Friday, January 25, 2008

A warning on drug trials!

Ora sent me this article (you need to be registered to access but it's free!) from the New York Times on how drug trials are misleading:
Researchers Find a Bias Toward Upbeat Findings on Antidepressants
The makers of antidepressants like Prozac and Paxil never published the results of about a third of the drug trials that they conducted to win government approval, misleading doctors and consumers about the drugs true effectiveness, a new analysis has found. In published trials, about 60 percent of people taking the drugs report significant relief from depression, compared with roughly 40 percent of those on placebo pills. But when the less positive, unpublished trials are included, the advantage shrinks: the drugs outperform placebos, but by a modest margin, concludes the new report, which appears Thursday in The New England Journal of Medicine.
My conclusion is two-fold. First, I am convinced that virtually all studies suffer from this publication bias effect (i.e. null or negative trials are conveniently not published or non-publishable) including stuttering research (not just on medication though). Second, I am always amazed at how strong the placebo effect is, and one can certainly argue that even the placebo is a significant improvement which the patient wouldn't have gotten if s/he hadn't taken the worthless or minimal-effect medication!!! Strange world...

Thursday, January 24, 2008

Lack of motivation a convenient excuse?

When I talk to therapists and ask why many (most?) patients are not succeeding in therapy over the long-term. I often here things like "you are not motivated enough" or "the moment is just not right" or "there are no cures" or "we just don't know enough about stuttering" or "you need to practise more". Could I not turn this around and blame relapse on the therapist because s/he was not able to motivate the patient hard enough or made practising unnecessary difficult due to bad advice?? Have you ever heard a therapist at a conference talking about their mistakes? I haven't to this day. I only heard talks about success, enriching personal development experiences for the patient, and so on. And they also revel in showing just how sensitive, caring, bonding, holistic, intelligent, hard working, smiling, and most importantly professional (what ever that term may mean?) therapist they are; attributes that, to put it mildly, people who stutter are not that interested in, all they want are results for them!

Here is my analysis. Yes, it is true that motivation plays an important role in long-term success. You do need extraordinary motivation. For example, I was absolutely motivated at each of my therapies, though not obsessively. Even a good therapist might not be able to motivate some patients. However, this argument even though it is correct to some degree is also a very dangerous argument, because mediocre therapists can use it as an excuse for their own failure. They can hide behind it. This is also true for non-professional therapists like the crackpot award winners that I nominated. A healthy advice would be: if you are a patient and you fail, look at yourself to find points of improvements; if you are therapist and your patient failed, look at yourself to find points of improvements!

And here is a challenge to all therapists: I will buy that therapist a drink who gives a talk with the topic "The mistakes I made and why they were mistakes"! I would even offer a meal, but a meal with me might be too much punishment for the winner! :-)

Monday, January 21, 2008

The Valsalva hypothesis

Virtually all theories on stuttering from amateur (and many professional) researchers that I have seen are just wrong and void of empirical evidence. However, there is one theory where I just cannot make up my mind on whether it is a piece of the stuttering puzzle or just complete non-sense! It's Malcom Perry's Valsalva mechanism hypothesis. Claire Walker nicely summarises the hypothesis here:
So, stuttering has been linked to differing neural patterns, but why does stuttering cause different parts of the brain to be more active? One idea is called the Valsalva Mechanism, which is a natural bodily function, but it may turn the extra effort put into speech into the block that stutterers fight with everyday (3). The Valsalva Maneuver was named after an Italian anatomist, Anton Maria Valsalva and its purpose is to bring more air pressure into the lungs to help a person exert more force on an object, such as weight lifters who hold their breath when lifting large masses. (3). To create more air pressure in the lungs, the abdomen muscles contract and press against the diaphragm, which in turn presses up on the chest cavity. For this mechanism to increase the air pressure in the lungs, the larynx has to tighten around the airway so that the air cannot escape and this is called the effort closure (3). All of the muscles involved in the Valsalva mechanism are connected neurologically, so that they can all contract at the same time and with the same force (3).

The reason that this mechanism is thought to be tied to stuttering is because stutterers put a lot of force and effort into the words that they stumble over. This force causes the lips and tongue of the stutterer to press harder together, thus creating more air pressure in the lungs, but also causing speech difficulties (3). Fluent speech actually requires very little effort, so when a stutter puts a lot of effort into speaking the Valsalva Mechanism does what it is supposed to do, it is an instinctive reaction when we are trying to force something out of the body (3). This confusion between the Valsalva mechanism and the neurological components of speech can happen because there is neurological tuning involved in the motor neurons that control all the muscles involved in a movement, this includes speech (3).

