Where does the recommendation for 40 hours per week of behavioral intervention come from?

In 1987, Ivar Lovaas, at UCLA, published the first study to suggest that children with autism could make significant gains with treatment. A treatment group of nineteen children received 40 hours per week of EIBT (including discrete-trial instruction; aided integration into regular preschools and kindergartens; and structured, supervised play dates with neurotypical peers). Nine of these children (47%) developed IQ’s in the average range and were able to attend regular education classrooms independently. When independently assessed at an average age of 11 years (McEachin, Smith and Lovaas, 1993), eight of these children had maintained their gains, and were no longer diagnosed with any autism-spectrum disorders. This study also included a control group of nineteen children who received only 10 hours per week of behavioral intervention. No children in this group developed average IQ’s, or attended regular education placements.

Dramatic claims about new autism treatments are made every year. However, Lovaas’ results received tremendous attention, in large part because they were from a renowned scientist whose previous two decades of findings had been successfully replicated by other researchers.

In 2005, Glen Sallows and Jane Howard published two of the most significant replications of Lovass’ 1987 findings.

Sallows and his colleagues provided children with 31-38 hours per week of EIBT services. When assessed at age seven, 48% of these children demonstrated IQs’ in the average range, and were attending regular education placements.

Howard and her colleagues compared EIBT to special education programs. Twenty-nine children received EIBT, 16 children attended a 30-hour-per week special education program, and 16 children attended a 15-hour-per-week special education program. Children in the EIBT group demonstrated an average IQ increase of 31 points, while children in the two special education program demonstrated no statistically significant improvements in IQ scores. Despite receiving twice as many hours of intervention, the 30 hour-per-week special education group did no better than the 15-hour-per-week special education group on any measure of treatment outcome.

For further information about these studies, and information about the scientific validity of a broad range of popular autism treatments, visit the Association for Science in Autism Treatment’s web site at: www.asatonline.org

References

Howard, J.S., Sparkman, C.R., Cohen, H.G., Green, G., Stanislaw, H. (2005). A comparison of intensive behavior analytic and eclectic treatments for young children with autism. Research in Developmental Disabilities, 26, 359-83

Lovaas, O. I. (1987). Behavioral treatment and normal educational and intellectual functioning in young autistic children. Journal of Consulting and Clinical Psychology, 55, 3-9.

McEachin, J. J., Smith, T., & Lovaas, O. I. (1993). Long-term outcome for children with autism who received early intensive behavioral interventions. American Journal of Mental Retardation, 97, 359-372.

Sallows, G.O. & Graupner, T.D. (2005). Intensive behavioral treatment for children with autism: four-year outcome and predictors. American Journal on Mental Retardation, 110, 417-38.

What kind of data is appropriate to expect classroom aides to take on an ongoing basis?

My son is fully included in a first grade class, and has a full-time aide provided by the school district. The teacher and aide always say things are going fine, and they send home a notebook every day that is supposed to let me know what is going on. When I have asked for some kind of data at our IEP, I was told that it is not possible for the teacher or aide to take any ongoing data on a regular basis. I really want to know what is going on. What should I ask for?

Momentary time sampling is any easy way for classroom aides or teachers to take data on whether or not a student is independently on task. With momentary time sampling, aides can even collect data while they are working directly with other students.

For example, an aide might collect data on a particular student for one hour per day, at five minute intervals. Ideally, the data collector carries a vibrating timer (http://www.bindependent.com/cgi-bin/shopper.cgi?preadd=action&key=gct100). When the timer vibrates, the collector records whether or not the child is on task, at that moment. The aide then records the behavior on either a counter (http://www.tallycounterstore.com/ ) or a data sheet. At the end of the hour, you have 12 data points, that took less than 30 seconds to collect.

This data is not terribly detailed, but it is significant, and it allows you to track growth or deterioration. Knowing that a child is independently on task 90% to 100% of the time tells you a lot. Alternatively, if you see your child’s percentages decreasing, you know additional help may be required.

Data should be collected during the times of day when your child faces his/her greatest academic or social challenges. In addition to tracking on-task behavior, you may take data on appropriate play or social engagement during recess. Which observation interval you choose (e.g., once every five minutes vs. once every two minutes) will vary based on activities. For example, a two-minute interval between observations would be more appropriate than a five-minute interval for tracking engagement in appropriate play during a 10-minute recess.

Do teachers’ unions really prohibit non-public agencies from serving as children’s aides in public schools?

Some special education administrators will initially mention this as being a reason why they cannot authorize non-public agency support for a child. It is true that some districts' aides are in labor unions. However, in the past 25 years, working with at least 40 different school districts, the union issue has never affected our ability to serve children in public schools. The issue generally disappears if you ask for specific details about the exact constraints that districts are working under.

Should We Have 30 Programs on Maintenance?

My daughter’s program currently has 11 programs on acquisition, and over 30 programs on maintenance. We end up spending more time reviewing old programs to make sure my daughter doesn’t lose them than teaching new ones. Is this normal?

No. You should probably have no more than 3-5 programs on maintenance at any one time. You should definitely not have to regularly practice all old programs. Your problem is most likely due to poor sequencing in the programming of your daughter’s case.

