Validating a second language perception model for classroom context - a longitudinal study within the perceptual assimilation model

Abstract

Abstract The present study verified whether adult listeners retain the ability to improve non-native speech perception and if it can be significantly enhanced in the formal context, a very impoverished context with respect to the natural one. We tested (i) whether perceptual learning is possible for adults in a classroom context during focused phonetic lessons, and (ii) whether it follows the pattern predicted for natural acquisition by the PAM-L2 [1]. The results showed that adult listeners are still able to improve foreign sound perception and this ability seems to occur also in formal contexts in line with the PAM-L2 predictions. Index Terms : non-native phone perception, foreign language acquisition, PAM. 1. Introduction The dominant theories in second language (L2) phoneme acquisition argue that perceptual similarity/dissimilarity between the sounds of L2 and native language (L1) governs their assimilation or non-assimilation and dictates their learnability in adulthood. According to the Perceptual Assimilation Model (PAM) [2], subjects with learning an L2 in classroom context (FLA), most of them limited L2 instruction, especially that typified by classroom-only education with instructors with a strong L1 accent, can be considered as naive listeners, i.e., functional monolinguals not actively learning or using an L2. The PAM assumes that the way naive listeners assimilate non -native phones to native phonemes is determined by the detection of commonalities between them [3]. The PAM predicts that the non-native phones can be Categorized or not consistently categorised (i.e., Uncategorized) as exemplars of native phonemes, falling between two or more L1 phonemes. Finally, non-native phones cannot be categorised at all as speech sounds. If two Categorized non-native phones are perceived as acceptable exemplars of two different native phonemes, a very good/excellent discrimination is predicted (Two Category assimilation, TC). Conversely, poor discrimination is expected for Single Category (SC) assimilation, where two non-native phones are perceived as equally good or poor exemplars of a single native phoneme. If two non-native phones are both perceived as a single native phoneme but differ in rating, intermediate degree of discrimination is predicted (Category Goodness assimilation; CG). For the Uncategorized phones, if one non-native phone is perceived as a native phoneme and the other is perceived as an uncategorised speech sound, the predicted degree of discrimination is good (Uncategorised-Categorised assimilation, UC). Two non-native phones assimilated to partially-similar native phonemes will be discriminated from poor to moderate (Uncategorised-Uncategorised assimilation; UU). Finally, the predicted discrimination is good/excellent for the Non assimilable typology (NA) since two non-native phones are not perceived as any speech sound and are easily distinguishable from each other. A recent extension of this model, i.e., the PAM-L2 [1], refers to adult L2 learners in an L2 immersion context, for which a common L1-L2 system (i.e., an interlanguage) is developing. The L2 perceptual learning seems to be determined by the increase in L2 vocabulary size that causes the learners to