Friday, October 22, 2010

On causality, correlation, experience, and interpretation of experience

Peter's comment is a good example of fuzzy thinking on causality, correlation, experience, and interpretation of experience, and a way of thinking that I often find with many but not all clinicians.
"David Seidler says that his stuttering was triggered by their moving to the US from Britain! Oh god. I hope he doesn't repeat that line too often. If you know him, tell him that millions moved to the US and did not start stuttering."

MANY PWS have stated that their stuttering was triggered by a stressful event. This should not be a point of debate, it's common knowledge. There is a difference between "triggered" and "caused". Why do you take issue with this?
Here is the resolution:

1) Yes, "many pws have stated that their stuttering was triggered by a stressful event", but they are deluded.

a. humans tend to reduce events into a single event. Look at 9/11, there is no single event at time T, but a series of events on the time scale of minutes, days, weeks, and years, on the spatial scale of
centimetres, meters, kilometres, and even continents away by, and on different regions of the brain, humans, groups, and even nations.

b. humans are re-writing memories when recalling and it becomes even more idealised.

c. humans tend to view their interpretation of what happened to them as the truth. But there are other interpretations of the events that had been happening to them.Now if a girl refuses me, my interpretation is that she is blind, and her interpretation is that he is an idiot. Same experience, but different interpretation.

2) We need to focus on what they say has happened to them and not on their interpretation.

3) So if we asked them about their experience and not their interpretation, they will tell us "My stuttering occurred at the same time as the stressful event X". If we dig deeper, they will say "Y told me that I started stuttering roughly around the period of the events of X happening. Y thought that the event X must have been stressful. I have similar memories but they are vague because I was so young."

4) So the real observation is a correlation between stuttering and a period of the events of X.

5) There are four possible interpretations:

a. S and X are randomly correlated, which is by far the most likely. Because whenever we start stuttering, there is almost always a period of event X close by, e.g. birth of younger sibling, divorce, death, moving, being scared, bitten by animal, fall, virus infection and so on. So you can always find an event X to correlate to stuttering.

b. S causes X. This is unlikely, but maybe stuttering causes divorce because mother blamed father and this was the last straw. And later it becomes the bloody father caused you to stutter.

c. X causes S. This option is possible but not the only option. Here we need to distinguish between trigger and real cause. A trigger is something that initiates the event but is not truly responsible, e.g. a flooding made the dam break but the dam had cracks and it was only a matter of time until ANY major event would break the dam. There might be cases in children who start stuttering just after a stressful event but would have started stuttering anyway. And a real cause, for example massive stress leading to real brain damage, a virus infection, a fall. But then we need to explain why it happened.

d. C causes X and S. Another event causes S and X. Like a virus causes stuttering and a long stay at hospital. But it was not the hospital stay and the crying that causes stuttering.

Conclusion: Because Peter does not understand the subtleties and fallacies, he assumes something to be true, and for him "it's not a matter of debate", and bases his theory on it, but he is dead wrong.

10 comments:

Peter Louw said...

Hi Tom

I agree with much of your response, yes indeed we should be careful before making inferences. But take the following example. I remember a case history, mentioned by a SLP, where a medical doctor started stuttering while in huge stress – he was treating patients during and after a bombardement in wartime. When this crisis ended, his fluency returned (though his stuttering would occasionally come back in later life when in stress). Would it be fuzzy thinking to conclude that the stress and the stuttering were somehow linked? I think it would be a reasonable inference. Surely it would be fuzzy thinking to see the stress and the stuttering as unrelated?

I am, of course, not saying that ALL stuttering is triggered by a single traumatic, stressful event, but some cases do indicate this.

Regards

Peter

Ralph said...

You are right. For years I believed my stuttering didn't start until I was 10 years old, which coincided neatly with the birth of my sister (I had been an only child until then), and so speech therapists and psychologists liked to pinpoint this as the stressful trigger event.

