Tuesday, March 03, 2009

Why look at stuttered speech?



I agree with Greg that studying stuttered speech is more or less time wasted! See here. It is a good example of what happens when people use quantitative tools without having any scientific common sense or vision. And I say it again, if you don't have a hard science background, do not work on quant stuff, you will fail. I would have tolerated such research decades ago, but not any more...

I mean just imagine we were to study the content of children pornography to find out what causes paedophilia. Yes, you can find a lot of "interesting" and intriguing patterns: what age, how many, what, who does what, and so on. But at the end of the day, it is sexual drive gone to the extreme crossing moral boundaries and hurting others, and crystallizing into behaviour depending on the environment, opportunity of the individual and culture...

Think of the drive as the water flowing and the forms into which water flows and freezes as the environment. Surely, studying the forms is not telling you anything about where the water came from?

7 comments:

Greg said...

Thanks for the link Tom. There's actually a few articles that make this argument in a very, very strong fashion. (From the early to mid '90s.) I can email you a few, if you'd like.

Armson & Kalinowski, 1994--in particular--goes to great length in detailing why the 'fluent speech paradigm' and other measures of stuttered speech provide uninterpretable data.

...And yet, the field has yet to significantly change... *smacks forehead* I've come to the conclusion that most of SLP is actually what R Feynman would call "Cargo Cult Science". By mimicking the scientific process, they think they're conducting science. But they're not, they're just mimicking the process!

Greg
http://stuttering.me

Anonymous said...

can you define hard science? what do you mean?

hard science must be the opposite of soft science?

Soft Scientist: will they admit they are soft?

Hard Science: Physics, Quantum mechanics, string theory???

So stuttering is fuzzy? How do you measure stuttering severity anyway?

ig88sir said...

All the speech programs I've participated in measure the audible vocal output yet I feel I am in my stuttering "state" even when not talking and someone walks in the room. I also feel articulator tension and heat while not talking. Is this stuttering? Defiantly!
Low white matter fibre density in the left hemisphere and a defective Basil ganglia and Striatum is probably what stuttering is. Please correct me if I'm wrong. I'm defiantly not an expert. Take a look at all the stuttering abstracts at NCBI:

http://www.ncbi.nlm.nih.gov/

Tom Weidig said...

Hard Science is where you are using quantitative methods to describe systems: physics and mathematics, but also engineering, chemistry, computing, genetics. So they are using quantitative methods all the time.

Soft science is biology, psychology, sociology, and so on. And they only occasionally use quant methods. Another difference is that in hard science you have constant feedback on the quality of your thinking in soft sciences you don't. Does your computer make you stupid?

Anonymous said...

do you have a bias against soft scientists?

There are plenty of good Scientists in Psychology?

Who are the good Scientists in SLP, especially Stuttering research?

Anyone????

Anonymous said...

do you have a bias against soft scientists?

There are plenty of good Scientists in Psychology?

Who are the good Scientists in SLP, especially Stuttering research?

Anyone????

Greg said...

Allow me to try and explain. When Tom (or some high-energy physicist) is working on smashing atoms together to see what squirts out, the atoms don't really care. They are not conscious. They simply react to the forces and variables that Tom (or other hard scientists) throw at them.

Behavioral 'sciences' like psychology and SLP have it much more difficult. The atom doesn't care; it doesn't think; it just is. But people are much, much different. People have free will; people have choice; they react and respond. Sometimes the body responds and/or compensates without conscious thought. This is a *huge* threat to internal validity. Is the treatment working or is the body compensating differently?

As a result, the behavioral 'sciences' are really cargo-cult sciences. We try and follow the process. But process alone isn't enough. We have to be able to control that which we cannot control; and account that for which we cannot account. Human responses and human reaction, coupled with the body's ability for attempted compensatory self-corrections.

Stuttering research is more or less cargo-cult science. My own publications included. It's just the nature of the beast.