The emergence of applied behavior analysis can be traced to the late 1950s and early 1960s,
during which a series of studies was conducted using the principles of operant conditioning
to improve the lives of persons with developmental and mental disabilities. The efficacy of
the science was established for both neuro-typical and mentally disabled persons. The use of ABA
with persons with ASD was made famous by Ivar Lovaas, Ph.D. at University of California Los Angeles;
however, since his initial research, the application of ABA with persons with ASD has changed
tremendously. Although the early inception of the use of ABA with persons with ASD included sterile
clinic rooms, robotic repetition of “mass trials” and highly artificial delivery of rewards and punishments
for appropriate and inappropriate behaviors, the application today is vastly different. Based on precisely
the same scientific principles, today’s use of ABA with persons with ASD can be highly naturalistic and
intervention can take place in the home and community environments.
Applied behavior analysis (ABA) is a science centered on understanding and improving human behavior.
Specifically, ABA is the study of changing socially significant behavior to improve a person’s experience
of his/her environment. Baer, Wolf, and Risely (1968) authored a paper entitled “Some Current Dimensions of
Applied Behavior Analysis,” which highlighted seven characteristics of applied behavior analysis. This seminal
paper is referenced by most in the field as the inception of ABA. The seven areas outlined include the following:
applied, behavioral, analytic, technological, conceptually systematic, effective, and generality.
- Applied denotes that this branch of behavior analysis is focused on human behavior
that impacts every-day life. The behaviors selected for change are socially significant and improve the lives
of the individual and/or significant others around them (e.g., parent, teacher, etc.). For example, a behavior
analyst is much more interested in developing the functional skill of requesting, rather than increasing
an IQ score.
- Behavioral indicates that the focus is the behavior that needs to be changed,
and it must be both observable and measurable. The science does not attempt to measure or change intentions,
but rather is focused on behaviors that can be reliably observed and measured by other people in the environment.
- Analytic indicates that a relationship can be demonstrated between the intervention and the change
in the behavior of the person under study. In other words, the intervention is the cause of the behavior changing.
- Technological implies that the procedure of behavior change is defined clearly enough that another
behavior analyst, who was not a part of the initial intervention, reading the procedure, could replicate
the same intervention.
- Conceptually systematic is the term used to describe the need for reports of intervention to be
related to the principles of ABA from which they were derived. To clarify, although there are innumerable
ways to change and improve behavior, most interventions result from comparatively few basic principles.
- Effective simply implies that the intervention improves the behavior under study to a socially
meaningful degree. To measure effective change it is critical that the participant and/or those
in the environment of the participant have experienced a significant difference in the occurrence of the
behavior. For example, although reducing the frequency of property destruction, from 5 occurrences per
hour to 2 occurrences per hour (a 60% reduction) may be statistically significant. This reduction may
still not yield an “effective” intervention if on a daily basis a person is breaking windows, albeit less windows.
- Generality is the idea that the outcomes, or behavior change, lasts across time and occurs across
multiple environments in which it was not treated. For example, if in the home environment a child is taught
to sit at the table with his family to eat dinner, generalization of this skill would mean that the family
could also go out to dinner at a restaurant and the child would remain seated. Additionally, behavior change
that lasts across time, or maintains, means that months after the intervention to address that behavior has
ceased, the new behavior is still in place.
Effective intervention using the science of applied behavior analysis is focused on manipulating the variables
surrounding a specific behavior. Specifically, the science is interested in what happens immediately before
(the antecedent) and what happens immediately following (the consequence) the behavior under question.
By examining and understanding the antecedent, the behavior analyst can assess what cues or signals the person
to engage in the specific behavior. The consequence, the event that immediately follows the behavior, is
the variable that either increases/maintains or decreases the behavior that just occurred.
A hallmark characteristic of applied behavior analysis is the individualization of intervention. Thus,
although the general principles of behavior change are the same, how these principles are applied with the student
in her environment will have great variability. This reality is exemplified in how behavior analysts assess
behavior. Unlike colloquial parenting books, which may indicate that a parent always “ignore” tantrum behavior
and always use “time out” for aggression, ABA requires an assessment of the specific environment and the people
in the environment before recommending how to respond to behaviors. This assessment process, referred to as
a functional behavior assessment (FBA), requires interviews with those in the environment and direct observation
of the behavior under study and the variables (antecedents and consequences) surrounding that behavior.
