The Role of Structural Position in L2 Phonological acquisition : Evidence from English learners of Spanish as L2 Nov. 3th. 2008 Student’s Name/ ID ： Yuri 9722616 Betty 9722609 Vivian 9722607 Instructor ： Philip Lin
Vokic, G.(2008). The Role of Structural Position in L2 Phonological acquisition : Evidence from English learners of Spanish as L2. Foreign Language Annals, 41(2). 347-363.
Introduction <ul><li>The background of study ： </li></ul><ul><li>It generally accepted that in the process of L2 acquisition, learners identify certain sound in the L2 with the L1 sounds . </li></ul>
Introduction <ul><li>The assumption ： </li></ul><ul><li>Structural position and thus distributional properties of sounds must be taken into account when determining the relative degree of similarity or dissimilarity between L1 and L2 sounds system. </li></ul>
Introduction <ul><li>The purpose ： </li></ul><ul><li>To contrast phonemic inventories of L1 and L2, and pointing out phonemic correspondences and opposition across L1 and L2. </li></ul>
Interlingual identification equivalence classification <ul><li>One position ： </li></ul><ul><li>Lado (1957), constructive analysis hypothesis, is that sounds similar in L1 and L2 are easier to acquire. Underlying the position that L2 learners will have more difficulty with new sounds. </li></ul>
Interlingual identification equivalence classification <ul><li>The opposing view ： </li></ul><ul><li>Flege's speech learning model predicted that late L2 learners will be less successful in learning similar L2 sounds precisely because they equate L2 sounds with sounds of the L1. </li></ul>
Research Questions <ul><li>Do adult L2 learners relay on distributional information in the process of acquiring L2 phonology? </li></ul>This pilot study represents a first step of the role that the structural position plays in the process of L2 phonological acquisition.
Research Questions <ul><li>Does structural position have any bearing on what is considered by L2 learners to similar to or dissimilar from their L1 sound system? </li></ul>It examined the structural position of a sound in a word has bearing on the way L2 speech is perceived and, in turn, produced .
Parameters for Analysis <ul><li>Structure positions ： </li></ul><ul><li>Word-initial position (W-I) = </li></ul><ul><li>Breath-group initial position </li></ul><ul><li>Word-medially position (W-M) = </li></ul><ul><li>Word-internal syllable onset position </li></ul><ul><li>Word-final position (W-F) = </li></ul><ul><li>Breath-group final position </li></ul>
Parameters for Analysis Hypothesis 1 ： If L1 and L2 share similar phoneme that has the same phonotactic distribution in both L1 and L2, then this situation is regarded as identical and sound in question will not a problem for the L2 learner.
Parameters for Analysis Hypothesis 2 ： If L1 and L2 share an acoustically identical or similar phoneme, but it has different phonotactic properties in L1 and L2, then this situation will represent a problem for acquisition.
Parameters for Analysis Hypothesis 3 ： If the L2 has a phoneme that appears neither at the phonemic nor at the allophonic level in the L1, this will be regarded as different by learner and will represent a problem in the acquisition of the L2 sound system.
Methodology 1. Participants ： 12 participants who were all adult native speakers of Midwestern variety of American English. They were all university students whose Spanish proficiency level was intermediate to intermediate high.
Methodology 2. Stimuli Design ： The following criteria were used in the process of word selection: A The word follows the CVCV separated pattern B The word is disyllabic C The word carries penultimate stress D The target consonant is surrounded by a variety of vowels
Methodology 2. Stimuli Design ： There were three signs per each parameter. This summed to a total of 33 stimuli . In order to make sure the purpose of the study easy-understanding to the subjects, the stimuli were re-shuffled, also, for each stimulus added one additional distracter .
Methodology 3. Data-Collection Procedure ： A. Both the stimuli and the distracters were placed into a normal media sentence , i.e., [I see the word STIMULUS on the screen] (English) Veo la palabra STIMULUS en la pantalla (Spanish)
Methodology 3. Data-Collection Procedure ： B. The participants on a computer screen in a ppt format with a 5-second break between each sentence.
Methodology 3. Data-Collection Procedure ： C. The participants read aloud and at a normal speaking speed, the sentences on the screen contained the stimuli. Each set of stimuli was repeated and recorded 4 continuous times.
Methodology 4. Equipment ： Place <ul><li>Sound-proof closet </li></ul>Equipment <ul><li>A Sennheiser cardioid microphone model e815s </li></ul><ul><li>A SONY DAT (Digital Audio Tape) recorder model TCD-D7 </li></ul>Setting used on the DAT recordings <ul><li>Sampling frequency: 48 KHz </li></ul><ul><li>Microphone sensitivity: high </li></ul><ul><li>Recording mode: manual </li></ul><ul><li>Recording level: 6-7 </li></ul>
Methodology 5. Data Analysis ： Collected data were transferred from the DAT recorder to the computer Redigitized at a 22050 Hz sampling rate using version 4.3.02. of Praat software and saved as a .wav file A mental analysis of the data acquired was then conducted by playing the data from Praat and listening to them using Sennheiser headphones.
Results Hypothesis 1 ： The stimuli used to test by the subjects for the Spanish phoneme /s/ are presented below: sano[s á .no](healthy) solo[s ó .lo](alone) silla[s í .ya](chair) casa[k á.sa] (house) mesa[m é .sa](table) piso[p í .so](floor) gatos[g á .tos](cats) gafas[g á .fas](glasses) comes[k ó .mes] (you(Sg.)eat)
Results Hypothesis 1 ： [s] [z] Average Rates : Word-initial: 100% Word-medial:100% Word-final:96.3% Average Rates : Word-initial: 0% Word-medial: 0% Word-final: 3.7% Terms of frequency The subjects didn’t encounter difficulties in pronouncing the target sound.
Results Hypothesis 2 ： The stimuli and the target sounds are shown below, they were: [ø], [h], and [ h ]: Target sounds Frequency rates reloj [re.l ó h] (watch) 53.8% pedicoj [pe. ð i.k ó h] (one-legged jump) 34.3% relej [re.l é h] (rut) 11.9% Similar phoneme with a different distributional pattern in the L1 and L2 will be difficult to feel .
Results Hypothesis 3 ： The Stimuli used to test this hypothesis and the sounds are shown below: carro [k á .ro] (car) burro [b ú .ro] (donkey) perro [p é .ro] (dog) rata [r á.ta] (rat) rojo [r ó .ha] (red) risa [r í .sa] (laughter) The frequency results point to the fact that a pattern like this one is indeed problematic for the learners.
Conclusion & Discussion 1. It was found that overall native speakers of English were more successful in producing target sounds with overlapping distributional patterns in the L1 and L2 than those target sounds in the L1 and L2, thus confirming that classroom L2 learners do rely on distributional properties of sounds in the process of L2 speech learning.
Conclusion & Discussion 2. The findings contribute to the existing knowledge of L2 phonology by the key issues in the filed-variability in the input is processed by learners in developing new categories. It suggests new and more detailed ways of computing similarities between the L1 and L2, and thus making more specific predictions about L2 phonological development.