What really happens during a stutterers speech is not wholly known, but if we take an example of someone stuttering on a word starting with p, say plane, then the brain remembers that p-words were difficult to say. Thus the brain ‘thinks’ that more effort has to be put into saying p-words, so the Valsalva mechanism kicks into gear and the stutterer is left squeezing their lips and trying to get any p-word to come out of their mouth (3). What makes p-words difficult for many stutterers is the fact that you have to close your lips, momentarily, to build up a little air pressure to say p-words. However if the nervous system is too excitable the brain may misinterpret this signal to mean that a Valsalva maneuver is being started and thus try and shut the air in the lungs, making speech very difficult (3).
Here is William Parry's own summary from his website:

The Valsalva Hypothesis postulates that excessively forceful closures of the mouth or larynx associated with certain types of stuttering, as well as difficulty in phonation, may involve a neurological confusion between speech and the human body's Valsalva mechanism.

Persons who stutter may have learned to activate the Valsalva mechanism in an effort to produce words, as if they were things to be forced out of the body. Such activation is most likely to occur when the stutterer anticipates difficulty or feels the need to use extra effort to speak properly. While this might instinctively feel like the right thing to do, it actually makes fluent speech impossible.

When a person who stutters prepares to exert effort in speaking, excessive neuromotor tuning of the Valsalva mechanism may occur. Such tuning may:

render the Valsalva mechanism over-excitable to triggering stimuli, such as the increase in subglottal pressure that accompanies the start of articulation;
interfere with normal prephonatory tuning of the vocal folds (because the larynx is instead being neurologically tuned for effort closure), thereby causing delays and difficulty in phonation; and
contribute to abnormal laryngeal behavior, either by increasing the strength of laryngeal reflexes (the Valsalva maneuver is known to strengthen reflexes), or by interfering with the suppression of reflexes, which normally occurs during speech.

The Valsalva Hypothesis views the Valsalva mechanism as only one of many factors involved in stuttering. For example, a stutterer's anticipation of difficulty might be affected, to varying degrees, by attitudes, expectations, neurological impairments, or emotional factors.

Furthermore, a child's original disfluencies may be caused by a variety of neurological, psychological, or developmental factors not involving the Valsalva mechanism. These original causes may be different for various individuals, but their commonality is that they create the perception that speech is difficult and will require extra physical effort. The child — already accustomed to using the Valsalva maneuver when exerting effort or expelling bowel movements — may instinctively assume that words can be forced out in the same way. Continuation of this behavior during certain critical years of childhood may influence the development of nerve pathways in the brain. Over time, these behaviors become deeply rooted in the nerve pathways of the brain, making them extremely difficult to change.

For further discussion of the Valsalva Hypothesis, see Stuttering and the Valsalva Mechanism: A Key to Understanding and Controlling Stuttering.

Sunday, January 20, 2008

Stuttering is cool - blog

My blog seem to become more and more popular. I receive 5 emails this week! Here is a new blog with audio comments, in case you are interested:
Hi Tom,
I'm a life long stutterer and also a regular reader of your blog. I never quite read up on the science of stuttering (you think I would have!) but your posts are fascinating. Anyway, just thought I'd send you a quick email in case you'd be interested to know that I've recently launched a podcast for stutterers. It's called "Stuttering is Cool" and it's focused around audio comments submitted by my fellow stutterers. For me, hearing other people stutter really helps me feel that I am not the only one. And it's great to meet other stutterers!

Feel free to send in your audio comments, too. The address is

It's fun to know - blog

I discovered a new blog called It's Fun To Know. He also talks about stuttering: see here. He seems quite witty.

Thursday, January 17, 2008

Twins: Little shared environmental effect

I have just re-read the 13000-twin study by Katherina Dworzynski et al and she clearly empirically confirms what I have been saying in my last posts:
Stuttering appears to be a disorder that has high heritability and little shared environment effect in early childhood and for recovered and persistent group of children by age 7.
And further,
With respect to the genetic effects throughout, it also needs to be emphasized that, even though substantial estimates of heritability were obtained, many monozygotic twin pairs were discordant for stuttering. This is further evidence for the importance of the child's unique environmental influences, in that specific stressors have unique effects when a genetic liability is present.
Again, the nature vs nurture fallacy is revealed. It is simply wrong to divide the world into nature and nurture. There are three categories: the genes (nature), unique environmental stressors (more or less random events), and shared environment effects (nurture). Think about being the CEO of a small start-up company. Unique events like you having a car accident or a virus infection, sheer bad luck, pregnancy of your most important employee, and so on can completely change the outcome of your company's success because you only have a few months of financing to hit a certain target before your bank or venture capitalist cuts financing (window of development for child). I have seen this many times. Of course, you are especially vulnerable to such events if you have a bad or weak business plan (think of this as the genes - the instructions on what to do). General market conditions are also important but not that crucial, and can increase or decrease vulnerability. I never thought that venture capital, the topic of a book I have written (see here), is relevant to stuttering! :-)

