Tuesday, 22 August 2017

Music for Autism? – an acquired taste, apparently

Today’s post is about music and music therapy.

A new study reports that music therapy does not improve autism symptoms.

In an earlier post we saw that singing reduces the level of your stress hormone cortisol; this was based on testing adults in a choir, so not music novices.

Music has actually been shown to do much more than just reduce your level of stress, it can actually affect the expression of your genes, but only in those who are “musically experienced”; in people with little experience of music it does nothing. 

Although brain imaging studies have demonstrated that listening to music alters human brain structure and function, the molecular mechanisms mediating those effects remain unknown. With the advent of genomics and bioinformatics approaches, these effects of music can now be studied in a more detailed fashion. To verify whether listening to classical music has any effect on human transcriptome, we performed genome-wide transcriptional profiling from the peripheral blood of participants after listening to classical music (n = 48), and after a control study without music exposure (n = 15). As musical experience is known to influence the responses to music, we compared the transcriptional responses of musically experienced and inexperienced participants separately with those of the controls. Comparisons were made based on two subphenotypes of musical experience: musical aptitude and music education. In musically experienced participants, we observed the differential expression of 45 genes (27 up- and 18 down-regulated) and 97 genes (75 up- and 22 down-regulated) respectively based on subphenotype comparisons (rank product non-parametric statistics, pfp 0.05, >1.2-fold change over time across conditions). Gene ontological overrepresentation analysis (hypergeometric test, FDR < 0.05) revealed that the up-regulated genes are primarily known to be involved in the secretion and transport of dopamine, neuron projection, protein sumoylation, long-term potentiation and dephosphorylation. Down-regulated genes are known to be involved in ATP synthase-coupled proton transport, cytolysis, and positive regulation of caspase, peptidase and endopeptidase activities. One of the most up-regulated genes, alpha-synuclein (SNCA), is located in the best linkage region of musical aptitude on chromosome 4q22.1 and is regulated by GATA2, which is known to be associated with musical aptitude. Several genes reported to regulate song perception and production in songbirds displayed altered activities, suggesting a possible evolutionary conservation of sound perception between species. We observed no significant findings in musically inexperienced participants.


Apparently there are about 7,000 music therapists in the United States and about 6,000 in Europe.  One of the target groups for these therapists is children with autism.
So should parents pay out their cash for music therapy classes?  Well a very recent large study carried out in nine countries by a team from Norway suggests you might not want to open your wallet.
I must say that I hold a different view and that this simplistic kind of research is rather unhelpful. 
From the research in this blog we know that people who develop a love of music express a measurable biological effect, which does indeed look beneficial.
How do you develop a love of music, or indeed dance? Well you have to be exposed to it and engage in it.
Music therapy is all about engaging in music.
Monty, now aged 14 with ASD, has been dancing almost since he was walking, in great part because his then assistant loved music.  Later on you can start to make your own simple music, later you can sing and eventually play an instrument.  This process takes years.  
Music therapy is just a start, years later you can be trampolining to Abba, lying in bed listing to classical music, or just playing the piano.  But it is a long road.
In the recent research they gave 5 months of music therapy to 364 children aged 4 to 7 and then tested their social skills using the Autism Diagnostic Observation Schedule (ADOS).  Their social skill score did not improve. I am not sure why they picked this variable to measure.
This is yet more flawed research, which will then be quoted as fact by others.
You could make a study on teaching judo to kids with autism. I think you would find after 5 months it did not improve their social skills, but those who continue for 5 years might benefit considerably, versus those sat on the sofa watching videos on their iPads.
Clearly not everyone likes music, or indeed judo. Many kids with more severe autism have little interest in anything and so they need a lot more encouragement than typical kids.
The only way to find out if children can develop an interest in music, sport or anything else is to expose them to it at a young age. This is all music therapy is supposed to be, it is not meant to be a cure for anything. 

Researchers found that children with ASD in nine countries scored similarly on a test of their social skills whether or not they had received the music therapy.

"Music therapy - like many other interventions that have been suggested - does not improve autism symptoms," said senior author Christian Gold, of the Grieg Academy Music Therapy Research Center and Uni Research Health in Bergen, Norway.

