Perkins School for the Blind Transition Center

New Treatment Directions for Autism Spectrum Disorders – Neuromodulation

Much of our psychiatric and neurologic knowledge is based on a “lesion model.” When there is an intact nervous system and something happens, such as a stroke or bleed, some tissue is destroyed and deficits may remain. In autism and related developmental disorders, instead of one discreet area being the “problem,” widely distributed brain circuits are malfunctioning; often without obvious telltale signs. This is the result of abnormal early brain development and leaves the “target” of treatment hard to define.

Eric London, MD

Eric London, MD

It has been observed that in individuals with autism, the level of functioning can vary dramatically from day to day (or sometimes hour to hour). For example, many “non-verbal” individuals with autism have blurted out language that their parents were unaware they were capable of. At certain times, individuals with autism appear to learn rapidly while on other days they struggle to learn. Interestingly, when individuals with autism spike a fever, it’s been observed that function may improve. We would not expect to see this with a stroke victim, for example. If a stroke victim lost the use of a leg, we would not see the leg function return for a day and then go away again. The variability seen in autism, in contrast, leads to a hypothesis that brain circuitry is relatively intact but that something prevents their optimal functioning at times (and, perhaps, most of the time). When a person is very tired, that would not be the time for him to attempt a very difficult task and so we wait until we are at a more optimal physiologic state to undertake that task. In a similar way, the concept for autism treatment would be to aim at improving the physiologic environment in which the brain is functioning. This is done naturalistically, by promoting adequate sleep, good food and an environment in which stress is minimized. For many years, behaviorists understood that creating an environment that was not over-stimulating would facilitate better learning. To improve treatments for individuals with autism, researchers and clinicians need to take this concept another step further.

A key sign of autism pathology is the deficit in tolerating and adjusting to change in the environment. The formal diagnosis refers to this as “a rigid adherence to sameness” which explains repetitive behaviors and behaviors seen during environmental changes. Social deficits might likewise be explained by the fact that social skills are based on the need to change the script constantly in response to the person with whom you are interacting. While we do not yet have a full understanding of this problem, one rather attractive hypothesis is that “neuromodulators” responsible for helping the brain adjust to the environment are not functioning adequately.

There is a neurochemical called noradrenalin (also named norepinephrine) which functions primarily as a neuromodulator. That is, it doesn’t really send messages from neuron to neuron but rather acts as a modulator of the functions of neurons and circuits – turning them on or off, changing the volume, etc. One might conceptualize it as an orchestra conductor whose job it is to quiet down the wind instruments so that the violins can play the melody. Later, he must bring in other instruments perhaps at high volumes to create the desired effect.

I am proposing that, in individuals with autism, the noradrenergic system is dysregulated and does not perform its function well. The chronic non-performance of this function leads to more permanent deficits as children develop. The clearest example of this is that the noradrenergic system is needed both to focus attention as well as to change focus. When we are out in the world we may be in a “wide-scanning” mode; for our ancestors that meant looking for predators or prey. Then when we encounter something, like food in a tree, we need to figure out how to access the food which requires attention, focus, and perhaps learning a novel task. Once we’ve eaten the food then we go back to scanning the environment to see what may come next (perhaps to find water or a female). Each one of these transitions requires the noradrenergic system to initiate the switch. If the noradrenergic system does not do its job at that time, there will be no switching of circuits and the person will continue in his current mode. Or conversely if the noradrenergic system fires at the wrong time the switch will take place prematurely in the middle of the task.

Aside from attention, there are many other examples of how the circuitry needs to be switched for adequate functioning. Behaviorists have noted for a long time that children with autism can perform much better when given “prompts.” Often a small touch in the right direction enables the student to perform a complex task that they could not seem to perform without it. In fact, the behavioral literature is voluminous on how to fade prompts as prompting is so effective. It is very likely that the function of prompting is to stimulate the appropriate circuitry because the noradrenergic stimulation has failed to do so at that time. Other functions dependent on the noradrenergic modulation include language acquisition, sleep, arousal levels when awake, memory formation and the ability to discard memories, learning consolidation, and autonomic functioning which goes hand in hand with our emotions and behavior. Put another way, the noradrenergic modulation – or the dysfunction thereof – can explain many of the deficits seen in autism.

The use of noradrenergic medications is not new in psychiatry or even for autism. Clonidine and guanfacine work on this system in short or long acting forms and are commonly used for individuals on the autism spectrum. All of the medications used for ADHD are adrenergic. A medication which is less commonly used, however, but which has shown more promise is propranolol (brand name: Inderal). David Beversdorf and his group at the University of Missouri have used small single doses of propranolol on high functioning individuals with autism spectrum disorders. They have shown improved cognitive functioning in many different testing paradigms. I have used large doses of propranolol for individuals with autism who have severe symptoms of aggression and/or self-abusive behaviors. About 85% of those individuals with whom I have used this medication have gotten “much better” or “very much better” and these were individuals who failed on an average of over six other medications (with some having tried up to 12 medications). I recently received grant funding and will soon begin a double-blind, placebo-controlled trial with propranolol for patients with severe aggression or self-abuse. This type of rigorous study is necessary to demonstrate the findings which I have observed in my clinical practice.

Psychiatric medications, as currently available, are a blunt instrument to address brain pathology. When administering a medication orally we need to deal with the reality that it will not be selective to the circuitry that we are interested in “fixing.” It is not that there is too much or too little noradrenalin; rather that is not well regulated and that does not do its job at the times that it is needed. Better ways of administering treatments need to be ascertained. Techniques such as deep relaxation or meditation do regulate the adrenergic system; however, my experience is that the effect does not last long so that in a few minutes the dysregulation is back. Experienced meditators may be able to overcome this but I would not be optimistic about individuals on the autism spectrum becoming expert at these techniques. Electrophysiologic methods deserve attention, such as Transcranial Magnetic Stimulation, which is being used by Manny Casanova’s research team in South Carolina. Deep brain stimulation is now being used successfully for obsessive compulsive disorder and Parkinson’s disease. This would allow for specifically stimulating important circuits; however, the electrodes need to be implanted surgically which limits the practicality of designing research for this modality. Vagal stimulation has potential and is used for epilepsy. Up until recently, surgical implantation was needed. However, there are now external vagal stimulators which are being tested for various disorders and only require holding the device against the skin.

Forty years ago, the best evidence-based treatments for Autism Spectrum Disorders was applied behavioral analysis and antipsychotic medications. Today’s best treatments are similar with some refinement but no breakthroughs have occurred. I believe that improved treatments targeting more sophisticated goals are likely to prove successful. With these prospective treatments “on the shelf,” what is needed is the careful study of these new modalities to provide the scientific evidence for their application to those who are in need of them.


For almost 50 years, the Institute for Basic Research in Developmental Disabilities (IBR) has been serving New Yorkers with developmental disabilities, initially through research alone, and currently through an integrated program of research, service, and education.  Discoveries made in the Institute’s laboratories help save lives and improve the quality of life of people who have developmental disabilities, from premature infants to the elderly.  For more information, visit

Have a Comment?