The Barrett Initiative team would like to thank Dr. Carl Binder for his help with this page. For more resources, check out his website at http://fluency.org
Dr. Beatrice H. Barrett’s Enduring Legacy
This archive includes publications from academic journals and books, unpublished manuscripts, and internal publications from two important research sites in Waltham, Massachusetts: The Behavior Research Laboratory at Harvard University (at the Metropolitan State Hospital) and the Behavioral Prosthesis laboratory at the Walter E. Fernald State School.
Drs. Ogden Lindsley and B.F. Skinner co-founded the Behavior Research Laboratory to study human operant behavior with psychiatric patients—the first lab to do so. Although Dr. Barrett first began as a postdoctoral researcher in the Behavior Research Lab in 1960, she secured funding in 1962 to establish her own lab—the Behavioral Prosthesis Laboratory—across the street at the Fernald State School, which she oversaw until it closed in 1992.
These research sites established a critical research base by replicating basic operant research with humans (Behavior Research Laboratory) and extending Skinner’s natural science to work in clinical settings (Behavioral Prosthesis Laboratory). Chief among the methods adapted from Skinner’s work were the use of response rate as the fundamental behavioral datum and the cumulative record to display behavioral data. These practices permitted the moment-to-moment measurement and analysis of behavior, detail not possible in other paradigms. Barrett spent her career researching stimulus discrimination and differentiation, reinforcer assessment, programmed instruction, classroom teaching for children with developmental disabilities, Precision Teaching (with collaborator Dr. Carl Binder), functional assessments of children with severe and profound disabilities, as well as staff, teacher, and parent training. From an early point in her career, she saw clinical potential in the measurement practices used in Dr. Lindsley’s human operant research and endeavored to apply them to behavioral assessment to better understand learning difficulties in adults and children. This archive organizes the many impacts of Dr. Barrett’s work in four main areas:
1. Behavior Analytic Neuroscience
2. Behavioral Measurement
3. Simultaneous differentiation and discrimination (SIDAD) procedure
An Interview with Dr. Beatrice Barrett from the Bailey Interviews
In the early nineties, Bob and Marian Bailey recorded detailed interviews with prominent behavior analysts. Recently, the team at the Constructional Approach to Animal Welfare and Training has endeavored to digitize and make these videos available for generations of behavior analysts to come, including this interview with Dr. Barrett.
Dr. Barrett’s Vision For Behavior Analytic Neuroscience
With emerging technology, Dr. Barrett saw the possibilities of expanding the behavioral level of analysis to include neuro-physiological changes functionally related to behavior. While discussing her endowment made to the Behavior Analysis department at the University of North Texas, she detailed her vision in exquisite detail. The following are quotes by Dr. Barrett from an unpublished document we use to guide our research:
“A comprehensive account of human behavior will encompass not only relevant range of visible social and cultural variables, but also the neurophysiological changes demonstrated to be functionally related to behavior actions. The gaps pointed out by Skinner (1989) between (1) stimulating environmental action and response of the organism, and between (2) consequences and resulting in behavior, must be filled to complete the account (not to make a new account of the same thing). Brain Science can do this. A partnership with behavioral science is necessary. The operant can be the unifying concept to focus the neurological with the behavioral sciences. It has potential for establishing a fundamental commonality of purpose between behavior analysts and neuroscientists. We must chart the limits of the operant.”
“We must understand the relation between behavior that we can see with our own eyes, whether amplified by instrumentation or not, and brain activity. We know that nerve cells can be conditioned. We have recent evidence that behavioral activity can generate new brain cells, but we still don’t have a clear picture of the synchronous relation we assume exists between what we do and what our brain does in close temporal relationship with our actions–both public and private.”
“The operant is our path toward that relationship between person activity and neural activity. Neuroscience as of now (1/1/01) has not developed, evolved, or adopted a unit of its subject matter that meets the measurability criterion of scientific analysis. Behavior analysis does have such a unit. Its limits of applicability in the analysis of behavior are as of yet unknown. What it can encompass depends on the astuteness of the investigators in arranging visible measured avenues of exchange between the molar and molecular levels of human understanding.”
The Importance of Measurement in Behavior Analysis
Measurement in Skinner's Natural Science of Behavior
Operant research involves the study of learning processes in which behavior is sensitive to, or affected by, its consequences (i.e., event(s) produced by responses). While at Harvard University, B.F. Skinner studied operant processes in a variety of organisms. His lab innovated numerous technologies and conceptual foundations related to the measurement and analysis of behavior. These innovations allowed for an extremely sensitive measure of behavioral changes in real-time. In fact, Skinner believed rate of response (the measurement unit made up of the number of behavioral responses per unit of time) and the cumulative recorder were his most important contributions to behavior analysis.
