The social interactions that come naturally to most people are difficult for people with autism and Asperger syndrome. Simple matters like making eye contact, reading expressions and working out what someone else is thinking can be big challenges, even for "high-functioning" and intelligent individuals. Now, a preliminary study of 13 people suggests that some of these social difficulties could be temporarily relieved by inhaling a hormone called oxytocin.
The participants, who either had Asperger or high-functioning autism, experienced stronger feelings of trust, showed stronger social interactions in a simulated game, and paid more attention to socially important cues like someone else's eyes. These results will need to be confirmed in larger studies in real-world situations, but for the moment, they're promising.
Oxytocin is involved a myriad of emotions and social behaviours including trust, social interactions, sexual arousal, the bond between mother and child (see SciCurious's epic oxytocin series for more). It has been linked to autism before. Autistic children have lower levels of the hormone coursing through their blood and what little there is appears to be made in an abnormal way. Some researchers are testing oxytocin as a treatment for some symptoms of autism, including repetitive behaviours, but the new results are some of the most interesting yet.
Elissar Andari gave oxytocin in the form of a nose spray to 13 autistic people and 13 typical ones of similar ages and genders. Each volunteer came into the lab twice and was given either oxytocin or a saline spray and neither they nor the scientists knew which was which until the experiment was over.
The recruits played an online game where they passed a ball to three other 'players', who were actually controlled by the computer. The aim of the game was the catch the virtual ball as many times as possible, so it's worth identifying and playing with cooperative characters. Most people do this easily, returning the ball most often to players who pass it to them. Autistic people, however, will pass the ball equally to all the players, regardless of their behaviour.
But a whiff of oxytocin changed all of that. The autistic players became more likely to engage with "good" players who passed the ball to them and shun "bad" players who ignored them (see left side of image below). And they became more and more entrenched in these preferences as the games progressed and the behaviour of the good and bad players became more evident (see right side of image below). They didn't quite match the behaviour of the neurotypical volunteers, but they weren't significantly different either.
After the game was over, a questionnaire confirmed that the autistics strongly preferred the good player and trusted them more if they played under the influence of oxytocin. This suggests that oxytocin helped the autistic players to intuit and interpret social cues that they would normally gloss over.
In another test, Andari tracked the eye movements of her volunteers as they looked at pictures of faces on a computer. As you'd expect, typical people look at the faces themselves but autistics clearly find this uncomfortable. They spend half of their time looking elsewhere and when they do look at the faces, they their dart their eyes around with anxious haste and specifically avoid staring at the eyes. Again, a whiff of oxytocin narrowed the gap between typical and autistic behaviour. The autistics spent more time looking at the face and the eyes in particular, and they explored the images with more deliberate and relaxed eye movements.
Reactions - caution and optimism
A lack of eye contact is one of autism's core problems and oxytocin's ability to overturn it is certainly promising. But looking at eyes in a laboratory test doesn't necessarily translate into real-world behaviour. Between this and a very small sample, it's unsurprising that other scientists are wary of the results. Everyone I spoke said that the study needed to be repeated with more people in realistic settings.
However, many people are optimistic. Larry Young from Emory University described Andari's work as as "promising" and "one of the better papers" in this area. However, he adds, "I would caution the public in their interpretation that this is definitely not a cure. They did these tests in a very controlled laboratory setting and they need more sophisticated studies of real social situations. Does oxytocin help people to read body language? Will they really look at the eyes more?"
Uta Frith from University College London agrees. "Against my initial scepticism... I found the paper very persuasive," she says. But she echoes Young's concerns and she wants to know exactly how the oxytocin effect works so that proper treatments could be designed. "Nevertheless," she says, "this form of pharmacological intervention would be a first in the treatment of the key social impairments of autistic disorder. This could be revolutionary."
Frith also wants to see if oxytocin could work on all autistics or just high-functioning ones. Even in Andari's study, some of the recruits responded strongly to oxytocin but others were totally unaffected. Other scientists have tried to categorise autistic people based on their style of social interactions, and it may be that oxytocin only 'works' for some of these groups.
For example, Andari noted that most of those who showed improvements in the ball-throwing game would be classified as "active-but-odd" - they normally approach other people but do so awkwardly. In contrast, most of the people who didn't show improvements fell into the "aloof" category - they avoid any close contact and actively reject social advances.
