The assertion that applied behavior analysis is the most powerful
approach in the treatment of autistic behaviors warrants examination.
Applied behavior analysis is characterized by discrete presentation of
stimuli with responses followed by immediate feedback, an intense
schedule of reinforcement, data collection, and systematic trials of
instruction. This highly structured format seems to meet the needs of
individuals with autism who typically respond to routine and directness.
However, a critical analysis of the approach is necessary to determine
the real potential of applied behavior analysis in the treatment of
children and youth with autism.
Autism is a spectrum disorder that encompasses many labeled
disorders such as autism, pervasive developmental disorders, and
Asperger syndrome (Jacobson, 2000). Symptoms of this psychological
disorder typically include varying levels of impairment in interpersonal
skills, emotional or affective behavior, and intellectual functioning.
One of the most pervasive characteristics of the disability, however, is
a delay or impairment in the ability to produce and respond to language
(Secan, Egel, & Tilley, 1989). Many children with autism do not
develop speech and other children with the disorder often exhibit
unusual speech patterns such as echolalia or the repetition of what has
been heard. Frequently, the tone of their speech is flat and
unexpressive. Language difficulties are compounded by social
impairments. Children with autism frequently are unresponsive to others,
fail to make eye contact, and miss social cues such as a person's
facial, verbal, postural, and gestural responses (Gena, Krantz,
McClannahan, Poulson, 1996). In addition to these areas of difficulty,
other common disturbances involve stereotypic behaviors,
self-stimulatory behaviors, self-injurious behaviors, repetitious
actions, preoccupation with select objects or topics, aggression,
inflexibility in routines, and over-sensitivity to sensory stimuli.
Autism is a developmental disorder that is usually identified
before 3 years of age. Strange behaviors appear early in the child'
s life and diagnosis has been expedited by more public awareness of the
disability (e.g., popular cinema such as Rain Man). This is certainly a
positive outcome of such attention, but inaccurate portrayal of the
disability can lead to public misunderstanding. For example, only a
small percent of individuals with autism possess the splinter skills of
"Raymond" in Rain Man (i.e., card counting ability).
Research has not been able to pinpoint any single causation of
autism; consequently, interventions for the disability have varied
greatly. Heflin and Simpson (1998) provide a very thorough overview of
interventions for children and youth with autism that incorporate
strategies from psychoanalytic, medical, educational, and behavioral
perspectives. The sheer number of possible interventions for parents
and/or guardians to consider is overwhelming and confusing. Researchers,
themselves, debate about the potential of various treatment options
(Lovaas, 1987). Some experts make optimistic claims to cure the
disability (Lovaas, 1993), while other professionals focus on
remediating a specific behavior or building a particular skill (Stromer,
Mackay, & Remington, 1996). Caregivers and educators are faced with
the problem of choosing the most promising treatments.
Treatments that concentrate on emotional disturbances related to
autism stem from the psychoanalytic approach. Holding Therapy, for
example, attempts to build a bond between the child with autism and the
parent/guardian. The caregiver is advised to hold the infant very
closely and tightly as s/ he speaks in a comforting tone, even when the
child tries to escape the embrace. Another approach that emphasizes
relationship building is the Son-Rise program (Kaufman & Kaufman,
1998). Parents are to repeat the actions of the child in an environment
that is not distracting or stimulating. The majority of the
caregiver/child time must be spent in trying to enter the child's
world and provide unconditional love and acceptance. The difficulty of
these treatments and others like them are significant. First, there is a
guilt-based underlying assumption that the family relationship is
problematic. Second, the intensity of the treatment may not be feasible
for many families with other work and parenting demands. Third, the
treatments are highly invasive with close and on-going proximity to the
Other treatments emphasize a physiological basis for the disability
and rely upon medical interventions. Medications for individuals with
autism range from tranquilizers, anti-depressants, anti-anxiety drugs,
and stimulants to anti-convulsants (Heflin & Simpson, 1998). More
natural, dietary treatments are also suggested. Rimland (1999) strongly
advocates the use of high dosage vitamin B6, magnesium, and
dimethlglycine. Concerns regarding natural and prescribed remedies are
numerous. What are the side effects of these treatments? New drug
therapies for this population lack supportive research and study. What
are the proper dosages for young children? What adverse effects can
result from the combination of biological treatments? Clearly, careful
physician monitoring of such interventions is essential to the health
and welfare of the child.
