What the Macnamara et al. research focused on was the variability associated with excellence in various disciplines or arts. Results varied widely. In a report on a meta-analysis described in "Psychological Science", Macnamara and colleagues note the following:
However, the domain itself seemed to make a difference. Practice accounted for about 26% of individual differences in performance for games, about 21% of individual differences in music, and about 18% of individual differences in sports. But it only accounted for about 4% of individual differences in education and less than 1% of individual differences in performance in professions.
There is obviously a lot going on there, but of particular interest for us is the overall range of "skill areas" sampled. In a very real sense, ALL of those relate to pronunciation proficiency, in part due to the relative degree of physical and cognitive involvement required, especially for adult-age learners. My guess is that pronunciation probably falls somewhere in the middle, around 10 to 15%.
So, if that is the case, what would that mean for instruction? One obvious question is how much practice is effective at different stages of the acquisition process. A new study getting underway here, which will be reported on in a Panel presentation on the role of homework in pronunciation teaching, at the TESOL convention in Seattle next March 27th, will address that question.
Some preliminary interview work with a broad slice of learners about their pronunciation practice suggests that something like the 26-21-18-4-1 ratios may actually map on to beginning through highly advanced L2 phonological proficiency and "accent retention".
In other words, as learners improve, the demand for pronunciation practice diminishes accordingly. That, of course, makes perfect sense--as long as the "bottom" is addressed. Without the 26-21-18 in the early stages--which entails significant degree of body or physical engagement--learning the sound system to "intelligibility" level can be seriously compromised for many learners.
When the "education" approach is taken from the outset, with its resulting 4% variance--and its generally strong cognitive vs physical practice approach to pronunciation--little wonder some conclude that practice (primarily insight, plus aural comprehension and oral drills) often does not appear to make much difference.
Reminds me of Tom Scovel's wonderful tongue-in-cheek definition of an "expert": "ex-" (former, "has been" out of touch) plus "spurt" (gush out forcefully but be gone quickly)
See you in Seattle, if not before!
Original source reference:
B. N. Macnamara, D. Moreau, D. Z. Hambrick. The Relationship Between Deliberate Practice and Performance in Sports: A Meta-Analysis. Perspectives on Psychological Science, 2016; 11 (3): 333 DOI: 10.1177/174569