Audiovisual english /r/−/l/ identification training for japanese-speaking adults and children

Purpose: This study tested the hypothesis that audiovisual training benefits children more than it does adults and that it improves Japanese-speaking children’s English /r/−/l/ perception to a native-like level. Method: Ten sessions of audiovisual English /r/−/l/ identification training were conducted for Japanese-speaking adults and children. Assessments were made of age effects on the increase in identification accuracy in three testing conditions (audiovisual, visual only, and audio only) and auditory discrimination of the primary acoustic cue (F3 frequency). Results: The results showed that both adults and children increased their identification accuracy in the audiovisual condition more than in the single-modality conditions (visual only and audio only). Their improvement in the visual-only condition was larger than that in the audio-only condition. Japanese-speaking adults and children improved their primary acoustic cue (F3) sensitivity to a similar extent. In addition, their identification improvement in the audiovisual condition was positively correlated with those in the audio-only and visual-only conditions. The improvement in the audio-only condition was also positively correlated with that in the visual-only condition and with primary acoustic cue sensitivity. Conclusion: It was unclear whether children had an advantage over adults in improving their identification accuracy, but both age groups improved their auditory and visual perception of the English /r/−/l/ contrast and showed additive effects in the multisensory (i.e., audiovisual) condition.