Another approach centers on educational options and is broader in
scope. Alternatives for consideration in this category of interventions
focus on the placement of the child. A continuum of educational services
is available from a segregated special education program to a fully
inclusive placement in general education classes with support services.
A highly individualized program is required to meet the specific needs
of the child. Of particular concern is the issue of the onset of
treatment. Repeatedly, early intervention programs initiated before the
age of 5 years have been strongly related to progress (Fenske, Zalenski,
Krantz, & Mc Clannahan, 1985; Rosenwasser, & Axelrod, 2002).
A final treatment approach is behavioral in nature. It focuses upon
increasing appropriate behaviors and decreasing inappropriate behaviors.
In order for behavioral treatments to work, there needs to be an
understanding of how children with autism use the stimulation around
them to predict an appropriate response (Schreibman, 2000). This
stimulus-response relationship attempts to build a connection for
children with autism. Applied behavior analysis, which emphasizes this
relationship, has been reported by the Surgeon General of the United
States to be the most effective way to treat autism (Rosenwasser et al.,
2002). A closer examination of this acclamation is warranted. The
remainder of this paper will critically analyze the potential of applied
behavior analysis in the treatment of children and youth with autism.
Applied behavior analysis had its beginnings with laboratory
experiments on animals and trials with humans in the severely retarded
range of functioning (Snell, 1978). After the technique was used
successfully in clinical settings, it extended to additional
exceptionalities of children in classroom settings. The process of
applied behavior analysis is very systematic. Children are first
individually analyzed to assess the behavior that needs to be altered.
Once the behavior is identified, intervention strategies are determined
to suit the situation and, then, used to modify the behavior. During
this time, the instructor provides reinforcement to elicit and maintain
the desired behavior. Evaluations are made throughout the modification
process to assess the effectiveness of the intervention (Simpson, 1998).
When an intervention is found to be ineffective, another strategy is
Each case of applied behavior analysis (ABA) must be conducted
around the context of the environment and particular characteristics of
the individual. The behavior that is targeted for change must also be
observable and measurable. Five more specific steps are followed in the
ABA process (Snell, 1978). First, the positive behavior is measured
directly. Second, the behavior is measured daily based on the target
responses. Then, systematic procedures are followed so that, if
successful in modifying the behavior, those procedures can be
replicated. Fourth, data is recorded on the individual level, usually by
graphing progress. Finally, the interventionist demonstrates that the
results were completed in a controlled manner in an attempt to prove
that the intervention accounted for the change in behavior.
Applied behavior analysis has been implemented in various areas of
learning, including language acquisition, self-help skills, vocational
skills, and daily living skills (Grindle, & Remington, 2002; Snell,
1978). Although applied behavior analysis can take many forms, the
common core procedure described above links all the attempts taken to
modify behavior. For example, techniques such as discrete trial
training, direct instruction, and response prompt systems (e.g.,
increasing assistance, decreasing assistance, time delay) provide
repeated practice and rigid presentation. This format of presentation
seems to fit the characteristics of the population of autistic
individuals. The need for routine, structure, and concrete examples
meshes with the applied behavior analysis approach.
Applied behavioral analysis has the best documented outcome data
supporting this approach as compared with other methods (Rosenwasser et
al., 2002; Jacobson, 2000). The first positive results of ABA with the
autistic population were demonstrated in the 1960s, when programs were
established in classroom sites (Schreibman, 2000). The strategy helped
to increase desired behavior and diminish undesirable behavior.
Many studies have revealed the successful application of ABA, and
many advocacy groups support its use. According to Jacobson (2000), the
only data that shows consistent improvements with autistic children is
applied behavior analysis. In a nation study called Project
Follow-Through, the findings supported the idea that direct instruction,
behavior analysis methods, and additional behavioral approaches were the
strongest ways of instruction for these children. There are also various
associations, such as Families for Early Autism Treatment (FEAT),
Parents for the Early Intervention of Autism in Children (PEACH), and
New Jersey Center for Outreach and Services for the Autism Community
(COSAC), which support behavior-analytic treatments for autistic
In another study at Princeton Child Development Institute, children
between the ages of 11 and 18 were included in an experiment to modify
stereotypic and disruptive behaviors (Gena et al., 1996). The children
and therapist sat face to face during the sessions. They were confronted
with scenarios, given 5 seconds for a response, and then presented with
a consequence. Each session was videotaped and, during the session,
twenty-four scenarios were presented. At specific points, training
trials were used to model appropriate responses. The individuals were
then verbally prompted to match the model. The therapist distributed
tokens based on the responses given to the scenarios. If twenty-three
tokens were attained, they could be exchanged for desirable objects. The
results were that an error-correction procedure and token economy
produced effective results in all participants. The Princeton Child
Development Institute researchers concluded that gains in their
system's effectiveness were directly tied to the use of applied
behavior analysis (McClannahan & Krantz, 1993).