When this information does not produce conclusive results, further investigation through the use of an experimental
functional analysis (FA) maybe warranted. A FA manipulates the variables in the environment to assess which
variables increase the likelihood that an individual will engage in the target behavior. For example, in a typical FA,
situations will be set up to assess whether escaping from demands, gaining attention, or accessing a preferred item
is the reason why an individual engages in a behavior. Additionally, a FA will assess if the behavior is not
maintained by social consequences, but is rather automatically maintained.
Interventions are developed based on the results of a FBA and provide consequences that increase appropriate behaviors,
called reinforcement (e.g., giving smiles, high-fives, toys, a break, etc) and decrease inappropriate behaviors,
called extinction and punishment (e.g., ignoring, blocking, and redirection). The consequences provided through
intervention are based on the function of a behavior, in other words, the purpose the behavior serves
(i.e., attention, escape, access to tangibles, or automatic reinforcement).
In the video, the treatment for the Isaiah’s tantrum behavior is to ignore the behavior because results of the FBA
indicated that Isaiah’s tantrum behavior is maintained by attention. Prior to intervention, Isaiah’s mother
would hold and comfort him whenever he started to tantrum. Unfortunately at the start of intervention,
Isaiah was not learning the appropriate ways to get his needs met (e.g., through language), but rather his mother’s
attention during tantrum behaviors was communicating to him that his tantrums were the most effective way to gain
access to her attention.
An integral application of the antecedent-behavior-consequence sequence is discrete trial training. Discrete trials
are units of teaching that have a clear beginning and end. Typically, the start of the trial is initiated by the
teacher’s use of a clear, simple instruction (e.g., “Do this”). The teacher then waits a predefined time
(e.g., 5 seconds) for the student to respond to the instruction. At the end of the time, if the student has
responded correctly (e.g., imitated the teacher’s actions) then the teacher immediately gives the student
access to a reinforcer (e.g., giving free access to a highly desired toy). If, however, the student responds
incorrectly or does not respond at all, the teacher will follow with feedback (e.g., “Oops, let’s try again”)
and then immediately redeliver the instruction with a prompt (e.g., pointing, modeling the correct response).
Applications of this teaching procedure are discussed in the next paragraph; however it should be noted that
these trials can be highly structured in format or can be very loosely structured given the needs and skills of the student.
These applications include Pivotal Response Treatments(Koegel & Koegel, 2006), Incidental Teaching
(Hart & Risley, 1975) , Milieu Language Paradigm (Hart & Rogers-Warren, 1978),
and Natural Environment Teaching (Sundberg & Partington, 1988), to name just a few applications. Although each
of these has subtle differences, they are firmly grounded in the principles of ABA. These applications have worked
to address the issues of generalization, as well as to create more shared control between the child and the teacher
to increase the level of engagement in the learning activity. For example, in the video, Isaiah is playing outside
with his mother and interventionist engaging in an activity that he has selected and is highly preferred
(i.e., writing with chalk on the concrete). His mother has learned through intervention to capitalize on Isaiah’s
excitement for this highly preferred activity and she starts writing numbers (one of Isaiah’s target goals is
to identify numbers). As the video shows, Isaiah becomes excited imitating his mother’s language models of
the numbers. Isaiah begins to independently say the numbers (a skill that he had not yet been demonstrating).
The actual activity of saying numbers is becoming more and more reinforcing, in and of itself, because of
the manner in which it is being targeted by his mother. This is a great example of the flexibility and malleability
of the principles of ABA.
Evidence-Based research on ABA:
The Kids Who Beat Autism1This feature-length article in the New York Times Magazine (July 31, 2014) considered whether some children can
can recover from autism. The article reviews the history of applied behavioral analysis (ABA), an intensive,
highly structured one-on-one treatment for children on the autism spectrum that was developed at UCLA in the 1980s by Ivar Lovaas. Based on the results of this technique, which deconstructs every human social behavior into sequences of tiny steps and seeks to teach children the steps using behavioral therapy tools, Lovaas claimed that nearly half of the children who received this therapy in its full form recovered. The claim created considerable controversy, and examination of the treatment and Lovaas’ data identified a number of problems. More recent case studies, and two systematic reviews of those studies, have shown that although a certain percentage of individuals with the diagnosis do recover (have successful, functional lives), the rate of recovery may not be higher among individuals who underwent ABA or other intensive behavioral-based therapies than among those who received much less intensive types of therapy. The article follows 5 young people and their families from birth through their diagnoses and treatment; 4 of the 5 have recovered. The fifth, a young man, never learned to speak and now resides in a group home, in spite of intensive one-on-one therapy, yet another is a popular and accomplished honor student in his senior year of high school, in spite of no intensive therapy. Nothing in the studies seems to suggest what might predict whether any one child will benefit more from intensive therapy or even outgrow the diagnosis on his or her own.
