Effects of native language on perception and neurophysiologic processing of English /r/ and /l/ by Native American, Korean and Japanese listeners

The perception of English liquids /r/ and /l/ is challenging for native Korean and Japanese adult speakers because these sounds are not phonemic in these languages. The Korean language has a partial phonetic model that could potentially facilitate processing of English /r/ and /l/ but the Japanese language does not. The purpose of this study was to compare the effects of native language on the neurophysiologic processing of English intervocalic /r/ and /l/ by native American, Korean and Japanese listeners using several event-related evoked potentials (ACC, MMN, & P3a) along with behavioral identification and discrimination. Three specific aims were investigated. The first aim was to examine the effects of native language on the perceptual identification and discrimination of English intervocalic /r/ and /l/. The second aim was to determine the effects of native language on the neurophysiologic encoding of English intervocalic /r/ and /l/ using the acoustic change complex (ACC). The third aim was to determine the effects of native language on the pre-attentive discrimination of and related attention shifting/orienting to English intervocalic /r/ and /l/ using the mismatch negativity (MMN) and P3a.;Stimuli were a synthetic vowel-consonant-vowel (VCV) continuum that generated percepts in American English listeners ranging from /iri/ to /ili/. Stimuli falling within- and across-phonetic category were presented using an oddball paradigm. The probability of occurrence of the deviant was 20%. Nine participants from each language group participated. Stimuli were presented via insert earphones at 70 dB SPL using an 1100 ms offset-to-onset interstimulus interval. The evoked potentials were recorded from surface electrodes using a Neuroscan system. Behavioral testing included a 2-alternative forced choice identification task and a 3-alternative forced choice oddity discrimination task.;English medial /r/ and /l/ were perceived in a categorical manner by Americans, in a categorical-like manner by Koreans and in a non-categorical manner by Japanese. The midline-central ACC P1-N1-P2 responses did not differ significantly between language groups, suggesting little effect of native language on the primary cortical encoding of these sounds. In contrast, the lateral-temporal ACC T-complex differed across language groups, suggesting that native language influences secondary cortical processing of these sounds.;The MMN also depended on native language, suggesting that native language influences automatic, pre-attentive discrimination of English medial /r/ and /l/. Both early (400-650 ms) and late (655-905 ms) responses were obtained for Americans and Koreans, whereas early responses were absent in Japanese. Early MMN responses were significantly larger for across-category pairs than for within-category pairs only in Americans and Koreans. Late MMN responses were significantly larger for across-category pairs than for within-category pairs only in Americans. Additionally, late MMN responses for the across category pairs were significantly larger in Americans compared to other language groups. The P3a had both early (600-700 ms) and late (900-1000 ms) responses, similar to MMN responses. Early P3a responses were present in Americans and Koreans and P3a latency for across-category pairs was shorter for Americans than for Koreans. Late P3a responses were obtained from all three language groups, and did not differ significantly across tokens or groups.;The partial language model available to Koreans appears to facilitate both neurophysiologic processing and behavioral perception of English /r/ and /l/. The absence of such a model in Japanese results in perceptual processing difficulties and alterations of the neurophysiologic processing of these sounds. Native language influences the neurophysiologic processing, including encoding at the level of secondary auditory cortex, pre-attentive discrimination, attention-related processing, and behavioral identification and discrimination.