Wednesday, January 16, 2008

The Genain quadruplets

The Genain quadruplets are a "beautiful" experiment of nature that can teach us a lot about genetic and environmental factors. All four girls share exactly the same genes, and lived in the same family environment. Unfortunately for them but fortunately for science, they share exactly the same genetic predisposition to schizophrenia. In exchange for free treatment, the quadruplets were and still are monitored; they are now about 50 years old. Interestingly, the expression of schizophrenia was very different between the quadruplets: Myra after an initial episode was never hospitalized and is married, Nora was hospitalized repeatedly but had marginal adjustment while outside the hospital, Iris experienced more-chronic hospitalization and her symptoms range from severe catatonic withdrawal to times of marginal adjustment when she was able to leave hospital, and Hester further deteriorated and remains hospitalized! Check out this review: here.

The case study shows that the same genes for schizophrenia (and many other brain disorders like stuttering) can lead to very diverse outcomes, though all quadruplets did develop schizophrenia. The family environment including parenting skills (nurture) were the same for all four, though parents might have treated them differently. However, a difference in treatment is mostly due to the kids being different. So the parents react to the difference of the child.

So why do the Genain quadruplets develop so differently? The answer lies most likely in environmental stressors unique to a child. I would split it in pre-natal and post-natal factors. Pre-natal factors include the position of the child in the womb in relationship to the others, differences in weight due to differences in nutritional supplies, injury, and so on. And post-natal factors include illness, injury, unique random events like a traumatic experience, reactions of parents to behavior proper to one child and so on.

Tuesday, January 15, 2008

What's environment?

In my last posts, I spoke about onset of stuttering and about triggers. I also said that I am very skeptical that environmental factors play a leading role, especially at onset. It's the eternal debate between nature and nurture. But the more I think about this issue, the more I am convinced that I and many others have been confused and it is the really the debate between nature, events and nurture. Anything that is not attributed to genes is often called environmental. This term is strictly speaking correct, but environment is not just nurture. Rather, environment is the general environment that the child (and his siblings and friends) lives in AND unique events that happen to the child within the environment: like virus infection, head injury, traumatic experience, and so on. I would argue that, relatively speaking, events are driving divergence between siblings because they cause a large difference between the one hit by the event and the sibling which is not, as opposed to the general environment with its similar impacts on siblings.

We are definitely the product of our genes, of our events, and of the immediate environment (of family and friends) and culture that you grew up. And, even nurture is not just nurture by parents, but also by other family members and most importantly by peers. So you can see that parents actually often do not have such a big role at all. In my next post, I will talk about the Genain quadruplets (mono zygotic) and their life with a genetic predisposition to schizophrenia in order to explain my ideas with a concrete example.

Finally, as a physicist, I cannot resist pointing out to all of you that the environment is not really well-defined. Where does it start and "you" begin? You are really nothing else than a bunch of atoms! Is the air you have in your lungs part of your environment or not? and your blood? The millions of bacterias in your stomach? The 90% of atoms especially water that are completely swap every so many years and you are not even the same person physically speaking than we were 10 years ago!

Thursday, January 10, 2008

Uphill struggle for the environment

I discussed the danger of associating causes to the onset of stuttering. Formulating the wrong questions is also confusing the minds of nearly all parents, and many therapists and researchers, though not all.

First, it is misleading to ask or try to answer the question: What causes stuttering? It is important to disentangle onset of stuttering and lack of recovery from stuttering. The onset of stuttering does not mean onset of life-long stuttering. An adult stutterer needs to go through both onset of stuttering and lack of recovery. If someone asks you about what causes stuttering, you should say: We must ask two questions: first what causes the onset of stuttering, and second what causes some not to recover. The two questions very likely have very different answers.

