- Behavior analysis posits that people and organisms’ environments can be arranged so that desirable behaviors become more probable and undesirable behaviors become less probable.
- Behavior analysis has three main branches: behavioralism, which encompasses the philosophy of how behavior analysis should be conducted; the experimental analysis of behavior, which describes basic research into behavior; and applied behavior.
- The 20th century saw a shift from describing behavior internally to describing behavior empirically and observable; B. F. Skinner provides the foundation for how psychologists think about behavior analysis today.
- Commonly applied behavior analysis methods for teaching behavior involve chaining, prompting, and shaping.
Table of Contents
Behavior Analysis Defined
Behavior analysis is a science based on the philosophical principles of behaviorism. The key point of behaviorism is that what people do can be understood.
According to the Behavior Analyst Certification Board
“Behavior is a product of its circumstances, particularly the events that immediately follow the behavior. Behavior analysts have used this information to develop numerous techniques and treatment approaches for analyzing and changing behavior, and ultimately, to improve lives.
Because this approach applied behavior analysis (ABA) is largely based on behavior and its consequences, techniques generally involve teaching individuals more effective ways of behaving and working to change the social consequences of existing behavior”.
About behavior analysis – Behavior Analyst Certification Board. Bacb.com. Published 2017. https://www.bacb.com/about-behavior-analysis/
Scientists, in the view of behaviorism, can look at behavior as a process that can have any number of causes, internal or external.
Behavioral analysts believe that people can arrange their environments in such a way that desirable behaviors become probable, and undesirable behaviors become less probable (Heward and Wood, 2003).
Cooper, Heron, and Heward, 2020 consider science to have several key characteristics: description, prediction, and control.
- This means that applied behavior analysis, as a science, intends to undergo systematic observations (creating collections of facts about observed events that can be quantified, classified, and examined for relations with other known facts);
- create predictions as to the probability that a certain event will occur;
- and then develop a set of functional relations that provide an understanding valuable to the development of technologies.
A fundamental relation comes about when a well-controlled experiment reveals that a specific change in one event can be produced reliably by manipulations of another event and that the changes in the original event are unlikely to be the result of other, confounding variables.
Techniques and Strategies
Behavior analysts use a variety of techniques to change behavior. These include chaining, prompting, and shaping.
Chaining involves breaking a task into smaller components, and can be used to help people gain proficiency in complex, multi-step directions.
The idea of chaining is based on the behavior chains, which are strings of individual behaviors that, when connected together, create an end-behavior.
For example, the behavior chain of putting on a coat could involve the verbal stimulus, “Put on your coat,” obtaining the coat from the closet, having the coat in one’s hands, putting one arm in a sleeve, putting the other arm in a sleeve, the coat being on, zipping up the coat, and finally instructor praise (Cooper, Heron, and Heward, 2020).
There are three main methods of creating behavior chains: forward chaining, total-task chaining, and backward chaining.
1. Forward Chaining
In forward chaining, the behaviors that make up a task are taught in their naturally occurring order. For example, a child learning how to tie their shoes may receive reinforcement when the very first step, “pinch lace, “ is performed correctly three times.
The instructor may give the child additional reinforcement when they perform the first and second times correctly in sequence another three times and so on. As someone master’s one skill, they can link it to the next skill, and this series of skills can eventually constitute a completed task.
McWilliams et al. acknowledge two main advantages to forward chaining: linking smaller chains into larger ones, and its ease of use (1990).
2. Total Task Chaining
In total task chaining, the instructor gives the learner training on each step in the task during every session, and assists with any step the person cannot perform independently.
The instructor trains the learner with needed assistance until the learner can perform all of the behaviors in the sequence independently.
In one study using total task chaining, Werts, Caldwell, and Wolery (1996) taught skills such as operating an audiotape, sharpening a pencil, and using a calculator to elementary-school students with disabilities.
The researchers probed the students on their ability to perform the entire task response chain, a peer who could do the task chain demonstrated the chain in its entirety while describing every step, and the researchers prompted the learning student to perform the chain.
All three students learned how to complete the response chain after peer modeling (Cooper, Heron, and Heward, 2020; Werts, Caldwell, and Wolery, 1996).
3. Backward Chaining
Backward chaining involves an instructor initially performing all of the behaviors involved in a task except for the final behavior, which the learner performs.
