Revisiting auditory integration training—new research finding
Home » Treating Autism » Revisiting auditory integration training—new research finding facebook

This editorial appeared in the Autism Research Review International (2016, Vol. 3, No. 3, pp. 3 & 7, a publication of the Autism Research Institute - learn more and subscribe

By Stephen M. Edelson, Ph.D.

Since the 1970s, the Autism Research Institute (ARI) has shown interest in an auditory-based intervention referred to as Berard auditory integration training (AIT).

Over our near 50-year history, parents would sometimes contact ARI to find out how to treat their children’s severe hearing sensitivity. Such painful hearing can be associated with rather severe behavioral issues, such as head and ear banging as well as violent tantrums.

After receiving numerous reports from parents worldwide, Dr. Bernard Rimland started recommending that they contact Dr. Guy Berard, a physician practicing in Annecy, France. Dr. Rimland even traveled to Annecy to meet Dr. Berard and to learn more about this intervention. 

Briefly, AIT involves listening to filtered music for two half-hour sessions a day for 10 consecutive days. The most common improvements reported by parents include a reduction in sound sensitivity, improvement in attention and listening, and a decrease in behavioral issues. 

Initially, AIT was simply referred to as “auditory training.” But because there already was a device called an auditory trainer, which is similar to a hearing aid, Dr. Rimland and I came up with the term “auditory integration training.” The term is still used today. 

Annabel Stehli, a mother who brought her daughter to Annecy for AIT, was so impressed with her daughter’s dramatic improvement that she wrote a popular book titled The Sound of a Miracle. Reader’s Digest published a summary of the book, and Women’s Day and the television show 20/20 also covered stories on AIT. At one time 60 Minutes planned to produce an exposé on AIT; but after meeting Dr. Rimland and then spending a week with me, they decided not to air their story.

Prior to the book’s publication, Dr. Rimland anticipated much media attention would be focused on AIT. He contacted Dr. Berard and convinced him to help us conduct a series of research studies. Dr. Berard flew to San Diego, and spent two weeks discussing research with us and teaching us the method. He also gave us one of his AIT devices.

During the 1990s, Dr. Rimland and I conducted three double-blind controlled studies on AIT. All three documented a reduction in behavioral problems. In one study, we found a dramatic improvement in a specific brain wave, referred to as the P3 (also known as P300), which is associated with auditory processing, and possibly retrieval from memory of auditory information.

Around the late 1990s, AIT became a controversial intervention. Many people began offering AIT who were poorly trained, charged a great deal of money, and made unsubstantiated claims. In addition, facilitated communication (FC) was popular at the same time; and many people who were using FC were also trying AIT. As a result, the two methods were grouped together and labeled as “questionable.”

In addition, five less than optimal studies were published showing no improvement as a result of AIT. Given that these studies are still cited as “proof” that AIT is ineffective, I thought it is important, for the record, to briefly describe their shortcomings

Bettison (1996). This study evaluated AIT using a beta version of a new AIT device. The device processed the sound output differently than the one developed by Dr. Berard, and the device was still going through rather extensive modifications.

Gillberg et al. (1997). This study involved only nine participants, and the authors concluded that no benefits resulted from AIT. A reanalysis of their raw data by Dr. Rimland and me revealed a significant reduction on one of their measures. Later, Gillberg et al. (1998) admitted that “... a moderate reduction in sensory problems may have occurred.”

Mudford et al. (2000). These researchers employed a crossover experimental design which is typically used when the behaviors under investigation return to baseline after the intervention is removed. This design is inappropriate for studying AIT because research conducted by ARI found that improvements from AIT may last at least nine months. Furthermore, Anabel Stehli’s daughter continued to show benefits from her one AIT session 25 years later. 

It is important to mention that the study’s authors could not explain why they found significant decreases in ear covering and hyperactivity in the control phase of their study. Most likely, these improvements could easily be explained by examining the behavior of participants who received AIT during the first phase of the crossover and then the placebo in the second phase. That is, the benefits of AIT almost certainly continued to occur and spilled over into the control phase. The authors declined to reanalyze the data because they stated this would increase the likelihood of statistical error when conducting an additional analysis of the data. However, when conducting research, scientists are expected to examine their data in ways that best describe the phenomenon under investigation and not to ignore plausible explanations of the results even if this may mean recanting their original conclusions.

