School-Based Neurofeedback for Autistic Spectrum Disorder

By Mark Darling

Mark Darling is a psychologist from Queensland’s Sunshine Coast who specialises in working with
children with Autistic Spectrum Disorder and associated developmental and behavioural disorders. He has
worked with Disability Services Queensland and served as Paediatric Resource Project Coordinator for
Queensland Health. Following a stint as a sole private practitioner, Mark now provides neurotherapy
services for Family Challenge, an organisation that provides mental health services in Australia and the
developing world, with a particular focus on the rehabilitation of child soldiers in African states. Mark was
the recipient of the 2004 Showcase Award for Excellence in Innovation (Education Queensland Regional
Winner) and the 2005 Commonwealth Award for Outstanding National Achievement in School
Improvement (Federal Dept of Education and Training) for his research in reducing ASD behaviours in the
class setting. He is currently pursuing doctoral research into the use of neurotherapy in the education
setting.

Neurofeedback is an intervention that is showing a lot of promise for people diagnosed
with Autistic Spectrum Disorder (ASD). While other childhood behaviour disorders such
as Attention Deficit Hyperactivity Disorder (ADHD) have been in the neurofeedback
limelight for some years, it would appear that ASD is about to have its day in the sun.
Recent research is showing that children with ASD are responding very well to both
electroencephalographic (EEG) and haemoencephalographic (HEG) neurofeedback.
Furthermore, our own research indicates that neurofeedback can be an effective schoolbased
intervention for children in the autistic spectrum.

Neurofeedback is an intervention aimed at training individuals to better regulate the
biological functioning of their own brain. This has generally involved the self-regulation
of EEG rhythmic activity, traditionally referred to as EEG biofeedback, neurofeedback or
neurotherapy. In recent years, however, the concept of neurofeedback has expanded to
include self-regulation of other neural substrates. For example, haemoencephalography
(HEG) is a recently discovered biofeedback technique that encourages improved regional
cerebral blood flow. Within a similar timeframe in the mid-1990s, Hershel Toomim
invented near infrared haemoencephalography (nir HEG) while Jeff Carmen came up
with passive infrared haemoencephalography (pir HEG). Toomim’s device utilises a
headband with sensors that measure reflected red and infrared light from brain tissue in
order to calculate the degree of oxygenation. Trainees are basically rewarded for
increasing red colouration. Carmen’s device employs an infrared thermometer to measure
the temperature immediately below the sensor, with trainees being rewarded for
increasing core brain temperature. While nir HEG can be used at multiple cortical
locations, pir HEG is only used at prefrontal location Fpz. Both methods have been
shown to improve metabolism of hypoperfused brain regions.

A review of ASD research

While many clinicians will be familiar with the scientific literature supporting the use of
neurofeedback as a successful intervention for conditions such as epilepsy and ADHD,
research into the effectiveness of neurofeedback for the autistic spectrum of disorders is a
more recent phenomenon. A couple of case studies were published in the mid-1990s, but
it hasn’t been until the new millennium that more rigorous investigations have taken
place. Briefly, case studies or case series have included three single case studies of
children with autism (Cowan & Markham, 1994; Sichel, Fehmi & Goldstein, 1995; Ibric
& Hudspeth, 2003), a series of three children with Autism and Asperger’s Disorder
(Thompson & Thompson, 1995), seven children with Asperger’s Syndrome (Ross &
Caunt, 2003), 60 individuals with ASD ranging in age from 5 to 51 years (Thompson &
Thompson, 2003), and 15 children with Autism and Asperger’s Syndrome (Linden,
2004). All studies reported improved functioning.

In research conducted in Thailand, a series of 181 students with ASD were given 20
sessions of nir HEG in which they were rewarded for increasing cerebral blood flow to
the prefrontal cortex of the brain. The result was a Grade Point Average increase of 0.94
on a 4-point scale, as well as a 53% increase in prefrontal blood oxygenation readings
(Limsila et al, 2004). Recently, Scolnick (2005) also conducted an EEG biofeedback pilot
study with five adolescents with Asperger’s Syndrome that reported mild improvements.
The first controlled group study into the efficacy of neurofeedback for ASD was
published in 2002 when an average of 36 neurofeedback sessions were given to a group
of 20 children diagnosed with autism. There was also a control group of 20 autistic
children who did not receive treatment. Pre- and post-treatment evaluation was in the
form of parent reports using the Autism Treatment Evaluation Checklist (ATEC). The
experimental group showed an average 26% reduction in autistic symptoms in the areas
of communication, social skills, sensory/cognitive awareness, and health and physical
behaviours. The control group recorded a 3% reduction in autistic symptoms.
(Jerusiewicz, 2002).

