| Language transfer is a common phenomenon in the process of foreign language learning.In terms of the direction of transfer,previous studies have focused on the transfer from the native language(NL)or L1 to foreign language(FL)or L2,which is known as “forward transfer”.In contrast,“backward transfer”,which refers to the transfer from FL or L2 to the NL or L1,has not received sufficient attention.“Language attrition” refers to the phenomenon of a certain degree of degradation of the NL during the process of learning a foreign language.Generally speaking,this is a long-term process,and is related to “backward transfer”.This study focused on changes related to phonetic perception.Unlike “language attrition”,the main focus of this study was to investigate whether there was a change in NL perception in the short-term foreign language learning process,which is known as phonetic drift.Spanish was chosen as the target language,and phonetic training was used to simulate the initial stages of foreign language learning process in the current study.It aimed to investigate whether the subjects’ perception of Chinese consonants changed after training,and if so,which phoneme(s)changed.The study also examined changes in the categorical perception of the phoneme /w/,which was markedly decreased in perception after training.This study sought to answer the following questions:(1)Will the perception of native consonant segments changes after a short-term foreign language learning process?(2)Is there any change in native Chinese speakers’ categorical perception of native language consonant(s)during foreign language learning process in behavioral experiments? If yes,what changes occur,and to what degree?(3)Is there any neurocognitive change in learners’ perception of the NL consonant categories? If yes,are changes observed from a neurocognitive perspective in the categorical perception of native language consonant(s)consistent with the behavioral experiments?Two series of experiments were conducted in this study.Experiment A recruited 10 subjects to observe changes in the perception of each consonant after training through the Chinese consonant identification task under noise condition,and determined /w/,which had a significant degradation in perception,as the research object of subsequent experiment.Experiment B recruited 21 participants,and in addition to the same tasks as Experiment A,it used a synthesized /ba/-/wa/ continuum as materials to investigate changes of categorical perception of /w/.Additionally,Experiment B also used Oddball paradigm to elicit mismatch negativity(MMN)to investigate the changes from neurocognitive aspects.The main findings of this study were as follows:(1)After four days of Spanish consonant training,the subjects’ perception of NL consonants changed,with the most remarkable change observed in the phoneme /w/,though it did not reach the level of significance in Experiment B.However,it was still the most varied among consonants that appeared to degradation in perception.(2)From behavioural perspective,the categorical perception of /w/ had changed after training.Formant rising time(FRT)is the perceptual cue of /b/ and /w/,and this study used FRT as the marker to observe the changes in categorical perception of /w/.The FRT at the categorical boundary position changed from 44.93 ms to 48.17 ms after training and reached significance level.But the boundary width did not change significantly,from 15.51 ms to 14.13 ms.(3)From the perspective of event-related potential,the average amplitude of MMN elicited by stimuli within-,across-,or near-category stimuli in Experiment B did not change significantly during the pre-and post-tests respectively.This result is understandable,as the relative perception distance of both the standard stimulus and deviation stimuli does not change after training,making it difficult to observe changes in the average amplitude of the MMN.The above results showed that short-term foreign language training does have an impact on NL consonant perception.From a behavioral point of view,changes in the categorical perception of specific consonants could be observed.However,from a neurocognitive point of view,given that no changes in the average amplitude of the MMN were observed,future research may need to explore more suitable experimental paradigms for related research. |