When programming is done correctly, each new skill is only introduced if it will be used relatively soon in a higher level program. Skills that are not subsequently used in more advanced programs will typically be forgotten (though there is a critical exception to this that we will get to in a minute).

Studying math offers a good analogy to this situation. Each skill taught is used in the next, more complex skill. Counting skills are used in addition and subtraction. Addition and subtraction is used in multiplication. Multiplication is used in division. Multiplication and division are used in algebra. Algebra is used in geometry. Geometry is used in trigonometry, and so on. Children taking algebra don’t have to continually review multiplication facts, because multiplication skills are practiced in the course of doing algebra.

EIBT programming should be conducted the same way. Verbal imitation skills are used to expressively label objects and actions. Object and action labels are used to speak in complete sentences. Complete sentences are used to describe previous experiences. Descriptions of previous experiences are used to carry on conversations (we left a lot of additional skills out of this chain).

Introducing pronouns such as “I,” “you,” “he,” and “she,” too early is a common mistake. Because they are easy to say, they are often introduced before children are able to speak in complete sentences. But guess what? They don’t really have any communicative value unless they are used in sentences. Consequently, children lose them pretty quickly.

Effective EIBT requires the rapid acquisition of hundreds of skills over the course of 2-4 years. This rapid acquisition is sabotaged if programs are not sequenced appropriately.

What’s that exception we mentioned earlier? It’s the real world. It’s fine to introduce a new skill that will not be used in subsequent programming, if the skill will be used in a real-world setting. For example, we may teach a child to receptively label her own printed name, because in preschool, she has to sit on the carpet square that has her name on it.

Do Children “Burn Out” on ABA/EIBT After 18 Months?

I have seen a lot of postings on websites where parents talk about their kids making good progress in ABA programs for a while, but then they start tantruming and their rate of learning falls off. Is this common?

Children may plateau at any developmental level. However, if a child’s rate of progress slows down due to tantrums, noncompliance, or lack of motivation, the cause is most likely poor programming or poor therapy.

Unfortunately, there are a lot of minimally trained EIBT providers out there. They know how to teach discretely, and teach some basic communication skills. However, they do not know how to integrate children into preschools, develop peer socialization skills, or introduce high-level language programs (all critical components of a competent EIBT program). So, they run out of things to teach. Then, they make kids mad. They pointlessly expand previous programs (for example, teaching a four-year old to label every item in every kitchen drawer), or, they try to teach high-level language skills randomly, completely out of any developmental order (for example, expecting a child to engage in a conversation about what they did over the weekend, when the child doesn’t have the verbal skills to tell you what he/she did two minutes ago.).

Many EIBT providers also lack the experience necessary to diversify and naturalize reinforcement procedures as children mature. Tickles, rough-housing and food can be very effective for early stages of an EIBT program. However, their effectiveness decreases substantially for those children who make rapid developmental progress. Two critical changes in reinforcement procedures must occur as children advance through the EIBT curriculum:

1) The types of reinforcers used must be expanded and varied. To maintain a child’s willing cooperation as they progress through hundreds of programs and thousands of hours of instruction, therapists need to use anything and everything that a child finds interesting or motivating. For example, playing outside, trips to the park, movies, videos, video games, toys, trips to restaurants, trips to stores, swimming, etc. Now, you can’t use these reinforcers on a trial-by-trial basis. This leads us to the second critical reinforcement change that must occur over the course of an EIBT program:

2) Reinforcement schedules must be naturalized. Children have to learn to work longer in order to get bigger reinforcers. In early stages, this may mean working for three minutes in order to get to play a video game for three minutes. By kindergarten, this means working for several hours or days before getting preferred activities or items (and sometimes the preferred activity is just going home). Children have to be on roughly the same reinforcement schedules as their classmates, if they are to potentially function independently in an academic setting. The gradual transition to these types of reinforcement schedules has to be an integral part of an EIBT program, and it has to start years before kindergarten does.

Children vary in how much they learn and how much they will benefit from EIBT programs. However, competent behavioral teams can adjust the demands placed on any child so that they can be compliant and cooperative, regardless of their rate of learning.

What does "ABA trained" mean?

My school district tells us that all their classroom aides are "ABA trained." What level of experience should I expect from a person who is "ABA trained?"

Unfortunately, the term "ABA trained," is pretty much meaningless. It can be used to describe a person with a doctorate in applied behavior analysis, or a person who attended a two-day training on discrete-trial teaching, 10 years ago. Typically, you will see personnel closer to the latter extreme.

The best strategy is to define the level of experience that would be appropriate for your child's educational needs (with the help of an ABA professional), and request a person with those specific credentials. For example, you may request, "a person who has received a minimum of six months of training and supervision from professional-level ABA personnel in the development and implementation of token economies in academic settings." Note that in this example, we specified that the person was trained by an ABA professional, for a specific amount of time, in a specific procedure.

In addition to these details, it is also critical to define what level of supervision your aide will receive (how many hours per week, by whom). Even the best aides will require ongoing professional supervision.