In more recent research conducted in New York by Robert Coben’s team, 37 children
with ASD were given an average of 20 neurofeedback sessions. A wait list control group
of 12 children matched for diagnosis, gender, age, race, handedness, IQ and previous
treatments was used. All subjects were assessed using quantitative EEG (QEEG)
analysis, infrared imaging, neuropsychological tests and behaviour rating scales. Posttreatment
assessment revealed an improvement in 89% of the experimental subjects, with
an average 40% reduction in symptoms on the ATEC, as well as significant
improvements on neuropsychological tests of attention, language, visual-perception and
executive functioning. QEEG and infrared imaging confirmed that positive changes had
occurred at a neurophysiological level, particularly in terms of reduced cerebral
hyperconnectivity (Coben & Padolsky, 2006).

From the original experimental group of 37 autistic children who completed 20 sessions
of neurofeedback, Coben (2006) then took 32 subjects identified as having frontal system
dysfunction based on QEEG, infrared imaging, neurobehavioural and neuropsychological
testing. The children in the experimental group each received 20 sessions or either nir or
pir HEG. Once again a wait list control group of 12 subjects matched for gender, age,
race, handedness, IQ and previous treatment was used. Following treatment, 90% of
subjects showed improvement, with an average 42% reduction in symptoms on the
ATEC. There were also statistically significant improvements on tests of
neurobehavioural and neuropsychological functioning, brain thermal imaging and QEEG
data. While both forms of HEG provided statistically significant improvements, pir HEG
proved more successful than nir HEG on some key measures of functioning.

Finally, Coben & Hudspeth (2006) identified 14 subjects from the previous group who
had significant levels of mu wave activity. Mu is an alpha-like rhythm that is associated
Mark Darling – Neurofeedback for ASD – June 2007 4
frontal disconnectivity. Half the group received 20 sessions of frontal bipolar
interhemispheric training while the other half received coherence training to increase
connectivity between central and prefrontal brain regions. Both groups improved
significantly on QEEG, neurobehavioural and neuropsychological measures, but only the
coherence trained group demonstrated a reduction in the mu rhythm. Overall, those who
completed the three phases of treatment (EEG 1; HEG; EEG 2) averaged an 80%
reduction in autistic symptoms.

School-based neurofeedback for ASD

My own foray into the use of neurofeedback for children in the autistic spectrum began in
1999 with my own son, but soon expanded into general clinical practice. Neurofeedback
proved effective, but also very time consuming, since autistic individuals seemed to
require longer-term treatment than children with ADHD. Consequently, whenever I
presented to educators I challenged them about the potential for neurofeedback as a
school-based intervention. The first group to take me up on this challenge was Hervey
Bay Special School. Following a further presentation to staff and parents at the school I
then provided three days of training for the staff who were selected to be technicians. My
main aim was to have them know where to place electrodes, how to get a good signal and
how to run a neurofeedback training session.

In 2004 we commenced a pilot project with six autistic students chosen by the school.
Each of the students was ascertained as ASD Level 6, the highest level of support
required within the Queensland education system. I provided all appropriate assessments
on site at the school, and then developed tailor-made neurofeedback protocols for each
student. Protocols were e-mailed to the school for downloading onto the neurofeedback
computer. Following each session, EEG data and session results were e-mailed back to
the clinic for evaluation and protocol adjustment as necessary. Full clinical and technical
support was provided and neurotherapy sessions formed part of each child’s Individual
Education Profile (IEP). I should point out at this juncture that school neurofeedback
technicians were unable to make adjustments to neurofeedback protocols, since we
utilised BrainMaster’s remote training system that makes on-site protocol adjustment
impossible. This provided a good safeguard since it meant that I was the only person who
had the capability of making changes to each student’s protocol.

Students completed an average of 40 neurofeedback sessions in the first semester with
improvements on the ATEC ranging from 2% to 47%. One of the main problems with
this subjective instrument is that parents and teachers often rated the same child in a
completely different light. For example, while one parent rated her son as having
improved by only 2%, the school principal acknowledged that he was the student who
had probably improved the most. It was later revealed that the parent concerned was
personally dealing with a cancer diagnosis at the time she was rating her son’s behaviour,
which surely impacted upon her objectivity. Similarly, two teachers were identified early
in the process as being opposed to having the neurofeedback program in the school. They
seemed somewhat threatened by the new approach and it will come as no surprise that
they each regularly rated the students from their classes as showing no improvement,
while anecdotal reports from other staff members about the same students were very
positive.