Behavioral measurement was a critical focus of Barrett’s work. The sensitivity of Skinner’s measurement technologies served as a foundation for Barrett’s research on stimulus discrimination and differentiation in humans—fundamental processes underlying learning ability. She and others envisioned diagnostic behavior laboratories that could help pinpoint behavioral performances and preferences under a range of conditions. If accomplished, they believed this type of facility could provide a wealth of information about human learning unavailable to clinicians through the use of traditional psychometrics.
Operant Conditioning and Rate of Response
To understand Barrett’s clinical applications of Skinner’s natural science of behavior, an introduction to how operant behavior is studied in the laboratory may be helpful. First, the development of the operant conditioning chamber occurred during Skinner’s years as a doctoral student at Harvard as he attempted to obtain a measure of discrete behavioral responses. The design of his apparatuses went
through many iterative changes before he landed on his final design (for a detailed discussion of the development of the operant chamber, see Skinner 1956). Over the years, operant conditioning chambers came to automate stimuli presentation, consequence delivery, and the collection of behavioral data. In short, the device provides a small controlled space designed to minimize environmental interference to an experiment, while presenting stimuli in a systematic fashion to occasion responses on various manipulanda (e.g., levers, buttons, etc.).
The research used rather simple and repeatable responses to study how the response rate of a particular response changes as a result of learning or due to changes in experimental conditions (e.g., alterations to schedules of reinforcement, changes over time to discriminative stimuli, states of deprivation, etc.). These simple responses were recorded on cumulative records via electrical circuits. Dr. Lindsley eventually adapted the operant chamber for human use, permitting the systematic replication of operant research with humans. The operant conditioning chamber serves as a foundation for all behavior analytic research and many behavioral assays in pharmacology and across the neurosciences. Many scholars consider the technological development of the operant chamber as paradigmatic, as the methods of basic psychological research changed thereafter.
While the operant chamber facilitated behavioral experiments with animals, measurement tools were needed to record the organism’s behavior. B.F Skinner invented another technology during his time at Harvard: the Cumulative Recorder. This measurement tool functioned as
essentially a “behavioral microscope,” permitting researchers to see changes in behavior in real-time (i.e., as a result of learning). Barrett and others saw how this technology could be used to study human behavioral differences. With this technology, human behavior could also be put under the microscope so to speak.
2. In B.F. Skinner’s book Behavior of Organisms (1938), he describes his initial cumulative records:
The number of responses per minute represented by each slope is indicated. The actual records are step-like. The movement of the lever is recorded electrically as a graph of the total number of responses plotted against time. The required apparatus consists of a slow kymograph and a vertically moving writing point. At each response the point is moved a uniform distance by an electrically operated ratchet. A step-like line is obtained, the slope of which is proportional to the rate of responding. The speed of the kymograph and the height of the step are chosen to give a convenient slope at the more frequent rates of responding.
The greater the response rate, the steeper the slope of the line produced. Other useful features were gradually introduced, such as the ability to reset the pen to its starting position (or baseline) after a particular event, the marking of a slight “pip” to indicate reinforcement, or other slight changes to the record noting a stimulus change or the occurrence of other experimentally important events. Eventually, Ralph Gerbrands, a scientific instrument maker at Harvard, started a company to mass produce operant conditioning chambers, cumulative recorders, and other equipment for psychological experiments.
Although Lindsley had begun to develop the Simultaneous Differentiation and Discrimination (SIDAD) apparatus and procedure before Dr. Barrett arrived at Harvard, she saw uses for Lindsley’s work in the area of clinical assessment. The data obtained during a SIDAD session provides an exquisite level of detail of human learning. Towards the end of her life, she envisioned how SIDAD could be of value to neuroscientists studying the effects of learning variables.
The SIDAD procedure aimed to explore the acquisition of stimulus discrimination and response differentiation using methods developed in the operant laboratory and extended to humans. Response differentiation and stimulus discrimination are two components necessary to make more complex operant learning possible. Before this, clinicians lacked a way to measure important component responses required for operant learning. SIDAD offered a diagnostically-oriented experimental analysis derived from single subject free operant methodologies. Most importantly, SIDAD assessed an overall ability to learn through operant methods. Using rate of response and several cumulative recorders, researchers obtained a rich datasets permitting analysis of how these important skills develop over time, potential issues hindering the acquisition of discrimination and differentiation, and an overall analysis of the length of time needed for a learner to acquire these responses. Notably, SIDAD provided the means for researchers and clinicians to measure the performances of children and adults lacking vocal-verbal repertoires.