Once the learner learns how to perform the last behavior, the instructor then performs all of the behaviors in the chain except for the final and penultimate behaviors and so on until the learner can perform all of the behaviors in the chain.
Pierrel and Sherman (1963) notably taught a white rat, Barnabus, to climb a spiral staircase, push down and cross a drawbridge, climb a ladder, pull a toy car by a chain, enter the car and pedal through a tunnel, climb a flight of stairs, run through an enclosed tube, enter an elevator, raise a miniature Brown University flag, exit the elevator, and press a bar to receive a pellet of food using backward chaining.
Another notable example of backward chaining was Hagiopian, Farrel, and Amari’s attempts to teach a 12-year-old male to inject food by mouth which he had previously expelled.
The researchers began by teaching the boy to swallow through placing a syringe in his mouth.
Eventually, they delivered reinforcement to the boy only once all three “drinking” responses occurred (accepting the juice and swallowing) (Hagiopian, Farrel, and Amari, 1996; Cooper, Heron, and Heward, 2020).
Backward chaining also has a modification, backward chaining with leap ahead, where the instructor does not instruct the learner on every step of the chain.
This is particularly effective when the learner has mastered particular steps of the chain ahead of time (Cooper, Heron, and Heward, 2020).
Prompting, meanwhile, involves using a prompt to trigger a desired response.
It can be used in combination with chaining or shaping, and encourages the learner to perform a task until they learn how and when to do it.
Prompts can involve instructions, demonstrations, touches, or other stimuli.
Shaping describes gradually altering a behavior until it becomes the desired behavior.
For example, a language therapist may use shaping when they develop speech with a client by first reinforcing lip movements, then sound production, and finally words and sentences.
The number of adjustments and approximations needed to create a desirable behavior depends on that behavior’s complexity.
Shaping can be useful in cases where behaviors cannot be easily learned by instructions, incident experience, or prompts (Cooper, Heron, and Heward, 2020).
One technique that instructors can use in shaping is differential reinforcement.
In differential reinforcement, instructors provide reinforcement for responses that share a predetermined dimension of quality with the desired behavior.
For example, a parent may pass food to a child at the dinner table when the child says “please,” but not otherwise. Because only the responses that share this similarity with the desired behavior are reinforced, other responses are performed less and less – they become extinct (Cooper, Heron, and Heward, 2020).
In order to shape a behavior, the instructor must determine a set of gradually changing criteria for reinforcement for approximations that look closer and closer to the desired behavior.
For example, a teacher may ask a child who normally speaks at the inaudible volume of 45 decibels to speak at 55 and finally 65 decibels, as demonstrated in Fleece et al. (1981).
The researchers found that shaping increased the children’s voice volume and that this increase persisted over four months.
Shaping can take place over several dimensions: topography, frequency, latency, duration, and amplitude/magnitude.
Topography involves the form of the behavior, such as in refining the motor movements of a golf swing.
Frequency involves the number of responses during a period, such as increasing the number of math problems completed per minute on an assignment.
Latency involves changing the time between the stimulus and the behavior, such as decreasing the time between a parent saying “clean your room,” and the child actually cleaning their room.
Duration involves changing the total time for the behavior to take place, such as increasing the length of time a student stays on task.
Amplitude/magnitude involves manipulating the response strength, such as increasing the height of a high jump bar for students in a physical education class.
While shaping can enforce behaviors in a positive way that would otherwise be difficult to learn through instructions, shaping also has several important limitations.
Shaping can be time-consuming, progress toward the desired behavior may not necessarily be linear.
Shaping required consistent monitoring of the learner, shaping can be misapplied (such as a child learning to scream at a parent when his quiet calls for ice-cream are ignored), and shaping can potentially produce harmful behavior (such as in enforcing a sequence of riskier and riskier behaviors) (Cooper, Heron, and Heward, 2020).
Experimental and Applied Behavior Analysis
Behavior analysis consists of three major branches influenced by three major scientists (Cooper, Heron, and Heward, 2020):
- Behaviorism, which is the philosophy of the science of behavior;
- Experimental analysis of behavior, or basic research into behavior;
- Applied behavior analysis, which aims to develop ways to understand and change behavior (Cooper, Heron, and Heward, 2020).
Behaviorism Assumptions and Philosophy
In the early 1900s, studies of states of consciousness, images, and other mental processes dominated psychology (Cooper, Heron, and Heward, 2020).