Yencer (1998).  The follow-up assessment in this study was administered four weeks after the last AIT listening session.  Research has shown that improvement is often first observable somewhere between six and eight weeks following the final AIT session.

Zollweg et al. (1997). In this study, the volume level was played as high as 122 decibels (dB), which is much higher than OSHA’s daily permissible exposure levels and has the potential to cause hearing damage. The recommended dB level for AIT is 80. In addition, 25% of the device’s filters were set incorrectly. And finally, only one-third of the participants had autism, and there is no indication in the literature that AIT may be beneficial for those with other developmental disabilities.

Due to these poorly conducted studies and AIT’s association with FC in the 1990s, interest in AIT within the research and autism communities has diminished over the last 15 years.

(Note: Numerous other studies, mostly supporting the benefits of AIT, have also been published in journals and presented at conferences. However, they lack scientific rigor, such as not including a control group for comparison and relying on the ratings of “non-blinded” evaluators.)

An important new study on the efficacy of AIT was recently published by Estate Sokhadze, Manuel Casanova, Allan Tasman, and Sally Brockett in Applied Psychophysiology and Biofeedback (online, 29 August 2016). Drs. Casanova and Sokhadze run one of the top psychophysiological autism research laboratories in the world, and Dr. Casanova is a well-published and highly regarded neurologist.

These researchers measured participants’ evoked potentials prior to, during, and after receiving AIT. Evoked potentials are brain waves that occur soon after the presentation of a stimulus. In this study, the stimulus was auditory-based.

The results revealed improvements in both early and late processing of auditory information. In addition, the researchers detected a decrease in hyperactivity, irritability, and repetitive behaviors. These results replicate the findings published by ARI in the 1990s regarding the effects of AIT on auditory processing (i.e., the P3) and behavior. Drs. Sokhadze and Casanova plan to conduct a more elaborate follow-up study.

We all know that the field of autism research and treatment has been rife with controversy. Many approaches—from Bruno Bettelheim’s parent-blaming psychoanalysis in the 1960s to FC, which actually led to the jailing of innocent parents—have actively done harm, while other approaches that seemed promising have not held up under careful examination. However, many other approaches that initially were greeted with skepticism, such as applied behavioral analysis and biologically-oriented therapies, have changed the lives of individuals with ASD profoundly for the better.

I urge the research community to re-evaluate past research on AIT and consider studying this intervention in an unbiased, truly scientific manner. With more research, AIT may someday become an accepted, evidence-based intervention.



Bettison, S. (1996). The long-term effects of auditory training on children with autism. Journal of Autism and Developmental Disorders, 26, 361-374.

Edelson, S.M., Arin, D., Bauman, M.B., Lukas, S.E., Rudy, J.H., Sholar, M., and Rimland, B. (1999). Focus on Autism and Other Developmental Disabilities, 14, 73-81.

Gillberg, C., Johansson, M., Steffenberg, S., and Berlin, O. (1997). Autism, 1, 97-100.

Mudford, O.C., Cross, B.A., Cullen, C., Reeves, D., Gould, J., and Douglas, J. (2000). Auditory integration training for children with autism: No behavioral effects detected. American Journal of Mental Retardation, 105, 118-129.

Rimland, B., and Edelson, S.M. (1996). Auditory integration training: A pilot study. Journal of Autism and Developmental Disorders, 25, 61-70.

Rimland, B., and Edelson, S.M. (1994). The effects of auditory integration training in autism. American Journal of Speech-Language Pathology, 5, 16-24.

Sokhadze, E.M., Casanova, M.F., Tasman, A., and Brockett, S. (2016). Electrophysiological and behavioral outcomes of Berard auditory integration training (AIT) in children with autism spectrum disorder. Applied Psychophysiology and Biofeedback, DOI 10.1007/s10484-016-9343-z

Yencer, K.A. (1996). The Effects of auditory integration training for children with central processing disorder (CAPD). American Journal of Audiology, 7, 32-44.

Zollweg, W., Vance, V., and Palm, D. (1997). The efficacy of auditory integration training: A double blind study. American Journal of Audiology, 6, 39-47.