George Anderson from the Yale Child Study Centre is more skeptical about the future potential of oxytocin. He notes that the hormone doesn't easily reach the brain it doesn't last for long in the blood. It has a half-life of just a few minutes. Even if future studies replicate Andari's results, it's not clear if oxytocin could ever be used in practice.
Young agrees and says, "I don't think that this particular treatment will become widely used." He sees these experiments as a starting point for developing drugs that mimic the effect of natural oxytocin but with longer half-lives. "We could try to target the oxytocin pathway with other drugs that do penetrate the brain and that have longer half-lives. But I personally think that this kind of treatment may be more beneficial when it's combined with behavioural therapy sessions, with their defined settings."
Advice for parents
For now, many people, especially parents of autistic children, may be tempted to try out oxytocin. But Frith warns, "In previous cases, such as treatment with chelation, hormones or large doses of vitamins, initially positive results which led to immediate applications, had either no effects or rather harmful ones." Young agrees, saying that the risks are still unknown. "I would definitely not recommend that people try to do this on their own," he adds. "It shouldn't be done without the assistance of a physician and I don't think physicians should be doing this until we have more data."
Anderson, despite his skepticism, takes a more pragmatic stance. "It's not recommended but it may be hard for some parents not to try," he acknowledges. "If anyone does go ahead at this point, they could try to blind their efforts by using saline in a similar bottle. One parent could administer the oxytocin or saline and the other parent [who doesn't know which was used] could rate the [child's] behaviour over the next hour. For a number of reasons, this wouldn't constitute a true double-blind study, but it might reduce false-positive responses."
Reference: Andari, E., Duhamel, J., Zalla, T., Herbrecht, E., Leboyer, M., & Sirigu, A. (2010). Promoting social behavior with oxytocin in high-functioning autism spectrum disorders Proceedings of the National Academy of Sciences DOI: 10.1073/pnas.0910249107
More on autism:
- Autism linked to common gene variants that affect the connections between neurons
- Autistic children are less sensitive to the movements of living things
- Same gene underlies two language disorders
- Log in to post comments
Ed, any woman who has ever nursed a child can tell you that oxytocin is one of the best natural highs around. The euphoria it produces makes the nursing worth doing (oh yeah, there are health benefits for the kid too) ;-)
This is actually very exciting result, even if oxytocin can't be used as a treatment (it is a peptide and might cause immune sensitization to it which would be a disaster).
What it does show is that the social neural pathways are substantially intact. Neuronal remodeling can't occur to a large extent in a few minutes. If people with autism can respond to something like this, the neural pathways are present, they just need a little bit different regulation.
I think people should really not try this on their own. Oxytocin is an extremely powerful neurotransmitter. It is what causes bonding. Inappropriate use of oxytocin might cause inappropriate bonding; attachment to abusive individuals (think Stockholm syndrome), or inappropriate sexual attraction (think pedophilia), or dependence on oxytocin (think getting worse and needing oxytocin to come back to what was the old normal). I am just speculating, but this is not something to experiment on children with.
I'm glad you covered this paper! I looked at it myself, and though I found the initial results interesting, it IS a really small study. VERY small. And this may only work in people who are high-functioning. But it's still very interesting, and to Sci it is most interesting because I think relatively few people have really looked at the MECHANISM behind low levels of oxytocin. I'd love to know where this effect is taking place in the brain and if there are other neurotransmitters affected (I bet there are, of course). It'd also be interesting to look at changes in cell firing and the anatomy of the systems.
Daedalus - Good point, and Andari says in her conclusion, "Although the effect we measured here is certainly transient, it serves to show that these patients are quite able to engage in social relationships." Indeed, the social neural pathways seem to be intact. I agree that people shouldn't try this on their own but I included Anderson's point if only because I thought it was interesting for someone to take a fairly pragmatic stance and say, "Well, some people are probably going to try it anyway - what would we say to them?"
Sci - all good questions. The paper talks about mechanisms much more extensively than I've done here, but there are still plenty of unknowns. Lots of great follow-on questions though, as you note. SCIENCE!
