Monday, December 05, 2005

Not much ado about hearing?

There is a discussion going on in the PDS literature on whether deficiencies or malfunctioning in the auditory cortex (responsible for hearing) causes or aggravates stuttering, or whether it hinders therapy. For example, auditory feedback frequency-shifted (FAF) or delayed (DAD) enhance fluency. And in brain imaging studies, the auditory cortex shows a different (lower) activation that for controls. And check out my last post, where I have described my difficulty to realise that my speech has been much more fluent after therapy most times.

These phenomena seem to point to a crucial role for the auditory cortex. But I dont really believe that hearing has much to do with the PDS. In my opinion, the FAF and DAF enhances fluency, because the brain switches to non-automatic speech due to input that sounds unfamiliar. Ingham and others have explained the brain imaging results for the auditory cortex using the concept of efference copy: see here. But, there is certainly truth in the fact that people with PDS often dont notice the progress they have made. I do not believe, that this is a defficiency of the auditory cortex, rather the consequence of a normal functioning. The stuttering is just left out after years of stuttering as a background noise. And the absence of no stuttering sound is just not registered.

I could be wrong and I am open to be converted. This new article out in Neurology Abnormal speech sound representation in persistent developmental stuttering does not really convert me but have a look at their abstract:

OBJECTIVES: To determine whether adults with persistent developmental stuttering (PDS) have auditory perceptual deficits. METHODS: The authors compared the mismatch negativity (MMN) event-related brain potential elicited to simple tone (frequency and duration) and phonetic contrasts in a sample of PDS subjects with that recorded in a sample of paired fluent control subjects. RESULTS: Subjects with developmental stuttering had normal MMN to simple tone contrasts but a significant supratemporal left-lateralized enhancement of this electrophysiologic response to phonetic contrasts. In addition, the enhanced MMN correlated positively with speech disfluency as self-rated by the subjects. CONCLUSIONS: Individuals with persistent developmental stuttering have abnormal permanent traces for speech sounds, and their abnormal speech sound representation may underlie their speech disorder. The link between abnormal speech neural traces of the auditory cortex and speech disfluency supports the relevance of speech perception mechanisms to speech production.

In plain English, the authors have looked at the following phenomena: If you play the same sound to someone several times and then change the sound either for duration, frequency, or related phonemes (the sound units that make up our spoken language), a different brain signal is recorded. This signal is the brain detecting that something has changed. But of course you can only know that something changed if you know what has been before! So the brain must have a representation (or memory) of the preceding sound with which it can compare the current sound. And the authors claim that the signals for complicated sound changes is different to people without PDS. They speculate that the PDS brain does not have good representation of sounds amd is therefore not as efficient in quickly discriminating between more subtle sound changes, and this could encourage development of stuttering.

I had a discussion with Per Alm about their paper, and will post some comments in my next post.


Anonymous said...

Hi Tom

The authors use the word 'enhanced' so we don't seem to be talking of a deficit here!

Also, I worry about their speculating that this may 'underly the speech disorder' when we only have an association. If we assume that speech is a concern to stammerers then an unconsciously enhanced response to phonemes might not be too much of a surprise. It might be interesting, and a better control, if the authors took folk with a spider or other phobia and used phonemes associated with this state and found whether there was an 'abnormal' 'enhanced' response!

Relating to the blog about your own self-monitoring, can their correlation with self-assessment stand much scrutiny?


Tom Weidig said...

I agree. There is only a different association to a changed sound as compared to the normal population.

I guess this different association is a reaction to the stuttered speech.

However, this effect could still be part of adult PDS, as it might make it more difficult for people with PDS to become more fluent.

Well, I do think that there is a correlation between self-assessed stuttering and objective dysfluency. But the link is not very strong. I did this kind of analysis with the data of the Kassel Stuttering Therapy, and got about a correlation of 0.4. As a measure stuttered syllables are probably better.