At Harvard, Dr. Ogden Lindsley wrote a dissertation on canine operant behavior. Around this time, researchers were focusing on extending B.F. Skinner’s research to other animals with the ultimate aim of demonstrating operant principles in human subjects. As a first step towards this, Dr. Lindsley modified the operant chamber to accommodate lever pressing by a dog. After accomplishing this demonstration, he developed an operant chamber and manipulandum appropriate for human use. Seeing that an operant paradigm is well-suited for human performances, Dr. Lindsley set his sights on more complex arrangements, leading to the development of the Simultaneous Discrimination and Differentiation (SIDAD) procedure and apparatus and Barrett’s ultimate involvement in the project.
The original SIDAD apparatus consisted of a small room with a chair and a single wall-mounted panel featuring two plunger manipulanda installed side-by-side below two lights. A small tray to the side of the panel facilitated reinforcer delivery
Four responses are possible in the SIDAD procedure. During a session, each light turned on for one minute on an alternating schedule. The participant could use each plunger independently or use both at the same time. Reinforcement was programmed on an FR10 schedule for only one of the four responses: pulling the left plunger when the left light was illuminated. None of the other three responses were reinforced during a SIDAD session.
This procedure used five cumulative recorders to record all possible responses:
1. Pulling the left plunger when the left light was on
2. Pulling the right plunger with the left light on
3. Pulling the left plunger with the right light on
4. Pulling the right plunger with the right light on
5. All simultaneous responses on any plunger
This design allowed researchers to determine whether response differentiation and discrimination were present in the repertoire of the learner—fundamental processes underlying operant learning. This procedure also allows researchers to observe individual patterns of learning in stages over time. This observation led Barrett to use the SIDAD apparatus as an assessment tool to evaluate learning difficulties, similar to various forms of medical assessment.
Dr. Barrett endeavored to bring Skinner’s natural science of behavior to education. Among her many accomplishments is spearheading the application of free-operant methods to teaching. Her classrooms at the Walter E. Fernald State School were among the first to make use of Programmed Instruction and Precision Teaching.
Contributions to Education
At the Fernald State School, Barrett was an early proponent of applying behavioral principles to teaching. Her programs at the Fernald School were among the first to extensively use teaching machines and programmed instruction. As a result of an ongoing collaboration with Ogden Lindsley (who was by now training teachers at Kansas University) and Dr. Eric Haughton, Dr. Barrett and Dr. Carl Binder began using Precision Teaching in their classrooms in 1974.
Additionally, one of the most referenced articles in behavioral education came from Barrett’s lab. Barrett (1979) illustrates the importance of the behavioral frequency over percent correct measures. This study evaluated the performances of state school students (aged 12-54), public school students (aged 5-7), and non-disabled adults on basic component skills. While all groups were 100% accurate at completing these tasks, large differences are observed in the frequency measures (include graph). This paper has had an indelible impact
on all behavior analysts working in education, and raises important questions about how learning is measured in educational institutions.
She promoted the use of free-operant methodologies broadly across the American educational system. In 1977, she wrote a still-relevant chapter on how behavioral principles could be applied in learning environments. This chapter displays a level of sophistication and understanding of behavioral principles beyond many of her contemporaries. Anyone working today in applied or clinical settings would benefit from reading this classic chapter (link to chapter).
In 1991, she chaired an ABA task force on the right to effective education with many experts in the behavior analysis field on the topic of education and schooling. This task force published an important 1991 paper outlining guidelines for effective education derived from behavior analytic research on effective teaching and educational systems. This group asserted that American youth were entitled to effective educational experiences. Students needed to learn skills that were useful to them long after schooling was over and also ethical conduct, but also cultural values and ethics of long term benefit to both the individual and society as a whole (Barrett et al., 1991, p. 79).
Towards the end of her career, she was asked to write a forward to a new edition of B.F. Skinner’s Technology of Teaching. Rather than writing a concise forward, she wrote a new book to accompany Skinner’s important text published 30 years prior. Although America’s educational demands had changed by the early 2000s, she eloquently reiterated how Skinner’s ideas on education remained helpful to a new generation of educators and behavior analysts.
E-book available from the Cambridge Center for Behavioral Studies.