ASDs are developmental disorders that can lead to social, communication and behavioral challenges. The U.S. Centers for Disease Control and Prevention estimates that one in 68 children in the U.S. has been diagnosed with an ASD.

The anecdotal link between music and ASD goes back many years, Gold and colleagues write in JAMA. During music therapy, a person helps a child spontaneously make music through singing, playing and movement.

There are about 7,000 music therapists in the United States and about 6,000 in Europe, the researchers write.

For the new study, the researchers recruited 364 children ages 4 to 7 years from 10 treatment centers between 2011 and 2015. The centers were in Australia, Austria, Brazil, Israel, Italy, Korea, Norway, the UK and the U.S.

All of the children received the usual care a child with ASD would receive in their region, but half of the children were randomly assigned to also get music therapy.

Usual care could range from early intensive behavioral interventions, to speech and language therapy, to sensory-motor therapies and medications, Gold told Reuters Health by email.

"Music therapy is also among the interventions that have been recommended when it is available," he said. "Some parents who are frustrated with behavioral interventions may experience it as bringing back the joy of being with their child in a natural way."

After five months of therapy, the researchers did not find a difference between the two groups of children on a measure of social skills.

Gold said parents should continue to pursue music therapy if they feel it's a good match for their children, but don't expect it to be a so-called treatment.

The article below is quite a good one:

The study itself:

In this issue of JAMA, Bieleninik and colleagues1 present the results of a large, well-designed, multicenter randomized clinical trial (RCT) of improvisational music therapy for young children with autism spectrum disorder (ASD). Music therapy is “a systematic process of intervention wherein the therapist helps the client to promote health, using musical experiences and the relationships that develop through them.”2 Among 364 children aged 4 to 7 years, over 5 months, the mean scores on the Autism Diagnostic Observation Schedule (ADOS), social affect domain, decreased from 14.08 to 13.23 among children randomized to improvisational music therapy and from 13.49 to 12.58 among those randomized to enhanced standard care, a mean difference in change scores of 0.06 (95% CI, −0.70 to 0.81), with no significant differences between groups.

How Much Music?
I think you need music lessons twice a week to have a meaningful impact and, as with all therapies, you need more practice at home.  Most kindergartens have music and dance as part of their activities. Taken together it is not so hard to get quite a lot of exposure to music at a young age. Then, if the child really likes music, you just keep going.  


Is music therapy a quick fix for autism? Definitely not.
Is music therapy a fun way to engage many young children with autism? The recent research does not say so, but it is clear that many people, with all levels of autism severity, can enjoy music and participate in it.
I think we should put music alongside sport, as a useful activity that young children should be encouraged to engage in.   It can be a struggle to get some people with autism to engage in anything, which is where a music therapist comes in.
Is it worth the investment in time and money? That all depends on the child and the therapist. Buying a piano, 7 years ago, was certainly one of my better investments; but you do also need a lot of lessons.  The end result is someone with a genuine love of many kinds of music and I expect he is now in the cortisol lowering, gene expression modifying category of the musically experienced.
Five months of unspecified music therapy may not be enough to see results and quite possible those results are not increased sociability anyway.


  1. Hello Peter, is it possible to email you? I had a few thoughts and questions about a possible future post topic. Thanks~

    1. MKate, just send it as a comment, not to be published.

  2. Peter I am glad you bring music up again. I still have your older post in mind which I found clearly applicable to my son.
    I am quite sure music can lower cortisol for some ASD and my son falls into this category.
    Is it aquired or innate, I don't know, maybe both.

  3. Hi everyone,

    Here are a couple of very interesting new papers:

    An RNA interference screen identifies druggable regulators of MeCP2 stability

    So in this paper, which we unfortunately only have access to the abstract of, the researchers appear to have found a few targets that stabilize MeCP2, so conceivably, antagonists to these targets could lower MeCP2 levels in those with excessive amounts (e.g. multiple copies of the MeCP2 gene)

    mGlu7 potentiation rescues cognitive, social, and respiratory phenotypes in a mouse model of Rett syndrome

    In this paper, increasing mGlu7 levels appear to be helpful in Rett syndrome. It appears that mutations in MeCP2 lower mGlu7 levels, and so increasing mGlu7 levels, or using a positive allosteric modulator on mGlu7 receptors may be a treatment path for Rett,

    Again, I would love to have the whole paper, and I may in fact just bite the bullet and get a AAAS membership just so I can read these articles.