These studies often used introspection, or the act of carefully observing one’s own thoughts and feelings, as the method of investigation.
Although this does not fully align with modern scientific methods, authors from the first decade of the 20th century nonetheless defined psychology as the science of behavior (Kazdin, 1978; Cooper, Heron, and Heward, 2020).
John B. Watson, scholars generally agree, influenced psychology to be more focused on the objective study of behavior in his article, “psychology as the behaviorist views it” (Watson, 1913).
Watson argued that psychology must focus on observable behavior over states of mind and mental processes; and, furthermore, that the objective study of behavior as a natural science consists of observing the relationships between environmental stimuli and the responses they evoke.
This idea of behaviorism became known as stimulus-response psychology (Cooper, Heron, and Heward, 2020; Watson, 1913). Although Watson’s stimulus-response psychology has faced criticism, it nonetheless influenced psychology to be seen as a natural science on par with the physical and biological sciences.
Experimental Analysis of Behavior
B. F. Skinner, scholars generally agree, established the foundations of behavior analysis formally in his 1938 book, The Behavior of Organisms. In this book – a summary of his research from 1930 to 1937 – Skinner emphasized two types of behavior: respondent and operant.
Respondent behavior, inspired by the work of Ivan Pavlov, posits that respondents are “brought out” by stimuli that occur immediately before the behavior.
For example, a bright light could be the stimulus for pupil constriction, and this phenomenon could be called a reflex. However, Skinner also acknowledged that the behavior of organisms can appear spontaneous or, as Skinner calls it, “voluntary” (Cooper, Heron, and Heward, 2020).
This so-called voluntary behavior, Skinner argued, could not be explained by Watson’s stimulus-response theory of behavior (Skinner, 1938).
To confront this contradiction, Skinner turned to the environment to seek the determinants of behavior that did not appear to have an antecedent cause (Cooper, Heron, and Heward, 2020).
Although Skinner did not deny the role that physiological variables played in determining behavior, he believed that this was the domain of other disciplines.
Through his research, Skinner found that the consequences that follow a stimulus are more important to behavior change than the stimulus itself.
This evidence formed the three-term contingency theory of behavior, which, unlike the stimulus-response model, accounted for how the environment can determine parts of learned behavior (Cooper, Heron, and Heward, 2020).
Skinner called behavior belonging to the three-term contingency model operant behavior. Operant behavior, rather than being elicited by a preceding stimulus (such as the smell of food triggering Pavlov’s dogs to salivate), is influenced by stimulus changes that had followed the behavior in the past.
For example, if a rat had learned that an electric shock would follow an attempt to drink water, the rat may avoid or skirmish when drinking water, regardless of whether or not the water was electrified (Glenn, Ellis, and Greenspoon, 1992).
Skinner named the analysis of operant behavior the experimental analysis of behavior and used quantifiable methods, such as recording the rate at which, say, a rat or a pigeon, demonstrated a particular behavior in a controlled and standardized environment.
In addition to influencing the development of the experimental analysis of behavior, Skinner created several treatises on how the principles of behavior could be applied to areas such as education, religion, government, law, and psychotherapy.
The two most famous of these texts are Walden Two (1948) and Science and Human Behavior (1953). In them, Skinner established approaches to the study of behavior such as mentalism.
Mentalism is an approach to the study of behavior that assumes that a mental or inner dimension differing from the behavior itself exists (Moore, 2003).
These unobserved mental processes – what psychologists such as Moore (1995) call hypothetical constructs – cannot be experimentally manipulated.
Hypothetical constructs can include concepts such as free will, readiness, innate releasers, language acquisition devices, storage and retrieval mechanisms for memory, and information processing (Cooper, Heron, and Heward, 2020).
For example, a rat could push a level every time a light comes on and receive food, but refuse to do so when there is no light.
Behaviorists would reject the idea that the rat has created an association (knowledge) between the light being on and the level dispensing food, calling it an explanatory fiction, saying that it contributes nothing to an understanding of the variables that develop or maintain the behavior (Heron, Tincani, Peterson, and Miller, 2005; Cooper, Heron, and Heward, 2020).
Different branches of behavioralism hold different views on events that cannot be observed objectively. There are structuralist, methodological, and cognitively-based behavioralism.