In an additional study, four students with delayed social
interactions, play skills, and behavior issues were taught responses to
what, why, and how questions (Secan et al., 1989). The study focused on
four types of probes--storybook questions, natural-context questions,
spontaneous questions, and maintenance probes. The students were
instructed each day for ten to fifteen minute sessions. The children
were shown pictures and asked questions corresponding with each picture.
When a correct response was given, praise was used as a reward. When
incorrect responses were given, the teacher would model the correct
response and question the student again. The study found that all
students reached or exceeded the desired criteria. There were increases
in responses, but students failed by 35% on meeting generalization
criteria for storybook and natural-context techniques, when the visible
cue was not present. The taught material was also maintained over time.
Through the increased request for ABA strategies, the public has
shown that recognition for the technique has become widespread. For
instance, in a web listing of schools that claim to use ABA as a primary
intervention, 59 sites were identified (http:/
/members.tripod.con/~Rsaffran/ schools.htm). Findings have reported that
ABA even produces improved results on standardized tests for this
population of students (Rosenwasser et al., 2002). Many researchers
agree that the most effective strategy for autistic children is the use
of an intensive intervention that is applied in high dosages
(Schreibman, 2000). Using the predictable and planned out organization
of applied behavior analysis benefits autistic children's learning
style. Although ABA is very effective with this population overall,
different forms of the technique may be more beneficial than others, and
finding those variables that influence effectiveness will be an on-going
Applied behavior analysis has not been without controversy over the
years. Although there are some advocates who state that ABA is the only
way to successfully teach autistic children, others will defend
different treatments. Rimland (1999), for instance, debates that there
are numerous other methods to the treatment of autism that include
documented evidence that the treatment works. Vitamin therapy,
casein-free diets, sensory integration, and auditory integration are
some of the other techniques that have been tested with autistic
populations. Studies with vitamin therapy have followed scientific
procedures that include double-blindness as well as factual evidence of
normalization in the brain waves (Rimland, 1999). With regard to the
various dietary studies, improvements in behavior resulted in many
individuals (Rimland, 1999). These additional findings show that ABA is
still a very powerful treatment, but it may not be the only means of
obtaining desired results.
Further, Lovaas (1987) argues that empirical results from
behavioral intervention with autistic children have been both positive
and negative. He accepts that the treatment is often primarily effective
in the original learning environment, although he cautions that the
reports of recovery from autism are false. Lovaas' (1987) own
research study targeted declining aggressive behaviors, increasing
correct verbal responses, teaching imitation, teaching appropriate play
at different at levels, functioning with peers, teaching appropriate
expressions and emotions, and learning pre-academic skills. The results
showed that only nine out of the nineteen autistic children succeeded in
a regular first grade classroom, after treatment had been delivered.
Also reported was increased intellectual functioning, with a gain of 30
IQ points (Lovass, 1987). Rimland's (1999) review of this study
questions the empirical support; however, the use of ABA must have had
have some bearing upon the outcome.
There are many limitations to the use of applied behavior analysis
treatments with individuals with autism. First, applied behavior
analysis is very intense and intrusive in its format and delivery.
Stressful reactions by the recipient of the procedure should be
carefully monitored. Sensitive and knowledgeable interventionists are
essential in observing adverse treatment outcomes. Second, setting
results may occur, with individuals with autism responding to stimuli in
one environment, but unable to generalize the learning to other contexts
(Schriebman, 2000). Care needs to be taken in selecting natural
environments for instruction in order to promote skills in real world
situations. Third, the spectrum of difficulties, range of abilities, age
of the child, culture of the family, and characteristics of the
individual combine to suggest that the use of a single treatment would
be poor advise. The many particular variables complicate the treatment
selection process. Obviously, treatments must be tailored to meet
Finally, new treatments require closer examination of current and
future empirical studies. Rimland (1999) pointed out that there are 18
studies related to the vitamin B and magnesium treatment alone. The
viability of other options that better suit the individual and family
cannot be overlooked.
