Session 4B

Every human language has its own phonological system, and acquisition of this system underlies language development (Christophe et al. 1997). Although phonological acquisition is now a dynamic branch of second language acquisition research, classroom studies on foreign-language (L2) phonology tend to focus on the effects of phonological awareness activities at intermediate levels of competence, with adult or teenage learners. This paper will report on phonological data from 253 children (ages 6 to 10) at early stages of L2 English learning, in various French primary-school settings. The data was obtained through elicited imitation tasks, and transcribed and analyzed with the PHON software package (Hedlund & Rose 2020). Elicited imitation is frequently used as a measure of general language proficiency (Tracy-Ventura et al. 2014), but when it is transcribed with a tool such as PHON, imitation data becomes a powerful means of investigating emergent foreign-language pronunciation. In addition to imitation data, we also collected information on variables that may play a role in instructed phonological acquisition: classroom methodology and the teacher's phonological competence, as well as each learner's attention span, phonological memory, linguistic profile and motivation for English. Our research questions concern the complex relationships between emerging phonological skill (phoneme production and prosody), and these external and internal variables.
The imitation data comes from three different studies, each initially focused on slightly different classroom factors: study 1 (67 learners, ages 6 and 8) targets the effects of visual speech (facial movements related to articulation) on the phonological features of a set of newly-learned English words; study 2 (54 learners, also 6 and 8) looks longitudinally at the effects of the teacher's phonological competence on phonemic and prosodic learning in two beginning-English classrooms; and study 3 (132 learners, ages 8 to 10) compares phonological acquisition in non-immersive and immersive classrooms. Automatic analysis of this large data set in PHON enables us to confirm the significant effect of a teacher's phonological competence on the production of English vowels and lexical stress. More detailed analyses of the realization of the English consonants /ɹ/, /h/ and /θ/ð/, and of voice onset time in the articulation of initial stop consonants reveal complex interactions of individual and institutional variables in children's L2 phoneme production, which will be briefly summarized. We will conclude our talk by drawing some conclusions for pronunciation learning at early and later primary school levels.
Christophe, A., Guasti, T., Nespor, M., Dupoux, E., & Van Ooyen, B. (1997). Reflections on Phonological Bootstrapping: Its role for lexical and syntactic acquisition. Language and Cognitive Processes, 12(5/6), 585-612.
Hedlund, G., & Rose, Y. (2020). PHON 3.1 Computer Software. https://phon.ca.
Tracy-Ventura, N., McManus, K., Norris, J. M., & Ortega, L. (2014). 'Repeat as much as you can': Elicited Imitation as a measure of oral proficiency in L2 French. In P. Leclercq, A. Edmonds & H. Hilton (Eds.), Measuring L2 Proficiency: Perspectives from SLA. Bristol: Multilingual Matters, 143-166.

The difficulty and complexity of language features for second language learners has attracted considerable theoretical and empirical research attention in recent years (Palotti, 2014; Housen & Simoens, 2016; Bulté & Housen, 2018). Understanding the relative difficulty of language features is important to determine whether simple or difficult features should be the focus of instruction, and whether particular instructional approaches are more effective for simple or difficult features. While there are several methods used to identify the complexity of language features, until now some of the best indicators of feature difficulty are the judgments of experts (e.g. teachers) and perceptions of students (Housen, 2014). With the rise of digital learning methods, such as digital game-based language learning tools, vast amounts of data are being generated by learners as they attempt to learn second and foreign languages. These data can be analysed statistically to determine more objectively which language features are difficult for learners to grasp.
In this study we aim at measuring feature difficulty objectively with data coming from a set of 16 minigames from the Horizon 2020 iRead project, aimed at developing primary school children's reading skills in their first language (5-6 years old) or in English as a Foreign Language (11-12 years old). Data were obtained from 744 Spanish EFL students playing 67,623 minigames using 225 language features across six categories (orthography, phonology, word recognition, morphology, syntax, and morphosyntax). The system automatically logs each game and feature a student plays, as well as their level of success. From this multitude of game logs, the lme4 R package was used to undertake an Item Response Theory based analysis (Kadengye et al., 2014, Debeer et al., 2021) separating student ability level, minigame difficulty, and language feature difficulty. This provides us with an objectively generated list of relative language feature difficulties for this population. Preliminary analysis suggests, for example, that on one end of the scale are adjectives, question words, and phonology of lower frequency consonants (e.g. q, x, j), while on the more difficult end are features such as anaphora, syllabification, and prefixes/suffixes.
Discussion will address the extent to which this bottom-up approach holds promise in our understanding of the difficulty of linguistic features, particularly as more and more data sets like this are generated and explored by SLA researchers.


