will gradually extinguish and disappear.

Pavlov received a Nobel Prize for his physiological work on digestive reflexes. He was slow to publicize his work among psychologists. Later, he did come to realize some of the psychological significance of his work, especially after his work became famous in the United States and became a language and a scientific foundation for the new kind of psychology brewing in that country.

ON THE OTHER SIDE OF THE POND

Meanwhile, animal psychology was welcomed as a useful tool by the growing school of functionalism in the U.S.A.; that is, What does behavior do for the animal? Many important animal studies were conducted by early American psychologists, including The Animal Mind by Margaret Washburn, one of the early female Ph.D. candidates in psychology. Particularly, we can point to Edward L. Thorndike's Animal Intelligence, published in 1898.

Thorndike was a young student at Harvard, interested in animal learning. He experimented mostly with cats. He put these cats in small boxes. If you have ever put a cat in a box, you may recall an interesting experience. Even a cat in a sack can be a challenge. It does a great deal of scrambling around, yowling, and trying to get out; getting out was the name of Thorndike's game. Each of the boxes had a different way of escape. The cat had to make a response which would trigger a mechanism to open a door. During of all the scrambling and carrying-on, the cat would hit Thorndike's trigger device and escape. The response might involve hitting a lever, pushing a plate, or pulling on a string. Then the cat was rewarded not only with its freedom, but with a drink of warm milk.

When graphed, these results demonstrated Thorndike’s famous learning curve. What the curve showed was that, as days went by and each cat gained more experience in its particular problem box, the time to escape dropped rapidly at first, then more slowly as the cat eliminated unsuccessful responses. Finally, by trial and accidental success, the cat was getting out almost immediately. With time, the ineffective responses dropped out of the cat's behavior. Incidentally, Thorndike did most of these wild and woolly experiments in the basement of William James' house because Thorndike's landlady took a dim view of cats. Perhaps science is not always elegant or pretty. Thorndike's theory has come to be called law of effect.

Shortly after the time of Thorndike's experiments, in 1901, another psychologist named W. S. Small took a lead from a famous feature at Hampton Court in England known as the Hampton Court Maze. In these days we would call it a tourist attraction. It was an enormous growth of tall hedge covering more than an acre, laid out in intricate patterns with true pathways and blind alleys. It was common that a person starting at the entrance lost his way, entering many blind cul-de-sacs before finally emerging at the end. A guard posted on a high perch guided those who panicked and became really frightened at being lost. W. S. Small copied the pattern of this famous maze, miniaturized it, and used it to study the learning abilities of laboratory rats. His rat maze (and others modeled on it) became probably the most popular apparatus for the study of animal learning for over half a century. The maze was used to study the behavior of chickens, pigs, sheep, people, and most particularly, the white laboratory rat.

Thousands of rats were to be run through this and other types of mazes. Gradually, by the 1920s, the rat was becoming the animal of choice for psychological laboratories. This development, incidentally, just about marked the demise of truly comparative psychology for many years to come. It wouldn't be until the 1950s that the European ethologists—zoologists who specialize in animal behavior—would begin to influence the course of experimental psychology. Ethology began in Europe in the 1920s and 1930s, and was popularized by such researchers as Tinbergen, Hinde, and Lorenz.

THE RISE OF BEHAVIORISM IN THE NEW WORLD

At about the same time Pavlov was at the peak of his work, John Broadus Watson was in the United States revolutionizing the field of psychology by tossing out all reference to the mind and any processes that could not be objectively observed. Watson was a student of the same pragmatic functional school as Thorndike. He advocated studying behavior only and his way of looking at psychology came to be known as behaviorism. By the 1920s behaviorism, led by its founder, was becoming very popular. Pavlov's discoveries were enough to arouse the attention of American psychologists who were scrambling around, trying to find some way of studying behavior objectively. When Pavlov's works became known in the U.S.A., Watson seized upon them and tried to make Pavlovian or classical conditioning the model for the objective study of all behavior. Many of Watson’s experiments dealt with human fears and other emotional behavior, particularly those specific and often irrational fears we know as phobias. Many psychologists of the time hailed his work as the most important development since psychology entered the laboratory. Others called it S-R or knee-jerk psychology and criticized it for oversimplifying behavior.

Watson soon left the academic world (or, rather, was driven out for some personal indiscretions) to go on to a very successful advertising career. You may thank Watson the next time you are assaulted by a hard-sell commercial, attempting to convince you how much you need something, such as a new deodorant. His psychological work was carried on and advanced by several later behaviorists, most of whom tried further to apply Pavlov’s and Watson's methods and ideas to all human and animal behavior as the fundamental learning process.

Watson had some success; dealing with emotional responses, for example. But as he and later psychologists like Clark Hull and his followers tried to expand the usefulness of their model they ran into some problems. What rapidly became apparent was that there is a huge amount of human behavior that does not follow the reflex paradigm. Most of our everyday behavior does not consist of physiological reflexes. In most cases, you would have a hard time pointing to any kind of a stimulus that elicits a response in the same way that an unconditioned stimulus elicits an unconditioned response. The forced nature of the Pavlovian reflex is characteristic of physiological reflexes and of emotional reflexes, which are physiological in nature. However, no simple stimulus comes on to cause you immediately to go to a movie in the way that a lemon makes your mouth pucker. If you pushed it far enough, you might be able to come up with some stimuli (My boyfriend asked me to go, and so on), but the analysis becomes far-fetched. Instead you ask such questions as, Under what conditions do you go to a movie? Some of Clark Hull's students like Mowrer and Neil Miller tried to accommodate everyday behavior by making some adaptations of the Pavlovian paradigm or system of thought, but with a few exceptions, their formulations have largely died out today. One of those who undertook a different kind of analysis was Burrhus Frederic Skinner.

