From meaning to sound: how word learning shapes non-native speech perception

Area of Science:

• Second Language Acquisition
• Phonetics and Phonology
• Cognitive Science

Background:

• Adult learners face difficulties perceiving and learning new speech sounds in a second language, particularly at early stages.
• Traditional perceptual training methods are often ineffective for beginners due to their reliance on low-level acoustic cues and lack of context.

Purpose of the Study:

• To investigate the effectiveness of cross-situational word learning (CSWL), a meaning-based approach, in enhancing the perception and discrimination of non-native vowel contrasts in adult learners.
• To explore the relationship between word learning and sound perception in second language acquisition.

Main Methods:

• Thirty-seven Hungarian native speakers participated in a single cross-situational word learning (CSWL) session using European Portuguese pseudowords.
• Training involved alternating passive and active learning blocks with feedback and exposure to multiple native voices.
• Participants completed pre- and post-training identification and discrimination tasks to assess word learning and vowel discrimination.

Main Results:

• Participants demonstrated successful lexical learning, evidenced by above-chance word identification.
• Vowel discrimination improvement was contrast-specific, with gains observed in three out of six contrasts.
• Performance on the most challenging vowel contrast remained low, impacting pseudoword identification accuracy.

Conclusions:

• Meaning-based training via CSWL can facilitate early lexical and phonological learning in second language acquisition.
• Perceptual challenges persist for difficult non-native speech contrasts, even with effective word learning.
• Lexically grounded approaches show promise for perceptual training at the beginning stages of language learning.