    If anyone has access to the full papers, or wants to share their insights on MeCP2, the targets mentioned in the paper that can lower MeCP2 levels, or the impact of mutant MeCP2 on mGlu7, and of course any ways to address these items, kindly share these insights. I will dig in a bit more but wanted to whole community to see these papers while I'm also doing more due diligence on these findings.


  4. Hi everyone,

    Just a quick point of clarification on my previous post - obviously, there is a difference between people who have a mutation in their MeCP2 gene which causes their issues (e.g Rett) versus people who express too much MeCP2 (e.g. gene duplication, maybe an epigenetic issue, etc.) These can be on the opposite ends of the MeCP2 spectrum (i.e. too little properly functioning MeCP2 versus too much functional MeCP2.)

    So the two papers I posted aren't both pointing to the solutions for the same people, but in fact for different populations that are likely mutually exclusive.

    For example, having mutant MeCP2 appears to decrease mGlu7 receptor activity, but we don't know the impact of having too much MeCP2 on mGlu7 receptors. So someone with ASD who has normal (non-mutated) MeCP2, or too much MeCP2 may have a negative reaction to increased mGlur7 activity.

    Just wanted to clarify as the two papers do share MeCP2 in common, but in very different ways, and for different populations in the ASD populations.

    One of the papers specifically identifies that it relates to Rett, but I just wanted to make sure I provided some clarification


  5. Here is some new research that shows the repurposing of an existing class of obesity drugs called GLP-1 agonists which was covered in one of Peter's previous posts here:

    for use in reducing intracranial pressure:

    Press Release:


    An interesting anecdotal theme I have noticed is that many interventions that have the side-effect of reducing blood pressure (which generally reduces intracranial pressure somewhat) whether it be Bumetanide, Diamox, Verapamil, or Agmatine Sulphate which I have had great success with is that they all reduce blood pressure to the degree they are clinically used for that purpose. They all acutely lower blood pressure with the exception of Agmatine which causes a short-term spike in blood pressure and then a longer-term reduction in blood pressure.

    So if excessive intracranial pressure due to macrocephaly related reasons happens to be causing cognitive problems, perhaps GLP-1 agonists might help out a lot with this subgroup of autism or at the very least rule out the intracranial pressure hypothesis. At the moment, there is good evidence Bumetanide helps significantly with cognitive functions of autism, but all we have to go on are the hypotheses of Ben-Ari at the moment which may indeed be correct to some extent, but may not be the primary method of Bumetanide's improvement in cognitive function in some with autism.

    It is also worth noting that a hyperactive HPA axis such as what you find in chronic anxiety as well as with a hyperactive amygdala and hypothalamus, both of which you find in autism, will cause a chronic rise in blood pressure which in populations already compromised by smaller ventricles for CSF, will lead to more intracranial pressure in the short-term and has brain tissue dies, you will find larger ventricles and hydrocephalus in the long-term (this would be an interesting comparison between 2 year olds with autism and deceased middle-aged people with autism). I know studies like this have been done on morphological differences in brain structure in the autistic population, but I am not aware of any that looked at these specific changes over a lifetime.

    For us parents, GLP-1 agonists might be another avenue to explore since they are already approved for diabetes/obesity.

    1. Tyler, Agmatine is indeed interesting. It has numerous effects (NMDA, NO, autophagy …). It also affects ion channels including ASICs (acid-sensing ion channels), which will be the subject of a later post.

      What dosages have you tried?

      How long did it take to show an effect?

      Has the effect been maintained at the same level since you started using it?

      Do you get different effects at different dosages?

      Agmatine is naturally present in the brain, has giving the supplement produced any negative effects?

