Can acoustic cues used in L1 really be used to perceive novel sound contrasts?

One issue in language acquisition is whether the acoustic cues used in the L1 can be used to perceive a different phonemic contrast in the L2. A previous study found that Japanese listeners can utilize their sensitivity to vowel duration in their L1 to identify the English coda voicing contrast when asked to classify each token as 'bit' or 'bid'. That is, they could use vowel duration in a way comparable to native English listeners when tested with a lexical identification task. The current study tested whether Japanese listeners could still utilize vowel duration to identify the same contrast when using a phoneme identification task: Japanese listeners were asked to categorize the last sound they heard as ‘t’ or ‘d’. Our results indicate that Japanese listeners rely on vowel duration only in the lexical identification task: the familiar acoustic cues are not applied to the novel phonemic contrast.