Evaluating the efficacy of the PEAK Relational Training System using a randomized controlled trial of children with autism.2The PEAK Relational Training System is a series of assessments and curriculum guides consisting of four modules, based on B.F. Skinner’s theory that language learning is best described by the relationship between stimulating events (A), the response of the speaker (B), and subsequent actions on the part of the listener (C). The PEAK Direct Training Module is designed to assess and teach language skills according to this “ABC” design, such that individuals are reinforced for providing the appropriate response to a verbal statement (stimulus).
Twenty-seven children diagnosed with pervasive developmental disorders (PDD) were randomized to either language instruction based on the PEAK Direct Training Module or treatment as usual (TAU, based on the child’s existing individualized educational plan [IEP]). All participants were evaluated using the PEAK direct training assessment prior to randomization. This assessment specifically evaluates an individual’s ability to learn language skills through direct contingencies (i.e., through reinforcement of specific verbal responses).
After one month, both groups were re-assessed using the PEAK direct assessment after one month. Children in the experimental group made significantly more gains in language skills than those who were assigned to the TAU group.A Randomized Clinical Trial Comparison Between Pivotal Response Treatment (PRT) and Adult-Driven Applied Behavior Analysis (ABA) Intervention on Disruptive Behaviors in Public School Children with Autism.3A recent study conducted in Iran randomized public school children with ASD to either Pivotal Response Treatment (PRT) or an adult-directed Applied Behavior Analysis (ABA) approach to language intervention. PRT is based on ABA but incorporates child choice, direct and natural rewards, reinforcing attempts, task variation, and interspersal of previously-learned and newly targeted language structures. The group that received PRT showed greater improvements in language acquisition, the primary outcome of interest, than did the group that received adult-directed ABA. The authors further hypothesized that because the PRT group showed greater improvements in language acquisition during the intervention, they would engage in less disruptive behaviors than the ABA group.
During the first session, the time spent engaging in disruptive behaviors averaged 9.6 minutes for the children in the ABA group, compared to an average of 11.5 minutes for children in the PRT group. However, after three months of intervention, the children in the adult-directed ABA group engaged in disruptive behaviors for an average of 8.4 min per session, while the PRT group showed a mean of only 1.6 minutes of disruptive behavior per session.
- ABA predictors of outcome4
- Large-scale, publicly funded ABA outcomes5
- Early and intensive ABA leads to better outcomes6
- ABA in School Setting7
- Overview of ABA research8
- ABA Changing Core Symptoms Study9
- Comparing ABA to Eclectic Treatments10
General Guides:
1 Padawer, Ruth. The Kids Who Beat Autism. New York Times Magazine. July 31, 2014.
2 McKeel AN, Dixon MR, Daar JH, Rowsey KE, Szekely S. Evaluating the efficacy of the PEAK Relational Training System using a randomized controlled trial of children with autism. Journal of Behavioral Education. 2015;24(2):230-41.
3 Mohammadzaheri F, Koegel LK, Rezaei M, Bakhshi E. A Randomized Clinical Trial Comparison Between Pivotal Response Treatment (PRT) and Adult-Driven Applied Behavior Analysis (ABA) Intervention on Disruptive Behaviors in Public School Children with Autism. Journal of Autism and Developmental Disorders. 2015.
4 Reprinted from Research in Autism Spectrum Disorders 5(1), 592-603. (2011), with permission from Elsevier
5 Reprinted from Research in Autism Spectrum Disorders, 6(2), 673-682. (2012), with permission from Elsevier
6 Reprinted from Research in Autism Spectrum Disorders 6(2), 829-835. (2012), with permission from Elsevier
7 Reprinted from Research in Autism Spectrum Disorders 6, 829–835. (2012), with permission from Elsevier
8 Reprinted from Clinical Psychology Review, 30 (4), 387-399. (2010), with permission from Elsevier
9 Reprinted from Research in Autism Spectrum Disorders 1, 304–317. (2007), with permission from Elsevier
10 Reprinted from Research in Developmental Disabilities 26, 359–383. (2005) with permission from Elsevier