References:
Bulté, B. & Housen, A. (2018). Syntactic complexity in L2 writing: Individual pathways and emerging group trends. InJAL, 28(1), 147-164.
Debeer, D., et al. (2021). The effect of adaptivity in digital learning technologies. Modelling learning efficiency using data from an educational game. BJET.
Housen, A. (2014). Difficulty and complexity of language features and second language instruction. The encyclopedia of applied linguistics.
Housen, A., & Simoens, H. (2016). Introduction: Cognitive perspectives on difficulty and complexity in L2 acquisition. SSLA, 38, 163–175.
Kadengye, D., et al. (2014). A generalized longitudinal mixture IRT model for measuring differential growth in learning environments. Behavior research methods, 46(3), 823-840.
Palotti, G. (2014). A simple view of linguistic complexity. Second Language Research, 31(1), 117–134.

There has been a surge of interest in L2 research investigating how practice schedule can influence various aspects of L2 learning such as grammar, vocabulary, and pronunciation (e.g., Kasprowicz, Marsden, & Septhon, 2019; Rogers & Cheung, 2020; Li & DeKeyser, 2019). Recent L2 distributed practice research has focused on oral fluency development-a dimension of L2 performance which hinges highly on L2 procedural knowledge (Kormos, 2006). Manipulating the timing of task repetitions has shown to affect the fluency of the repeated performance (Bui, Ahmadian, & Hunter, 2019), and the effects of practice schedule have been found to transfer to a performance on a novel task (Suzuki & Hanzawa, 2021). Research in cognitive psychology suggests that an ideal distribution of repeated practice rests on the ratio of the interval between practice sessions (i.e., the intersession interval; ISI) and the time gap between the final practice session and the time of testing (i.e., the retention interval; RI). However, no research to date has examined the effects of distributed practice on L2 oral fluency development by systematically manipulating the ISI–RI ratio. An investigation of specified ISI–RI ratios is necessary to gain a better understanding of distributed practice effects on L2 fluency development, and how the research findings from cognitive psychology can be applied to a rather complex skill of L2 speaking. The current study, thus, aimed to fill the research gap by examining the effects of distributed practice using the ISI–RI ratios of 10–30%, an optimal range suggested by cognitive psychology research (Rohrer & Pashler, 2007). To this end, 116 Japanese university students participated in an online experimental study. The participants were randomly assigned to one of four groups, which consisted of two experimental groups (a short-spaced group [1-day ISI] and a long-spaced group [7-day ISI]) and two control groups. The experimental groups engaged in four narrative-task practice sessions which were identical in terms of content and procedure, with the only difference lying in the distribution of the practice sessions (1 day vs. 7 days apart). The control groups, by contrast, only took the three tests (pretest, posttest, delayed posttest) which followed the same schedule as each corresponding experimental group. A total of 348 speech datasets were analyzed in terms of speed fluency (e.g., articulation rate), breakdown fluency (e.g., frequency and duration of mid-clause and clause-final pauses), and repair fluency (e.g., repetition). Linear mixed-effects modeling showed the advantage of short-spaced practice on the first posttest conducted one week after the final practice session in terms of the mean duration of mid-clause pauses, but this effect ceased to be observed on the delayed posttest (conducted three weeks later). On the other hand, the long-spaced group showed consistent improvement on the mean duration of mid-clause pauses throughout the experiment, demonstrating greater retention of enhanced fluency performance. The present findings contribute to the existing body of L2 research by yielding insights on how distributed practice may benefit the long-term development of L2 oral fluency.