    2. I have tried very low doses which more or less had a neutral effect. Moderate doses (250mg-500mg) seemed to have a cognitive boosting effect but came with unpredictable bouts of anxiety/rage so I stopped the intervention at that point. Once I read the paper on a dose used on rats, I realized that a higher dose was maybe needed to meet what was reported in rat behavior improvement in a valproate model of autism.

      My reasoning is here:

      The effects are very acute (good or bad). I am not sure if the lose doses were the problem or whether tolerance needed to build up. The literature says agmatine has a very short half-life peripherally but lasts up to 12 hours in the brain. Several researchers dosed themselves for 2grams per day for several years with no adversely reported results. Some bodybuilders/athletes might use even higher doses. I used it myself and there were some peripheral effects, but the strangest one was when I took it shortly before bedtime, I had very, very, very lucid dreams (I almost never remember my dreams except after waking up and then going back to bed again). Others have reported similar anecdotal findings from its use so it might have some effect on the REM/NREM ratio.

      As for chronic use, the cognitive effects are still there, but he does not get the anxiety/rage effect anymore.

      The main purpose for trialing it (this was even before I read the valproate study) was that it is thought to normalize the opioid system in the brain via its action on adrenoreceptors. I wasn't terribly interested in it for its effects in NMDA antagonism (which compared to many drugs is mild), its effects on lowering blood pressure, or autophagy and ASICS (that is new information to me after all I have read).

    3. Tyler, so for your 50kg son, is your dosage still the 1.2g from your June post?

    4. Yes. One teaspoon worth, though today and yesterday he had none because we ran out and my wife and I miscommunicated on our last amazon order. He should have it again tomorrow sometime. I will also note he has a lot more speech at this dose. Not hyperactive gibberish, but real speech.

    5. Tyler, just to confirm, you mean 1 regular teaspoon, which is about 5 ml. I do not know the density of the powder, but is this not more than 1.2g ?

    6. Well on the label of the packaging they have a volume to weight conversion for 1/4 of a teaspoon to milligrams. Obviously weighing the product would be more precise, but those are the guidelines I go by for measuring.

  6. Hi everyone,

    Another potential piece of the puzzle may be in the following paper, called "Novel biomarkers of metabolic dysfunction is autism spectrum disorder: potential for biological diagnostic markers"

    If anyone can access to the full text paper, it would be really appreciated if you would kindly be able to list which way the noted biomarkers are off in ASD (i.e. are they high or low in ASD kids), ideally at least lactate dehydrogenase and caspase 7, but all if possible.

    Thanks very much in advance to anyone who kindly posts this information as it may be very relevant.


    P.S. I don't know if anyone else has noticed this, but many of the more interesting ASD papers I've been seeing lately are coming from King Saud University - always glad to see researchers outside the US and Canada also involved, especially in a country where I didn't even think good research was taking place in ASD.

    1. AJ, most papers are available via , just copy and paste paper's DOI in the search field.

      As for this one:

      "We confirmed previous studies indicating abnormal biomarkers in children with ASD including pyruvate, CK, ETC Complex 1, GSH, GST and Caspase 7. We also identified that only the most severe children demonst
      rated abnormalities in ETC Complex 1 activity and GST.

      Additionally, we found that several biomarkers could be candidates for differentiating children with ASD and typically developing children, including Caspase 7, GSH and GST by themselves because of their favorable sensitivity and specificity and consistency with previous studies. LDH and Complex I were also good candidates when added to other biomarkers in combination."

  7. WOW, anon, I don't know who you are but THANK YOU!!! I didn't know about this site, and was just able to get the full text. I have a few others I was hoping to get the full paper on.

    It was very kind of you to share!


  8. Peter, Tyler, anyone who might have something to comment on Bespar as a novel anxiolytic, would be much appreciated.

    Buspirone an update on a unique anxiolytic agent NCBI

    There is an open label trial of Buspirone for the treatment of anxiety in youth with ASD. No study results posted, first received in 2013, last update in July 2017.

    1. Petra, Bespar affects certain serotonin and dopamine receptors rather than GABA, which is the target of many anti-anxiety drugs. It has been shown effective in ADHD and seems to have less side effects than most anxiety drugs.

      I think it all depends on what dysfunction is underlying the specific person's anxiety, which we can only guess at.