This is a really interesting article. For some time oxytocin along with the related hormone vasopressin has been implicated in many social behaviors. These two hormones have also been a major focus for many researchers studying autism spectrum disorder. It is nice to see psychophysical data in human subjects that is promising. I would caution you on your headline. While a "sniff" or "whiff of oxytocin" is a fun way to describe the application, it is not apparent in your subsequent article that it is not in fact the odor of oxytocin that is responsible for the effects observed. Being a peptide oxytocin does not readily cross the blood brain barrier. This makes the administration of the hormone to the brain difficult. For a variety of reasons the olfactory epithelium can be somewhat leaky, allowing the transport of certain large molecules (peptides, heavy metals and even parasites) access to brain while they wouldn't be passed along if in normal blood circulation.
This is great news, to see that legitimate research to help autistic kids is being done. I fear, however, that it won't be long before AoA, Generation Rescue and the like are jumping all over this, making oxytocin the next miracle cure.
Still, huzzah for steps in the right direction!
Are you getting this by way of an embargo that has yet to be fully lifted or some sort of press release or other scheme that people in general don't have access to? Excuse my raising this possibility, but I'm trying to understand why I can't find it. The paper simply doesn't appear to exist at present.
I have searched both at PNAS and PubMed and "Andari" yields zip. Ditto for floundering around trying the most recent results on 'oxytocin'. (Limiting the title to containing 'oxytocin' and in 2009 and 2010 yields only two matches in PNAS, neither the article you refer to.)
Where would one actually GET oxytocin?
Grant: Ed says "If the link doesn't work, it's because this a PNAS paper - you might have to wait for up to a week". PNAS has dumb embargo rules.
You can buy it on the internet. In fact, I'm pretty sure that if something exists, you can buy it on the internet. For obvious reasons, I'm not going to link to anything but a simple Google search reveals multiple vendors. This is why I asked all the scientists about what they'd advise people who went out and bought the stuff.
Cool. It is more than a little troubling that this kind of thing is available so easily. Let's hope pedophiles don't find out.
Making eye contact? That's my biggest social problem... and it's a very real problem... and I can't figure out a way to solve it... and I've tried. Hard. I have to be in a really lovely, serene mood to do it consistently. I have a friend who tried to give me some Ecstasy to mellow me out. I considered it.
Is oxytocin really THAT dangerous?
Oh. never mind.
Oh, but don't you know? Oxytocin repels the mercury that
came from the vaccines that caused the autism.
(Yes, I'm joking. You never can tell these days.)
The neural pathways that involve oxytocin and social behaviors also include NO as a neurotransmitter. If nitric oxide synthase is inhibited in ewes, they do not bond to their lambs.
They say âolfactory memoriesâ, but by how they are measuring it, I think it is really âbondingâ that they are measuring. Maternal bonding is the archetypal mammalian social behavior. All mammals exhibit it, even mammals that are considered non-social. I think that one of the reasons that maternal bonding involves NO is so that it can be coupled to energy status because lactation is so energy intensive. If the mother doesn't have sufficient physiological metabolic resources, that shows up as low NO, and inhibits bonding. I think it is low NO from stress in the early postpartum period that causes stressed rodents to abandon their pups.
Low NO (due to nNOS knockout) does impair long term social behaviors.
as does inhibition of nNOS.
In looking for more stuff on NO and oxytocin, I found this paper
Which demonstrates strong and non-linear effects of neonatal exposure to oxytocin on maternal behaviors and pair bonding in adults. Giving oxytocin to children could have major and unpredictable effects on their psychosocial and psychosexual development. These are non-linear effects, so dose, dose rate, dose timing, and likely many other things will affect the outcome. Anyone who says they know what will happen is lying.
I have no doubt that there are quacks who will sell and administer this to children. There are quacks who administer Lupron to children and adolescents too.
speedwell, I would be very careful of using ecstasy (I never would). I think it works as a stimulant by inducing ATP depletion and the euphoria of the near death metabolic state. That is an extremely dangerous state to induce, it nearly always causes neuronal damage. Damage that can't be measured until it accumulates (because young people have a lot of redundancy), and then it is too late; think Alzheimer's.
