Abstract
This paper investigates the perception of English voiceless plosives (/p/, /t/, /k/) by EFL learners from Cantonese, Mandarin, and Changsha dialect backgrounds, addressing gaps in existing research. We employed a rigorous experimental design involving 60 participants who underwent perception tests using audio materials generated by both native speakers and artificial intelligence. The results revealed significant differences in perception patterns among the three groups, with Cantonese speakers exhibiting the most pronounced transfer effects. Mandarin and Changsha dialect speakers showed relatively similar perception patterns, suggesting the absence of final consonants may lead to uniform transfer effects. This study highlights the importance of considering phonetic transfer in language teaching, offering insights for developing targeted teaching strategies to improve English education quality in China. Our findings contribute to the theoretical understanding of language transfer and have practical implications for second language acquisition, suggesting future research should include broader linguistic contexts and longitudinal studies.
Introduction
In second language acquisition, "language transfer" describes the phenomenon where a learner's native language (L1) influences their acquisition of a target language (L2). This influence can be positive (facilitating learning) or negative (interfering with learning). For Chinese learners of English as a Foreign Language (EFL), the English phonetic system presents unique challenges, particularly regarding voiceless plosives (/p/, /t/, /k/) in various word positions.
Previous research has often been limited to theoretical comparisons or single-dialect studies. This paper addresses this gap by empirically comparing three distinct L1 backgrounds: Cantonese, which features "checked tones" ending in -p, -t, and -k, and the Mandarin and Changsha dialects, which generally lack syllable-final plosives. The goal is to quantify how these structural L1 differences impact the perception of English sounds.
Methodology
The study utilized a quantitative experimental design to test phonetic perception among 60 undergraduate non-English majors. Participants were divided evenly into three groups based on their native mother tongue: Cantonese, Mandarin, and Changsha dialect.
Stimuli and Task
The researchers developed 18 English nonsense syllables to test plosives (/p/, /t/, /k/) in three specific positions:
- Beginning (CV): e.g., pae
- Middle (VCV): e.g., api
- End (VC): e.g., eip
To ensure acoustic variety, each syllable was recorded by two sources: a human (an 11-year-old Australian male) and an AI text-to-speech module. Participants listened to 36 audio clips and were given 5 seconds per clip to identify the correct syllable from a multiple-choice list of six options.
Results
The experimental data revealed distinct performance patterns based on the participants' native dialects. Statistical analysis showed that structural differences in L1 directly impacted listening comprehension.
Overall Performance
As shown in the descriptive statistics, Changsha dialect speakers achieved the highest overall mean score (32.00 out of 36), performing particularly well on /p/ and /k/ sounds. Conversely, Cantonese speakers recorded the lowest overall scores (mean: 30.50), though they outperformed other groups specifically on the /t/ sound. T-tests confirmed significant differences between Cantonese and Changsha speakers in perceiving "k+vowel" sounds, and between Cantonese and Mandarin speakers regarding "vowel+p+vowel" and "vowel+k+vowel" structures.
Positional Accuracy
Analysis of word positions indicated that participants were generally more accurate at perceiving consonants at the beginning of a word (CV format) compared to the middle (VCV) or end (VC) positions.
Discussion
The findings suggest that the presence of syllable-final plosives in Cantonese creates a distinct transfer profile compared to Mandarin and Changsha dialects. Interestingly, the most common error across all groups was the confusion of voiceless ("soundless") plosives with their voiced counterparts (e.g., hearing /p/ as /b/).
Furthermore, the relatively narrow performance gap between Mandarin and Changsha speakers may be attributed to the "Mandarin Effect." The aggressive promotion of standard Mandarin in China means that young adults are often highly proficient in Mandarin regardless of their local dialect, which may be homogenizing their phonetic perception baselines.
Conclusion
While this study provides valuable insights, it is limited by its sample size and the advanced English proficiency of the participants, which may mask stronger transfer effects present in earlier learning stages. Future research should expand to include younger learners and investigate oral production (speaking) in addition to perception. Understanding these dialect-driven differences is crucial for developing targeted phonetic training in second-language education, moving toward a more nuanced, dialect-aware approach to English instruction in China.
Summary updated