Third International Conference on Spoken Language Processing (ICSLP 94)
The present study examined the usage of spectral and temporal acoustical features in the production of the phonemic contrast between the [r] and  sounds in the prevocalic position by Japanese speakers with different perceptual skills in English. The acoustical analysis of recorded tokens demonstrated that Japanese speakers, who perceptually were able to integrate information carried by the spectral and temporal acoustical cues, were also able to use both of those cues in their productive distinction between the [r] and  sounds. A lack of ability to integrate the temporal and spectral cues of the perceptual task demonstrated by the second group of Japanese listeners was reflected in their poor productive distinction between the [r] and  sounds.
Bibliographic reference. Slawinski, Elzbieta B. (1994): "Perceptual and productive distinction between the English [r] and [l] in prevocalic position by English and Japanese speakers", In ICSLP-1994, 2031-2034.