I can read ;-)
My reading of Ed's words where that that *DOI link* might not work, not that the article might not be available (he linked to dx.doi.org, not pnas.org) and that the delay he was referring to was the provisioning of the DOI for the article. I went directly to pnas.org to find the paper to bypass the DOI link, but couldn't find it. It really ought to be there IMO. It seems very strange to have a paper "released" to some for general public release, but not be available to the science community for examination and comment.
Oy. See here and here and here and here
Grant: it's odd indeed. But that's PNAS for you. Journos get it first, then the scientists. Anyway: it IS now available.
"What [the oxytocin treatment] does show is that the social neural pathways are substantially intact. Neuronal remodeling can't occur to a large extent in a few minutes. If people with autism can respond to something like this, the neural pathways are present, they just need a little bit different regulation."
They've found disorders in other neurotransmitter pathways, too. I suspect the cognitive architecture is fundamentally not actually that different from a non-autistic person.
Will they test this on non-high-functioning autistic people? What biologically differentiates high-functioning autistic people from non-high-functioning autistic people?
"I'm glad you covered this paper! I looked at it myself, and though I found the initial results interesting, it IS a really small study. VERY small. And this may only work in people who are high-functioning. But it's still very interesting, and to Sci it is most interesting because I think relatively few people have really looked at the MECHANISM behind low levels of oxytocin. I'd love to know where this effect is taking place in the brain and if there are other neurotransmitters affected (I bet there are, of course). It'd also be interesting to look at changes in cell firing and the anatomy of the systems."
Have we figured out the mechanism behind HIGH levels of oxytocin? That might be a key.
Also, I wonder what happens when autistic women give birth.
There's also a theory milling around, which I'm approaching with a certain amount of skepticism, that autism might possibly be contributed to by oxytocin during labor, either natural oxytocin or artificial oxytocin, causing the oxytocin receptors in the infant's brain to downregulate.
According to at least a few studies, this theory has been pretty much tossed out.
Because Ed requested links -
http://research.yerkes.emory.edu/Young/Getz/2003%20Carter%20PB.pdf (page 11)
Have you considered that there might be an explanation for the autistics' behavior besides a "social deficit?"
Consider what they are doing sans oxytocin:
Is it cluelessness or altruism that we're curing them of?
I say this because I'm an adult autistic, and I've always seemed to be more sensitive to others' suffering than the neurotypicals around me. I'm vegan, I buy organic Fair Trade-certified food, I worry a lot about people or animals that I see in person which appear to be in pain, and I can't even stand to watch movies where innocent people are getting hurt. And I realize that this is anecdotal, but it's still worrying to me that no one is considering this as an alternate explanation. Is everyone's opinion of autistic people as headbanging kids that entrenched? Does nobody know (or know they know) an autistic adult?
If I had to play this "game," I'd probably pass the ball more often to the player who didn't unfairly favor anyone. Although if she caught on to what was going on and started to favor me in return, I'm not sure I'd feel bad about excluding the other two players. After all, they each didn't consider one of us to be important, so I'm not sure they deserve anything.
Oh ... altruism and a lack of hypocrisy, I mean, may be what's being cured. What else could allow them to see certain players as "good" and others as "bad" when they're both doing the same thing and excluding other people?
Readers may also be interested in autism researcher Michelle Dawson's review of this paper at The Autism Crisis and scientist Emily Willingham's objection to how people with autism are characterized A Life Less Ordinary.
"Cool. It is more than a little troubling that this kind of thing is available so easily. Let's hope pedophiles don't find out."
I take your point, but I've been injected with ultra high doses of oxytocin (or pitocin, to be precise), and I'm not so sure you'd have to worry about it. And, in this study, it *helped* the HF-ASDs identify the bad apples.
"Is oxytocin really THAT dangerous?"
I'd argue it's probably better than poorly prepped MDMA. But it can hurt like a bitch, if you've got a uterus.
"I say this because I'm an adult autistic, and I've always seemed to be more sensitive to others' suffering than the neurotypicals around me."
anecdote- I appear to be more indifferent to *general* suffering since my blast(s) of oxytocin... and, much, *much* more sensitive to my baby's suffering than I've ever been to another person before. Might be evolutionarily advantageous, that. Or possibly just me rationalizing post-hoc.