Second, it is misleading to ask what factors are to be blamed for the non-recovery of some kids. This point might be a bit more subtle. Factors, that cause some not to recover, are not really to be blamed for the non-recovery, but they are just not strong enough to overcome onset of stuttering. Think about overcoming stuttering as climbing the Mount Everest. Are you going to blame the bodies and minds of 99% of the population to cause non-ascent to Mount Everest? No, you rather say that 99% are just not good enough to climb Mount Everest as compared to the Top 1%, and you would re-assure them that they have normal bodies and minds. A better but still not perfect way to formulate the question is what causes most to recover. The question should really by: Under which (exceptional) circumstances are kids able to recover from onset of stuttering?

Let me be a bit more concrete. I will give you my big-picture guess on what is happening, and you can hopefully see why these questions are misleading us. The onset of stuttering is the failure to develop certain standard neural pathways because the construction plans are non-standard due to abnormal genes or because the construction was hampered by a neurological incident like head injury, virus infection, low birth weight. The moment of onset is the moment the system gets its first real test in the same way that your brand new but defective car engine runs smoothly until you hit high speeds on the motorway. There is no or little environmental impact like parenting skills or endured stress, except for the out-of-the-ordinary events like injury or virus infection. The onset is entirely caused by internal factors that mostly happen long time before onset like the genes or like the head injury at age 2. Your car engine has a design fault or a manufacturing problem in the factory long time before you hit the high way. Environment and outside events might make stuttering appear a bit earlier or later, but it would have appeared anyway. You could have hit the highways a week later.

At the onset, stuttering children have abnormal neural pathways which cause them to have dysfluencies to which everyone reacts someone somewhat differently in terms of stuttering symptoms. The latest brain imaging results suggest that all children (even those that later recover) have structural abnormalities. Most kids are recovering but some are not. Why? Here is how I think about it, and I use the same escalator analogy for explaining why matter cannot escape black holes! :-) Think of a downward escalator that you want to walk up. You have to overcome the downwards speed by walking up faster. The downward speed is the difficulty to create a second compensatory pathway related to the magnitude of the neurological abnormality. In order to reach the top of the escalator i.e. to recover from stuttering, you need to have an upward walking speed greater than the escalator's downward speed. The upwards speed is determined by whether the kid does the right things, which depends again on genetics like its brain plasticity, semi-genetics like temperament and purely environmental factors like good parental feedback. The key point is that the kid needs ABOVE AVERAGE compensatory skills and environment, in the same way that you need above-average walking speed to climb up a escalator with a standard speed. So those kids that only have average skills and environment to learn compensation or those who have extreme downward speeds, will fail. The majority who recovers was lucky because either the escalator had a slow downward speed which was well compensated by their average skills and environment, or the escalator had a serious downward speed but they had exceptional compensatory skills. For example, girls are more likely to recover because they have a higher ability to compensate. So finally can you really blame their average skills and their environment for stuttering? No.

Wednesday, January 09, 2008

The fallacious mind.

I want to talk more about the onset of stuttering and how parents, therapists and researchers often draw fallacious conclusions. They often see the onset of stuttering being related to X (be it the birth of the younger sibling, the kid being attacked by a dog, high fever, below-average language & speech capabilities, and so on), and conclude that there is a direct causal relationship. However, this is fallacious, because there are other explanations for a relationship that are not causal. The human brain has evolved to automatically search for explanations. If an event happens, you scan your environment or immediate past for explanations, and stop when you found one. This effect is especially pronounced when the event occurs suddenly like the onset of stuttering often does. We need explanations. If you cant find them, we still find them: Blame the gods.

Example: My kid starts stuttering. His brother was born a few weeks ago or I am heavily pregnant. Therefore, onset is caused by stuttering. Or, some kids that start stuttering have slower-than average language development. Therefore, slow language development causes stuttering. Or, I heard my kid stuttering the first time a day after a big dog scared her and she broke out in tears. Therefore, the dog causes stuttering. Or, my kid cried a lot and became more withdrawn, and then she started stuttering. Therefore, something cause it to be distressed which causes stuttering.

Lets be rigorous and systematic, if A is related to B, then many combinations are possible in order of likelihood:

A and B are not causally related but a pure coincidence i.e. related by chance.
A and B are not causally related because C caused A and B to happen.
A and B are not causally related (in the right order) because B caused A.
A and B are causally related.

Each of the four example can be explained by a non-causal explanations.

Example: My kid starts stuttering. His brother was born a few weeks ago or I am heavily pregnant. Therefore, onset is caused by stuttering.

Possible Non-Causal Explanation: Both onset and birth are indirectly related, i.e. they both often happen at roughly the same time due to the nature of human development and off-spring production.