While structuralism and methodological behaviorism reject events not operationally defined and objectively observed (Skinner, 1974), methodological behaviorists often acknowledge the existence of mental events but do not consider them to be within the realm of the analysis of behavior (Skinner, 1974).
Skinner himself created radical behaviorism, which viewed thoughts and feelings (what he calls “private events” “) as behavior to be analyzed with the same experimental methods used to analyze publicly observable behavior.
Skinner assumed firstly that private events such as thoughts and feelings are behavior; behavior that takes place within an organism is only distinguishable from “public” behavior by its inaccessibility; and private behavior is influenced by the same kinds of variables as publicly accessible behavior (Cooper, Heron, and Heward, 2020).
Applied Behavior Analysis
Applied behavior analysis is a science that seeks to understand and improve human behavior.
In contrast to other branches of psychology, applied behavior analysis focuses on objectively defining socially significant behaviors and intervening to improve the behaviors studied, and uses scientific methods such as objective description, quantification, and controlled experimentation (Cooper, Heron, and Heward, 2020).
Fuller (1949) conducted one of the first studies reporting the applications of operant behavior in humans.
Fuller studied an 18-year-old boy with severe developmental disabilities and squirts a small amount of a warm sugar-milk solution into his mouth every time he moves his right arm, which he would otherwise move infrequently.
By the fourth session, the boy had learned to move his arm three times per minute. A number of other researchers established through their research that the principles of behavior studied in animals are also applicable to humans (e.g. Bijou, 1955; Baer, 1960; Ferster and DeMyer, 1961; and Lindsley, 1956).
Scholars agree that applied behavior analysis as a branch could be traced to Ayllon and Michael’s 1959 paper, “The Psychiatric Nurse as a Behavioral Engineer,” where the authors described how nurses in psychiatric hospitals used a number of behavioral techniques to “improve” the functioning of patients.
This application of behavior analysis stretched to education in the form of teaching practices such as contingent teacher praise and attention, token reinforcement systems, curriculum design, and programmed instruction (Hall, Lund, and Jackson, 1968; Cooper, Heron, and Heward, 2020).
Applied behavior analysis began formally with the publication of the Journal of Applied Behavior Analysis and the publication of the paper, “Some Current Dimensions of Applied Behavior Analysis” (Baer, Wolf, and Risley, 1968).
This paper outlined the criteria for judging research and practice in applied behavior analysis. Baer, Wolf, and Risley (1968) argued that applied behavior analysis should be applied, behavioral, analytic, technological, conceptually systematic, effective, and capable of appropriately generalized outcomes (Cooper, Heron, and Heward, 2020).
Applications of Behavior Analysis
Goltz, Mayer, and Orr (2020) examined how applied behavior analysis can be used in encouraging development agencies to design agroforestry programs that reduce erosion and climate change through case studies.
The people enacting these changes could be individuals or large groups. In successful cases of using behavior analysis to modify agroforestry behavior, the groups attempting to create change gave reinforcements to agroforestry “adopters,” occasionally making the reinforcements either more personal or more immediate in order to increase their salience.
Goltz, Mayer, and Orr established that the antecedent-behavior-consequence contingency used by applied behavior analysts could be used to guide the behavior of change strategists, change agents, and change adopters involved in preserving forests (2020).
Dance in People With Neurodevelopmental Disorders
Pontone, Vause, and Zonnevelt (2020) reviewed 19 studies of teaching dance to those with neurodevelopmental disorders and found that eight listed at least one behavior analytic component as part of their dance intervention package.
For example, five studies used positive reinforcement such as praise to increase desired behaviors, and verbal, model, and gestural prompts were also common.
One study, for example, used auditory feedback to reinforce dance, where the instructors played a clicking sound in response to correct movements, and no sound in response to incorrect movements (Carrion et al., 2019).
Another study used chaining to teach steps of a “dancercise” and transferred this stimulus between different trainers in different settings (O”Connor and Cuvo, 1989).
Ayllon, T., & Michael, J. (1959). The psychiatric nurse as a behavioral engineer 1. Journal of the experimental analysis of behavior, 2(4), 323-334.
Baer, D. M. (1960). Escape and avoidance response of preschool children to two schedules of reinforcement withdrawal. Journal of the Experimental Analysis of Behavior, 3(2), 155-159.
Baer, D. M., Wolf, M. M., & Risley, T. R. (1968). Some current dimensions of applied behavior analysis. Journal of applied behavior analysis, 1(1), 91.
