Does how we hear ourselves affect how we talk? Shira Katseff, a post-doctoral fellow of the New Zealand Institute for Language Brain and Behaviour at UC, blogs about her current work.
Talking is like tightrope walking. At first, getting your lips and tongue in position to make the right sounds is really hard (think about how difficult it is to understand a 2-year-old!), but eventually you figure out what all the sounds of your native language are, and how to adjust your articulators to make them. Well, almost. Speaking in the real world is like tightrope walking on a rope that keeps changing its width. If you don’t adjust the motor commands that you send to your jaw and tongue while wearing a dental appliance or while chewing on a pen cap, your speech will sound like gibberish. This is why you, and all other speakers, are constantly listening to yourselves. You monitor feedback from your voice and from your jaw and tongue to make sure you sound the way you expect to be heard.
In our lab, we research how we use feedback by changing it experimentally. We do this with a specialised headset. To use it, you speak into the microphone; your voice passes from the microphone through a computer; the computer makes a subtle change to your vowel sounds; and then your altered voice is played back to the earphones. This means you hear your own voice in real time. The speech alteration process happens so fast that you don’t hear any delay between your speech and your altered feedback.
What happens to speakers whose auditory feedback is altered? It turns out that, as their feedback changes, so does their speech. If their voice is shifted to a higher pitch, they will start speaking with a lower pitch so that their voice sounds more normal to them. If they hear the vowel in ‘head’ as ‘had’, they’ll change their vowel production to make themselves hear something more like the expected vowel. This tells us that we’ve learned some sophisticated techniques for adjusting our native language sounds. The adjustments vary somewhat from person to person and language to language, but they’re measurable in any group of people.
Our latest experiment asks how auditory feedback is used by language learners who haven’t yet mastered their second language. To make things more interesting, we gave them a vowel feedback change that should make them say a more native-like vowel. This is worth doing because language learners are special– they don’t know what their vowels are supposed to sound like. For them, speaking is like trying to make a certain pose on the tightrope, without being sure what the pose should look like. So what happens when learners are exposed to altered feedback: can they respond to small changes to their “pose”, or are they too concerned with their “balance” to notice?
This experiment is currently underway. We’ll soon know whether even learners know how to adjust their language poses… and if learners can sound more like native speakers simply by altering the way they hear themselves.
To find out more about Shira’s work, visit: http://www.nzilbb.canterbury.ac.nz/katseff.shtml
To find out more about the work of the NZILBB, visit: http://www.nzilbb.canterbury.ac.nz/