Fortunately, the school also established a system of independent direct observation of
behaviour in the classroom for each for the students. Pre-determined autistic behaviours
were observed and quantified at pre-determined windows of time each day. The result
was that across the first 28 neurofeedback sessions, autistic behaviour in the classroom
setting reduced by 64% for the students involved in the study. Improvements were also
reported in the areas of sleep, mood, speech, academic performance, attention, memory,
and social skills; as well as reductions in anger, aggression, seizures, hyperactivity, and
impulsivity. The two students who had epilepsy both stopped having seizures, while two
non-verbal students both began speaking a few words.

The neurofeedback program at Hervey Bay Special School won a Commonwealth Award
for Outstanding National Achievement in School Improvement and was a regional
winner of the Courier-Mail Showcase Award for Excellence in Innovation. Subsequently,
in 2006 a contract with Education Queensland was established in which I provided
training for school staff in a number of regions in Queensland. The aim was to commence
neurofeedback programs in a number of schools and start collecting data for my PhD
research. Unfortunately, concerns about litigation have led to a suspension of the program
by Education Queensland while they do their own literature review into the efficacy and
safety of neurofeedback. We can only hope that Education Queensland sees the benefit of
neurofeedback as a school-based intervention and that program gets back on track sooner
rather than later. In the meantime we continue to provide programs for non-state schools
and individuals who attend the clinic.

References

Coben, R. (2006). Hemoencephalography for autistic spectrum disorder. Presented at
the14th Annual Conference of the International Society for Neuronal Regulation,
Atlanta, Georgia.
Coben, R. & Hudspeth, W. (2006). Mu-like rhythms in Autistic spectrum disorder: EEG
analyses and neurofeedback. Presented at the 14th Annual Conference of the
International Society for Neuronal Regulation, Atlanta, Georgia.
Coben, R. & Padolsky, I (2006). Assessment-guided neurofeedback for autistic spectrum
disorder. Manuscript submitted for publication.
Cowan, J. & Markham, L. (1994, March). EEG biofeedback for the attention problems of
autism: A case study. Presented at the 25th Annual Meeting of the Association for
Applied Psychophysiology and Biofeedback.
Ibric, V.L. & Hudspeth, W. (2003). QEEG and Roshi use in Autism post-toxic
encephalopathy – a case study. Presented at the 11th Annual Winter Brain
Conference, Palm Springs, CA.
Jarusiewicz, B. (2002). Efficacy of neurofeedback for children in the autistic spectrum: A
pilot study. Journal of Neurotherapy, 6(4), 39-49.
Limsila, P., Toomim, H., Kijvithee, J., Bunthong,W., Pookjinda, J., & Utairatanakit, D.
(2004). Hemoencephalography (HEG): An additional treatment for autism.
Retrieved from http://www.biocompresearch.org/Thai_autism_study.htm.
Mark Darling – Neurofeedback for ASD – June 2007 7
Linden, M. (2004). Case studies of QEEG mapping and neurofeedback with autism.
Presented at the 12th Annual Conference of the International Society for Neuronal
Regulation, Fort Lauderdale, Florida.
Ross, J. & Caunt, J. (2003). A comparison of QEEG characteristics in pediatric
asperger’s syndrome and attention deficit disorder. Presented at the 11th Annual
Conference of the International Society of Neuronal Regulation, Houston, TX.
Scolnick, B. (2005). Effects of electroencephalogram biofeedback with asperger’s
syndrome. International Journal of Rehabilitation Research, 28(2), 159-163.
Sichel, A.G., Fehmi, L.G., & Goldstein, D.M. (1995). Positive outcome with
neurofeedback treatment in a case of mild autism. Journal of Neurotherapy,1(1),
60-64.
Thompson, L. & Thompson, M. (1995). Autism/asperger’s/obnoxious child, 3 case
histories: How we get positive results with complex ADD clients. Presented at the
Annual Conference of the Society for Neuronal Regulation.
Thompson, L. & Thompson, M. (2003). Neurofeedback Treatment for Autistic Spectrum
Disorders: Review of 60 Cases-Principles and Outcome. Citation paper presented at
the Proceedings of the Association for Applied Psychophysiology and Biofeedback.