Home > Stuttering.Microblog > Valid stuttering research requires a proper review of the stuttering literature

Valid stuttering research requires a proper review of the stuttering literature

Here’s an example of a poor research question that results in data that is both meaningless an uninterpretable.  And yet, it still gets published.  Since one of my sub-themes for this site is to empower the stuttering community by enhancing our research-consuming sophistication, let’s look at why…

The study is trying to measure one’s speech rate as a function of their overt stuttering frequency.  So they had people of different overt severities talk, and they try to measure speech rate.  Well–the concept of speech-rate is a tar-baby in and of itself; are you measuring X number of syllables in Y amount of time?  If so, that’s short-sided.  What if there are just longer spaces between words or syllables?  Or–what if each syllable is longer in duration?  The concept of speech rate needs to be better defined.  Articulatory rate may be a better gamble…

But here’s the real flaw.  Of course, the very act of stuttering adds time; so any added time (i.e., moments of stuttering) would show up as reduced speech rate.  (Duh!)  So the authors decided to throw out any speech tokens in which they heard stuttering.  Great.  Yippee.  What’s the fatal flaw?  They’re assuming that stuttering is only an audible phenomenon.  That if stuttering isn’t heard, then it’s not happening.  Wrong, Wrong, Wrong.

There was a nifty study (Armson & Kalinowski, 1994) that demonstrates sub-perceptual stuttering moments.  Moments of perceptual “fluency” were analyzed, and stuttering was still found to occur on the physiological level (as measured by any numer of physiological measures).  Subperceptual stuttering exists, people.

So this latest publication is yet-another example of how poor theoretical thinking results in silly research questions and wholly invalid data.  The authors apparently think that stuttering is (only) an audible phenomenon; and if it’s not overtly heard, then it must not exist.  So the use of perceptually fluent speech tokens is OK.  So they find, shockingly, that as overt severity increases, speech rate decreases.  They’re totally ignorant that their data is likely being contaminated by subperceptual stuttering–the stuttering that remains in existence even though their ears can’t detect it.

The authors of this paper are likely people who fluent; had they the sense to ask a stutterer (and actually listen to the response) they would have known better…

Be Sociable, Share!
Categories: Stuttering.Microblog Tags:
  1. Bond
    April 12th, 2009 at 01:13 | #1

    Hi, so you suggest “time out” for kids who stutter?


    Greg @ Stuttering.me Reply:

    Well–it’s tough to say for all cases. But yes, I have suggested the use of ‘time-outs’ (done in an appropriate fashion) in the past.

    To be honest, I think that any number of treatment approaches will work, because they’re all likely equally flawed. So why not look at the big picture and try to the most appropriate approach for each particular case?


  2. A stutterer…
    April 13th, 2009 at 07:42 | #2

    (An anonymous coward w/ the IP posted this comment. Still deciding whether or not we’ll accept anonymous comments.)

    Time outs for young children who stutter? Great way to send the message to just shut up kid.


    Greg @ Stuttering.me Reply:

    Well–again, if it’s done in an improper way–it’s not good. But it doesn’t have to be that way; when made into a fun game. But it doesn’t sound like you care to think outside of your opinion, so I won’t waste my time on an explanation.


  1. No trackbacks yet.