Bijou, S. W. (1955). A systematic approach to an experimental analysis of young children. Child Development, 161-168.
Carrion, T. J., Miltenberger, R. G., & Quinn, M. (2019). Using auditory feedback to improve dance movements of children
with disabilities. Journal of Developmental and Physical Disabilities, 31(2), 151–160. https://doi.org/10.1007/s10882-019-9630-0
Cooper, J. O., Heron, T. E., & Heward, W. L. (2020). Applied behavior analysis. Pearson UK.
Ferster, C. B., & DeMyer, M. K. (1961). The development of performances in autistic children in an automatically controlled environment. Journal of Chronic Diseases, 13(4), 312-345.
Fleece, L., Gross, A., O”Brien, T., Kistner, J., Rothblum, E., & Drabman, R. (1981). ELEVATION OF VOICE VOLUME IN YOUNG DEVELOPMENT ALLY DELAYED CHILDREN VIA AN OPERANT SHAPING PROCEDURE. Journal of Applied Behavior Analysis, 14(3), 351-355.
Fuller, P. R. (1949). Operant conditioning of a vegetative human organism. The American journal of psychology, 62(4), 587-590.
Glenn, S. S., Ellis, J., & Greenspoon, J. (1992). On the revolutionary nature of the operant as a unit of behavioral selection. American Psychologist, 47(11), 1329.
Goltz, S. M., Mayer, A. L., & Orr, B. (2020). Applied behavior analysis as a development tool: Examples from agroforestry. Journal of Sustainable Forestry, 39(8), 785-799.
Hall, R. V., Lund, D., & Jackson, D. (1968). Effects of teacher attention on study behavior 1. Journal of applied behavior analysis, 1(1), 1-12.
Hagopian, L. P., Farrell, D. A., & Amari, A. (1996). Treating total liquid refusal with backward chaining and fading. Journal of Applied Behavior Analysis, 29(4), 573-575.
Heron,T. E.,Tincani, M. J., Peterson, S. M.,& Miller, A. D. (2005). Plato’s allegory of the cave revisited. Disciples of the light appeal to the pied pipers and prisoners in the darkness. In W. L. Heward, T. E. Heron, N.A. Neef, S.M. Peterson, D. M. Sainato, G.
Cartledge, R. Gardner, III, L. D. Peterson, S. B. Hersh,&J.C. Dardig (Eds.), Focus on behavior analysis in education: Achievements, challenges, and opportunities (pp. 267–282), Upper Saddle River, NJ: Merrill/Prentice Hall.
Heward, W. L., & Wood, C. L. (2003). Thursday afternoons with Don: Selections from three teleconference seminars on applied behavior analysis. A small matter of proof: The legacy of Donald M. Baer, 293-310.
Lindsley, O. R. (1956). Operant conditioning methods applied to research in chronic schizophrenia. Psychiatric research reports.
Moore, J. (1995). Radical behaviorism and the subjective-objective distinction. The Behavior Analyst, 18(1), 33-49.
Moore, J. (2003). Behavior analysis, mentalism, and the path to social justice. The Behavior Analyst, 26(2), 181-193.
O”Conner, Y. M., & Cuvo, A. J. (1989). Teaching dancercise to persons who are mentally handicapped: Programming
transfer of stimulus control to a community setting. Behavioral Residential Treatment, 4(4), 289–311.
Pierrel, R., & Sherman, J. G. (1963). Barnabus, the rat with college training. Brown Alumni Monthly, 63, 8-12.
Pontone, M., Vause, T., & Zonneveld, K. L. (2021). Benefits of recreational dance and behavior analysis for individuals with neurodevelopmental disorders: A literature review. Behavioral Interventions, 36(1), 195-210.
Skinner, B. F. (1938). The behavior of organisms. New York: Appleton-Century-Crofts.
Skinner, B. F. (1948). Walden two. New York: Macmillan.
Skinner, B. F. (1953). Science and human behavior. New York: MacMillan.
Skinner, B. F. (1974). About behaviorism. New York: Knopf.
Watson, J. B. (1913). Psychology as the behaviorist views it. Psychological review, 20(2), 158.
Werts, M. G., Caldwell, N. K., & Wolery, M. (1996). Peer modeling of response chains: Observational learning by students with disabilities. Journal of Applied Behavior Analysis, 29(1), 53-66.