However, my last speech therapist, which I started seeing just last year, said that could not be quite right, since childhood stuttering typically only starts as late as 8 years old due to some characteristics of the brain. So she said I was probably showing some early signs of stuttering that were not as dramatic, but were still of my stuttering.

I called my mother to ask her about it and she said I had been having trouble speaking since I was 5 years old, but they didn't make a big deal of it because I didn't seem to mind.

I only started to mind it when I was around 10 years old, and that's why I used to think it began at that age. But I think it's impossible to tell if it became worse and it got me more self-conscious about it, or if it became worse *because* I got more self-conscious about it.

Anonymous said...

Stuttering Brain?

It is easy to see why Judy Kuster put you on the "DO NOT REPOND TO" email she sent out to the "Ask the Prof" group during this ISAD Conference.

Tom Weidig said...

@Peter:

Yes, it would be fuzzy thinking to conclude after one observation.

The issue is whether this correlation happened by chance or not.

The more often you see this correlation the less likely it is random.

So once you tell me that it happened several times, i.e. the correlation stuttering-stress, it becomes more likely that it's a real correlation.

It's also important that stress seems to modulate stuttering rather than cause stuttering.

Peter Louw said...

Tom, I think you are trying to defend the indefensible. The indications that stuttering is TO SOME EXTENT linked to some form of localised tension and/or general stress are overwhelming. Yes, there are many PWS who stutter while relaxed, but that can be ascribed to the effect of learning and conditioning. Many, many PWSs and case histories attest to this link, and to just say that they are all deluded is rather ... well, unsatisfactory. When a PWS says that stress was a factor in him beginning to stutter it should not be dismissed out of hand as you have done. Just because a PWS is not a therapist does not mean that he doesn't know anything about his own disorder. Each case should be checked as carefully as possible to establish whether stress may have been a trigger in him beginning to stutter.

I agree that stress / tension, in any of its many forms, does not in itself cause stuttering, but it is definitely a factor in many cases if not in all. It's the elephant in the stuttering room ...

Tom Weidig said...

But I am not defending it????

I am just saying that observing it once is not enough. You need to see the correlation several time across several people to exclude a spurious correlation.

Again, if you are right that stress is modulating stuttering, then it's clear that stuttering will first appear when under stress. But then I would not really take about a cause or even a clear trigger.

It's like a dysfuntional car engine. You will only notice the imperfection when the engine is put under stress, e.g. heat or top performance.

Working on stress will certainly help pws even in the cases where they still stutter the same, at least they do it without stress.

Tom Weidig said...

But I am not defending it????

I am just saying that observing it once is not enough. You need to see the correlation several time across several people to exclude a spurious correlation.

Again, if you are right that stress is modulating stuttering, then it's clear that stuttering will first appear when under stress. But then I would not really take about a cause or even a clear trigger.

It's like a dysfuntional car engine. You will only notice the imperfection when the engine is put under stress, e.g. heat or top performance.

But the heat or top performance has nothing to do with the cause. You cannot even talk about a trigger. If the engine blows up, you could talk about them being a trigger. But if it just makes a strange noise or underperforms, is not really a trigger?

Working on stress will certainly help pws even in the cases where they still stutter the same, at least they do it without stress.

purity12lover said...

If you think that more people should know about Purity12 then be part of the network marketing opportunity they offer. It's a good way to spread the health and the wealth that Purity12 is known for. http://if1s.com?114

Anonymous said...

different people who stutter have different causes...what is stuttering? What is not stuttering?

What happens if what we think is stuttering is not really stuttering, then we are in Big trouble.

have to compare apples to apples, if you think an orange is an apple, and give the definition of an apple based on your experiments on an orange and your observations of it, then everything is trash....and need to start over

Cesar said...

Wow, that's alot of philosophy just to refute someone's interpretation of their reasons for stuttering! Everyone is entitled to their own opinion, especially one thatrelates to their disability. Let's not be overly-sensitive and believe that every comment made regarding stuttering has to be philosophically concrete and scientifically verified or forever silenced.