Sunday, June 22, 2014

Prompt (haptic) pronunciation prompting: Why does that work?

This post is an edited repost of a piece I posted on a list-serve recently, "The dark side of real-time, spontaneous pronunciation correction & feedback." Haptic anchoring of pronunciation change, where we have learners move and speak along with us, would probably be technically termed a type of "prompting" (Lyster and Saito, 2010). The question posed in the post is: How can you know when or why feedback works, based on research studies where what students knew or had been taught before the feedback event is not adequately specified? (In AH-EPS haptic pronunciation work, which emphasizes in-class spontaneous feedback, that connection is fundamental.) 
Have been working through the various studies and meta-analyses of corrective feedback in pronunciation teaching, (e.g., Lyster and Saito, 2010) looking for one feature: What did the students know about the feature in focus before the intervention and when did they learn about it? In other words, if learners have been introduced to the vowels system in some way, we probably will assume that in class correction or feedback on a specific vowel (or even out of class self correction) has a better chance of working. 
In all of the studies I have reviewed so far that investigate the range of feedback mechanisms, both in lab settings and in classrooms, I can find almost nothing that adequately characterizes the assumed cognitive schemata or understanding of the learners relative to that phonological feature prior to receiving some kind of feedback. It is occasionally referenced in passing, for example, that students, “had been introduced to X earlier, etc.,” but never in any systematic analysis.  The irony of contemporary theory giving such credence, prima facie, to “behaviours” without reference to what learners may be bringing to the party, must be enough to make the ghosts of Skinner, Lado and friends smile.

Haptic pronunciation work is based on prior, systematic introduction of specific features of the sound system before classroom intervention.  (Some of that is very much “Gilbertesque” in nature.) 

Do you know of an accessible study of classroom correction/feedback where the description of the formal pedagogical approach in place prior to the classroom “event” (or just what students are assumed to have known about the target) is sufficiently detailed and explicit so that the connection between formal (or even informal) presentation/knowledge – intervention--and effect or change is at least traceable? 
I didn't think so . . .  


Angelina Van Dyke said...

Good question - for my classes I focus on the intonation on focus vowels in each tone group - that give the vowel import in the breath and places the consonants in rhythmic order. That's why AH-EPS works!

Tom Tabaczynski said...

I'm inclined to view us moving into the 'language emergence' paradigm as the successor to cognitivism, and whereas cognitivism made corrective feedback a more complicated matter, it seems even worse in the language emergence paradigm.

To me the notion of corrective feedback always smacks of behaviourism. I've been reading the newly published Brown et al "Make It Stick: The Science of Successful Learning" which describes research in favour of spaced repetition and retrieval practice. Spaced low stakes testing provides a kind of feedback, but I think that's different from what is meant by "corrective feedback".

Today I had to point out to the Japanese student that he's pronouncing "Atlantic" with vowel #6 instead of #5. I would tend to put that in the context of a spaced quizzing model, in that it seems that it is the effort to produce the right answer (eg., vowel) with feedback as in a quiz (not immediate), on regular spaced-out occasions that produce lasting improvement.

Rod Ellis pointed out that the danger of correcting is that there may be psycho-affective consequences whereby the student may cease to learn anything. I think that outside of a system of 'buckets' whereby the information can be filed and systematically retrieved, correction simply leads to cognitive overload. And for that it seems that you need the explicit knowledge to systematise your learning.

Bill Acton said...

Good point, Tom, but let me "point out" something. You note that you "pointed out" a vowel issue to a student. How did you do that? Did you simply say, "Hey, X that is v5, not v6?" Did you have the student do the PMP w/you 3x? Did you "order" that word be noted and entered into the student's current practice word list or pronunciation diary? Did you offer another v5 word to associate it with? I assume that you had earlier done the AH-EPS rough vowel module in some form earlier and that students had been through in some form the related section in Gilbert's Clear Speech. I personally find Ellis' warning on "psycho-affective consequences" overblown. Granted, if you do not have a good professional, interpersonal working relationship with your students, nearly any aspect of the teaching experience gets caught up in the "affective filter." Haptic engagement done well should by pass at least that aspect of the experience of "correction." (In part that is because the warm ups and video mirroring defuse it.) Very much like the "spaced out" characterization, from a couple of perspectives!

Tom Tabaczynski said...

You're right Bill. The answer is 'yes' for most of that. The psycho-affective consequences is an interesting question. I train Capoeira. I started with a new group recently. There is a thing I find in some places that some people are just dying to 'teach' and establish dominance. They don't bother to try to establish a relationship. Even guys I'd otherwise respect and accept feedback from have a tendency to start off with lengthy explanations which apparently inevitably are a complete turn off ... ie., 'psycho-affective consequnces'.

There are in fact sound arguments from cognitivist research as to why movement teachers (in dance, etc.) should reduce the amount of verbal feedback which tends to create cognitive overload (eg., Journal of Dance Education).

So I think that your point about about a professional working relationship is absolutely correct. But what that effectively would mean is that the professionally involved teacher will eventually start to notice that certain kinds of feedback cause cognitive overload and affective filter. That teacher will then 'proactively' anticipate the problems by construcing a pre-emptive syllabus, as contrasted with 'reactive' feedback.

The procedure that you describe is precisely what I would call a system of 'buckets' or 'filing' whereby the student is provided with a way of coping with the information without overload. This idea actually comes from the 'Getting Things Done' scheme which has been argued to be in line with neuroscience research that the brain needs these 'external tools' for coping with information, known as the 'extended mind' metaphor. So then I guess it's a semantic issue whether we still call that 'corrective feedback' or 'correction'.

Post a Comment