Wednesday, April 23, 2008

We only look under the lamppost

I have become more and more aware of a fallacy committed by researcher in stuttering; a fallacy that is hard to avoid unfortunately but not very strongly stated anywhere. Many are not really aware just how prevalent and strong it is.
It is the looking-under-the-lamppost fallacy. Here is the story: A policeman sees a person searching the ground in the light cone under the lamppost at night. The policeman approaches, and notices that the man is a bit drunk and is looking for his key. Trying to be helpful, he asks: So where exactly have you lost them? The drunk points with a finger to a location a bit further away from the lamppost. Surprised, the policeman says: So why are you not looking there? Well, it is too dark to see anything. I have to look for my keys somewhere, don't I? I don't want to give in, and there is a small chance that I actually lost in around the lamppost!
In the last years, the view that we should study children rather than adults to understand the causes of stuttering has become increasingly popular and rightly so. I wrote a post about it. I also heard it repeated again at the Antwerp conference by Nan Rantner and Ed Conture. And indeed brain imaging work has moved in this direction, with work by Chang and by Watkins. Children's brains are still fresh, and unspoilt from decades of stuttering.
However, current stuttering research is not exactly doing what it preaches. In fact, the absolutely most important time window in children is completely, and I mean completely, ignored. It's the time before onset of stuttering and at onset. The dormant phase where the deficiency is present but unnoticed. I just know of one study that looked at infants. 99.999% of the research starts a few weeks and months after onset. Many researchers do work on children precisely, because they want to avoid the complexity of adult stuttering. However, they are not consequent enough, and stop at onset. Onset of stuttering is an arbitrary moment for the search of causes of stuttering. Think of a rock emerging from a receding tide. Is the moment of emergence relevant for the existence of the rock? No!
I am sure some of you are not happy with my comments. After all, how can researchers study kids before onset because the kids don't stutter yet? And who am I to criticise the researcher when I have no solution ready. First of all, my criticism is valid irrespective of whether I have a solution: if you are over-weight, and I point it out without offering a solution. Is my observation incorrect? Second, I have never seen the fallacy stated clearly and explicitly. In fact, every study on kids on onset should have a health warning: "We did this study knowing full well that the time before onset is as or even more important that the weeks and months after onset to discover the causes".
And, there are possible avenues. Of course, 20 children at onset means that you need at least 20 times 20, 400, children before onset to get the stuttering children. So any reasonable study would include 1000 children. That's difficult to do. What might be possible is to only monitor all children from stuttering families!


Scott said...

Tom -

Have you ever examined or reviewed the research done by Dr. Ehud Yairi and Dr. Nicole Ambrose of the Stuttering Foundation of America regarding childhood stuttering? They are considered the foremost experts on the development of stuttering in children over here in the States.

Thanks -

Tom Weidig said...


No, I have not looked at their articles in detail, just the summaries. I should definitely look closer at them.

Again, they exclusively focus on the kids starting a few weeks/months after onset and commit the fallacy!

I am not "blaming" anyone. I am just saying that we artificially restrict our observation for practical reasons, and we do not talk about our bias explicitly.

Best wishes,

Susan said...

The speech therapist recommended to me by the Stuttering Foundation of America recommends a Easy Relaxed Approach with Slow Movement Therapy for the treatment of Stuttering with stuttering children. In the workshops I've attended, the presenters they have speak of curing about 75% of children who stutter using this. Is this a cure therapy? Should I use it with my child?

Any suggestions, Tom?

Tom Weidig said...

First of all, between 70% and 80% of all children recover naturally with or without treatment!!

So if they say that they cure 70% to 80%, it is a bit misleading. Yes, 70% of the kids they treat are fluent, but they would be any way without treatment.

In my view, there is no clear evidence that treatment makes kids who would not have recovered fluent.

Of course, as a parent, doing something is better than doing nothing. What do you have to loose? At worst, your child does a therapy that does not work, at best it works. So there is no real downside. And I cannot point you to one treatment that stands out beyond doubt.

Therefore I would just do it, but I am not convinced they found a cure. They certainly have not done any scientific trials to prove the efficacy of their treatment.

Try to get a therapist that has long experience and that you and your child like personally.

Sorry if I am vague, but I am just telling what I think is the reality...

Good Luck,

Mike said...

Tom -

I do agree with you on your thought regarding 70 - 80% recovery with out treatment.

Do you -- as a person who stutters -- differentiate between disfluency and stuttering? Are they the same?

And in your therapy experiences and knowledge of stuttering, do you think there is any risk the therapy proposed by the SFA professional would/could increase the serverity of stuttering in the child?


Mike said...

Tom -

I am wondering if you would be so kind as to answer the second question of my post re: the recommended thearapy potentially could/would increase the severity of stuttering as a possibility?

Tom Weidig said...

Hi Mike,

sorry I don't check older posts often.

>> And in your therapy experiences and knowledge of stuttering, do you think there is any risk the therapy proposed by the SFA professional would/could increase the serverity of stuttering in the child?

No, in general I do not think so. Maybe a kid could become traumatised, but the therapy itself should not be harmful. Of all the studies, they did not show an increase in stuttering. My guess is that most benefits are short-lived, so if there are any downsides they are too.


Susan said...

I have a friend who is a speech therapist and suggested that I ask the SFA therapist to put me through the proposed therapy for my child.

After experiencing the therapy of slowing down my speech and the methods tried on me, I chose not to do it with my daughter. It created to much "speech" stress in me, and I did not feel comfortable at all with the disfluencies being pointed out in my speech. My daughter would have hated it.