Substantial contributions have been made in using applied behavior
analysis as the basis for designing treatments that are useful in coping
with a wide variety of behaviors related to autism (Laties, & Mace,
1993). The most scientifically effective treatment seems to be based
upon a behavioral model, which is done intensively and early in the
child's life (Schreibman, 2000). Besides these overall findings,
treatment for autism depends upon the individual as well as family
variables. Autism is such a fascinating disability and many questions
about the condition are still unanswered. Therefore, there is no current
treatment that completely addresses the needs of the disability. Perhaps
a complement of eclectic strategies is necessary to meet the complex
challenges and spectrum of characteristics associated with autism. As
Lovaas (1987) indicates, one may have to intervene on all behaviors, in
all environments, and with the help of all significant persons.
Realistically, a plethora of individualized treatments may have to start
early in life and continue intensively for a long period of time.
ABA Schools-Behavioral Intervention for Autism (n.d.). Retrieved
July 25, 2002 from http://members.tripod.com/~Rsaffran/ schools.html
Fenske, E. C., Zalenski, S., Krantz, P. J., & McClannahan, L.
E. (1985). Age at intervention and treatment outcome for autistic
children in a comprehensive intervention program. Analysis and
Intervention in Developmental Disabilities, 5, 49-58.
Gena, A., Krantz, P.J., McClannahan, L.E., & Poulson, C.L.
(1996). Training and generalization of affective behavior displayed by
youth with autism. Journal of Applied Behavior Analysis, 29, 291-304.
Grindle, C.F., & Remington, B. (2002). Discrete-trial training
for autistic children when reward is delayed: A comparison of
conditioned cue value and response marking. Journal of Applied Behavior
Analysis, 35, 187-190.
Heflin, L. J. ,& Simpson, R. L. (1998). Interventions for
children and youth with autism. Focus on Autism and Other Developmental
Disabilitiies, 13(4), 194-211.
Jacobson, J.W. (2000). Early intensive behavioral intervention:
Emergence of a consumer-driven service model. The Behavior Analyst,
Kaufman B., & Kaufman, S. (1998). The option institute.
[on-line] Available: http:// www.option.org/guidedintro/intro4.html
Laties, V.G., & Mace, F.C. (1993). Taking stock: The first 25
years of the Journal of Applied Behavior Analysis. Journal of Applied
Behavior Analysis, 26, 513-525.
Lovaas, O. I. (1993). The development of a treatment-research
project for developmentally disabled and autistic children. Journal of
Applied Behavior Analysis, 26, 617-630.
Lovas, O.I. (1987). Functioning in young autistic children. Journal
of Consulting and Clinical Psychology, 55(1), 3-9.
McClannahan, L.E., & Krantz, P.J. (1993). On systems analysis
in autism intervention programs. Journal of Applied Behavior Analysis,
Rimland, B. (1999). The ABA controversy. Autism Research Review
International, 13(3), 1-4.
Rosenwasser, B. & Axelrod, S. (2002). More contributions of
applied behavior analysis to the education of people with autism.
Behavior Modification, 26(1), 3-8.
Schriebman, l. (2000). Intensive behavioral/ psychoeducational
treatments for autism: Research needs and future directions. Journal of
Autism and Developmental, 30(5), 373-378.
Secan, K.E., Egel, A.L., & Tilley, C.S. (1989). Acquisition,
generalization, and maintenance of question-answering skills in autistic
children. Journal of Applied Behavior Analysis, 22, 181-196.
Simpson, R. L. (1995). Children and youth with autism in an age of
reform: A perspective on current issues. Behavioral Disorders, 21, 7-20.
Snell, M. E. (1974). Systematic Instruction of the Moderately and
Severely Handicapped. Columbus: Charles E. Merrill publishing Co.
Stromer, R., Mackay, H.A., & Remington, B. (1996). Naming, the
formation of stimulus classes, and applied behavior analysis. Journal of
Applied Behavior Analysis, 29, 409-431.
Alexis Ann Schoen, Graduate Student, Counseling Psychology, La
Correspondence concerning this article should be addressed to:
Alexis Ann Schoen, Graduate Student, LaSalle University, Olney Hall,
Room 354, Philadelphia, PA 19141; Email: firstname.lastname@example.org.