THE EMERGING FIELD OF BEHAVIORAL SCIENCE AND TECHNOLOGY

B.F. Skinner was one of a group now called neobehaviorists, the continuers of the behaviorist tradition. Operant conditioning, or, as it is frequently called now, behavior analysis, is probably the most well known and most effective method for studying and changing behavior, both for humans and for the lower animals. If a date had to be picked as the beginning of modern behavioral psychology, it would probably be 1938, when Skinner's historic book The Behavior of Organisms was published. Other psychologists were working hard to bring behaviorism to its mid-century position of leadership in psychology. But it was chiefly Skinner, and later his followers, who developed modern behaviorism to its present useful state.

Skinner came to psychology from a background in English. He had intended to be a writer, but discovered he had nothing to say, so he went back to school at Harvard to study psychology, which he approached from the direction of the physiology of behavior. His approach to science and to psychology was colored by his friend, Fred Keller, and his exposure to earlier philosophers: Bacon, Mach, and Bertrand Russell. Skinner was also strongly influenced by the zoologists Loeb and Crozier and by Sherrington, the brilliant British physiologist who initially described many of the laws of the reflex. Along the way, Skinner encountered Pavlov's work and realized both its promise and where it needed help. From this vantage point he undertook an experimental program to develop a new approach. Out of this program grew what he called operant conditioning, the basis for most of the behavior modification or management methods we will be discussing.

Skinner's analysis originally followed the reflex terminology. He wished to show how what he now called operants could be described using the reflex paradigm. These operants were bits of voluntary, overt, large muscle, human or animal behavior. Skinner later dropped the reflex paradigm; however, he and his followers continued to use, and still do use, the terms extinction, discrimination, generalization, and others that Pavlov had used. The commonality of terms used by Pavlovian and Skinnerian paradigms has sometimes led to more than a little confusion in the minds of beginning psychology students and the general public.

An analysis like this was to prove the foundation of an experimental program which was to be continued by Skinner himself for 53 years, until his death in 1990. This program, developed in the laboratories at the universities of Minnesota, Indiana, and finally Harvard, prospered and bore many fruits. Not only did it attract hundreds of students and colleagues to concentrate on the laboratory aspects, but it also branched off into numerous applied disciplines, where psychologists trained by Skinner and his first and second generations of students took the now well-tested principles into the field and put them to practical use for solving problems. Two of the first-generation students were Marian Kruse (later Marian Breland), and Keller Breland, the authors of the book you are about to read, and who were the first to apply Skinner’s ideas to commercial and large scale applications.

In Skinner's system, known as operant conditioning or behavior analysis, researchers have taken human and animal behavior into the laboratory and submitted it to the same rigorous experimental analysis that characterized early research on simpler stimulus-response behavior. As a result, behavior analysis can handle human behavior in many of its aspects. Operant psychology even deals with choice behavior, inner events, and self-control. With all these developments, it is the most powerful tool available for effective behavior change.

WHY THE BRELANDS?

Marian Kruse was a language student at the University of Minnesota. She could speak five languages. She took a class (proseminar) with Skinner in 1937 as an elective. Skinner soon invited her to become his student and later, his assistant. Keller Breland, an industrial psychology student with a practical bent, joined the team in 1940. That same year, Marian and Keller married. WW2 began in December 1941. Skinner soon proposed to the military that he could train pigeons to guide a specially designed bomb called Pelican. The Pelican program began in early 1942, and the birds were reliably and accurately guiding the bombs in simulators by 1943. The program demanded complex behavior and mechanical processes. The Brelands were deeply involved in developing this revolutionary animal-machine system with Skinner. The rapidity with which the Pelican program developed was breathtaking. The program, and the people who worked on it, are stories all their own.

The success of Pelican inspired Marian and Keller to promote the new technology by creating a commercial enterprise based on Skinner’s work. In 1943, the Brelands bought a farm in Mound, Minnesota, using money borrowed from Marian’s mother. They began building equipment and training animals in their barn. Much of what they did for the first year was based on what they had learned working for Skinner. When they realized that a business had to have paying customers and not many people would pay to see rats in boxes, they began using many other animals and radically changing the working environment. However, the Brelands still employed most of the training efficiencies they saw in Skinner’s laboratory procedures, including the use of electric feeders. They began training many kinds of animals in groups. This group training, and other efficiencies, presaged the Brelands’ future as pioneers in the mass production of precise and reliable animal behavior for commerce. In 1943, they named their new company Animal Behavior Enterprises and began to prepare for the end of the war, and the beginning of their efforts to bring operant conditioning to the world at large.

By 1950, the Brelands were moving to Arkansas and producing hundreds of trained animals of many species and their business was nationwide. A more detailed account of their early growth is chronicled in the video Patient Like the Chipmunks (Bailey & Bailey, 2010.) They were making increased demands on the