    2. Petra, we tried Buspar many many years ago at recommendation of a DAN we were working with but it did not help - like Peter mentions, could be how this works was not the underlying issue for him at that time. I am sure this is old news to you and you may have already tried, like we did many times before, but right now Omega complete fish oils has made a huge improvement in my son's overall mood (he is now 18). I realize some of the moodiness and OCD had increased since his brother left for college but I think a big underlying issue is inflammation causing the effects on mood etc. We have tried other things for inflammation in the past, even short course ibuprofen etc., and no improvement on mood - but this time with the fish oils, it's been amazing. I think the difference may be the protocol we are using. We are doing the recommendations of high EPA as per CHERAB foundation site.. On another tangent, Tyler has been commenting on agmatine, and one study i read using mice of course, they created OCD by socially isolating the mouse. hmmmm. interesting. OCD created from social isolation. I think this is a big piece especially when the kids get older and realize their differences and the circle gets smaller. This is where having more opportunities and awareness from the public to be more open and welcoming helps. I know when my son is having fun out and about and with people doing fun things as a group, like white water rafting, things out in nature that focuses on the activity with the group as a whole and not about singling him out, zero OCD or anxious moments. Someone mentioned in comments about the Amish and living close to land with no electro smog, farm life exposed to germs etc and connecting with no autism - the other interesting factor with the amish is their big circle of community and everyone caring for each other. That has to make a difference. Maybe not so far to say it would prevent autism but it seems to me it would have to help the outcome of severity. Makes me think of that study in the 1950s -the Rosetto effect and no heart disease. Sorry for the ramble there. Just thoughts that are circulating in my head at the moment.

    3. I probably know about as much as you do about Buspirone, though I think it would do more harm than good in most ASD's because it seems to increase dopamine signaling as a side effect and while some dopaminergic pathways seem to be inefficient such as the ones affecting social reward, in general most ASD's have too many catecholamines in their blood. I would say limit too much serotonin into the brain via BCAA blocking of tryptophan to upregulate serotonin receptors would be a better first choice because too much Buspirone is going to desensitize the 5-ht1a autoreceptors paradoxically leading to excessive serotonin signaling over time as they stop working and serotonin expressing neurons don't ever get the proper signals to stop firing so much, which of course should cause the post-synaptic receptors to downregulate and eventually become desensitized themselves.

      I would not give it to my child except if there was some evidence showing the 5-ht1a autoreceptors are genetically faulty and that a hard push to them will get the proper inhibitory signals going again.

    4. Peter, Tyler, Tanya, thank you for your responses.

      I intend to give 5mg buspirone and no more than 10mg in case of extreme anxiety and not as a treatment. I only used it once and the least I can say is that it didn't have an adverse effect but I can't be sure if it helped.

      Tanya, it's good to hear that your son has been helped with fish oil. The same happened to my son. Is his vitamin d3 ok? Maybe a serotonin dysfunction is their phenotype?

      Tyler, I follow your discussion about high dopamine levels in some autism and you may be proved right, but for my son, the fact that he has such low LDL cholesterol (55), makes it hard to believe he can make neurotransmitters. His vitamin d deficiency may also support this speculation.
      When he was on fluoxetine for about 6 weeks his uric acid dropped to normal. Fluoxetine doesn't normally lower uric acid which may mean that impaired level of serotonin was the case. I stopped fluoxetine when he experienced some kind of psychomotor agitation and speculated that there might have been an increased serotonin to dopamine ratio. I am just guessing of course and don't know if this makes sense.

    5. Hi Petra, I think it is the higher EPA to DHA ratio in this oil that must be making the difference - we did not see this with cod liver oil or supplementing vitamin D. He is ++ for mthfr a1298c which supposedly means lower bh4 levels and therefore low synthesis of serotonin, dopamine, norepi, epi and melatonin. But none of those protocols for that mutation made a difference..
      Have you ever tried higher doses of niacin for anxiety issues - not as treatment like you say, but occassional use in acute situation? I have used this for my son, and myself too actually, and it really calms us. It also increases uric acid , but maybe that only happens with frequent use?


Post a comment