Example: Some kids that start stuttering have slower-than average language development. Therefore, slow language development causes stuttering.

Possible Non-Causal Explanation: Kid had a neurological incident (head injury, virus infection, perinatal hypoxia, low birth weight) or genetic makeup that produced havoc in the developing brain leading to BOTH slow language development AND stuttering.

Example: I heard my kid stuttering the first time a day after a big dog scared her and she broke out in tears. Therefore, the dog causes stuttering.
Possible Non-Causal Explanation: Both event happen by chance at the same time. The kid like millions of others just got scared and cried, and that's it! You can always find a event to relate to onset. Kid started stuttering and something more dramatic ALWAYS happens like death of family member or friend, distressing situation like doctor's visit, thunderstorm, and so on.

Example: My kid cried a lot and became more withdrawn, and then she started stuttering. Therefore, something cause it to be distressed which caused stuttering.
Possible Non-Causal Explanation: This is a cool one. Actually stuttering caused the distress! The kid started to feel that it couldn't pronounce well in private, and it got distressed. The mother noticed this but not the dysfluency which she only noticed a few days later!!! The mother made the mistake in that she thought she had all informations on the events which she didn't have. She just saw one side of the story line!

I explicitly wrote possible non-causal explanation, because if such an explanation exists, it does not imply that it is the right one. There might be a causal explanation that is the correct one or that the effect is a combination of 2 or more explanations. It is only possible to find the right one from the many possible one by careful research. But my rule of thumb is: If there is a non-causal explanation, it will be the right one in 95% of the cases! (because most relationships in life are non-casual)

Gareth clueless on research

Gareth Gates, runner-up of Pop Idols and best-selling pop star, is definitely an inspiration for many of us on how you can really become more fluent when you work hard and are very very focused. The McGuire program is providing a good framework. However, in terms of him explaining why he stutters and why he is fluent while singing, I am sorry to say but Gareth is just a COMPLETE DISASTER. Unfortunately, he is very misinformed. So he (and the other participants) is a mixed blessing for people who stutter, because they re-enforce the message that it is just psychological or "I don't stutter when I sing because I act and I am not myself" (So who is singing, Gareth?) Success does not imply that their theory on stuttering is correct.

Gareth and McGuire people, read my blog on the latest stuttering research! Your methods are very helpful for many, but your theories are confused, fallacious and wrong.

Tuesday, January 08, 2008

Drayna talks on genetics at is encroaching on my territory by content, interviewing Dennis Drayna on genetics research (download here), and overtaking me by design...

Dennis Drayna doesn't really say anything new. I think childhood stuttering is a key part of the genetics equation. It might very well be that there are genes for onset of childhood stuttering and genes for recovery of childhood stuttering, and they are not correlated!

Shut up or go to sleep!

This piece of research gives interesting insights into the learning process, which might be helpful for learning speech techniques more efficiently. Two conclusions I would draw. First, you should take a nap after an intense practise. Second, you should avoid at all costs performing conflicting behavior (like using "anti-speech techniques") after practise. For example, you practise hard and then take a break and do not focus on technique while speaking. It is probably much better to just shut up for a few hours (at least 1 hour, actually a nap would do the same thing), or practise in the evening and go to bed because then you don't speak for 9 hours!

This research shows nicely that we might be able to improve behavioural stuttering therapy not by learning about stuttering but by learning more about learning in general.

Monday, January 07, 2008

Lies, damn lies, and younger siblings

I have spoken about the danger of linking the birth of a younger sibling to the onset of stuttering before. I feel obliged to write about this again because I have read it in Barry Guitar's newly update book which is supposed to be a bible among speech therapists.

I have been looking at the empirical numbers. Unfortunately, I could only find the birth spacing of women in developing countries. The younger sibling is born

<2 years after birth of last child with a probability of 25%
2-3 years later with a probability of 25%
3-4 years later with a probability of 25%
>4 years with a probability of 25%.

Lets assume the older sibling starts to stutter at age 3, then in 50% of the cases his younger sibling is born within a year, and in 25% (50% divide by 2) within half a year.

So in 25% of the cases, parents can perceive a strong correlation between onset of stuttering and birth of younger sibling. And it's a complete illusion!!!! It is a beautiful example of the common logical fallacy of confusing correlation with causation.

Note: I am not saying that no relationship exists to exact time of onset (i.e. when it starts now or a few weeks later) or strength of onset (i.e. whether the frustration or general well-being level is higher resp lower). But the coincidence explains very well any observation... and if there is a relationship it is very weak.