Session 4F

Usage-based approaches to language learning view multi-word sequences (MWS) as essential building blocks for language learning and processing (Christiansen & Chater, 2016). Collocations, a specific type of MWS, hold a prominent position in psycholinguistics, corpus linguistics, and language pedagogy research. Various psycholinguistic experiments have demonstrated that collocations are processed faster than novel phrases by L1 and L2 speakers (e.g. Wolter & Yamashita, 2018). However, relatively little attention has been paid to the effect of single word-frequency on collocational processing. We therefore first investigated the prominence of single word and collocation frequency information for processing high- and low-frequency collocations. Secondly, we examined whether there is a difference between L1 and L2 speakers’ sensitivity to transitional probabilities of collocations in relation to the specific association measures used. A number of previous studies used the Mutual Information (MI) statistic to identify strongly-collocated word combinations. However, one of the main issues with this methodological choice is that MI scores are negatively linked to frequency. Hence, these studies do not only tend to extract strongly associated, but also low-frequency collocations, which have only a weak link to language exposure (Gablasova, Brezina, McEnery, 2017). It is therefore important to explore alternative association measures, which are better suited for identifying collocations in SLA research. This study offers a validation of an alternative collocation association measure for extracting adjective-noun collocations using the Log Dice (LD) measure. LD is calculated as the harmonic mean of two proportions, which indicates the tendency of two words to co-occur relative to the frequency of these words in the corpus. Therefore, unlike MI, LD highlights exclusive, but not necessarily rare combinations. We used an on-line acceptability judgment task to explore L1 English (n=30) and advanced level L2 English speakers’ (n=32) processing of adjective-noun collocations. A total of 120 adjective-noun combinations were extracted from the BNC (98 million words). The items fell into one of the three critical conditions: (1) high-frequency collocations (e.g. dark hair), (2) low-frequency collocations (e.g. lovely house), and (3) non-collocational (baseline) items (e.g. general eyes). Single word frequency, collocation frequency counts, and the MI and LD scores of the items were obtained from the BNC. Response times were analysed using mixed-effects regression modelling. As for frequency, both groups of participants showed sensitivity to both individual word and collocation frequency counts. However, there was a reduced effects of individual frequency counts for processing high-frequency collocations compared to low-frequency collocations. That is to say, the effects of word-level and collocation level frequency information changes for processing low- and high-frequency collocations for both L1 and L2 speakers. This study contributes to the growing body of research that both L1 and L2 speakers are sensitive to the frequency distributions of MWS at multiple grain sizes. Both groups of participants were similarly sensitive to the MI and LD measures. However, the model including LD measure is a better fitting model than the model including MI measure. The findings are in line with the predictions of usage-based approaches that the cumulative experience speakers have with a target language appear to impact both L1 and L2 processing similarly. REFERENCES Christiansen, M. H., & Chater, N. (2016b). The Now-or-Never bottleneck: A fundamental constraint on language. Behavioural & Brain Sciences, 39, e62. Gablasova, D., Brezina, V., & Mcenery, T. (2017). Collocations in Corpus-Based Language Learning Research?: Identifying , Comparing , and Interpreting the Evidence. Language Learning, 1–25. Wolter, B., & Yamashita, J. (2018). Word frequency, collocational frequency, L1 congruency, and proficiency in L2 collocational processing: What accounts for L2 performance? Studies in Second Language Acquisition, 40(2), 395–41

Pop songs have been defined as “the motherese of adolescents” (Murphey & Alber, 1985) and could thus provide a noteworthy source of L2 native input both inside and outside the classroom, especially for this population. Despite the wide availability of songs and the frequent engagement with music that teenagers report, very little empirical research exists that examines the potential of songs to promote L2 vocabulary acquisition. Pavia, Webb, and Faez’s (2019) is one of the few existing studies on this topic, and their results suggest that listening to two songs repeatedly can successfully promote incidental vocabulary learning. To the authors’ knowledge, there are no equivalent studies for intentional vocabulary learning through songs in classroom settings, which is a context in which songs are sometimes used and are always welcome by the students. The aim of the present study is to further analyze incidental vocabulary learning through songs in class and compare it to more focused learning, in which, apart from listening to songs, learners receive explicit vocabulary instruction and practice the target words. In the case of reading and viewing, it has been shown that pre-teaching or directing the students’ attention to the target vocabulary, typically promotes more learning than no focus (Pujadas & Muñoz, 2019), and it would be interesting to analyze whether this is also the case for songs. Moreover, the present study will examine whether having access to the written lyrics (which is common classroom practice) is more beneficial for vocabulary learning than just oral input, in line with other studies which have shown advantages for bi- or multimodal input (Webb & Chang, 2012). The research questions that guide this study are the following: 1) Can the use of songs promote vocabulary learning in an EFL class? 2) Does type of instruction (with or without direct focus) and input modality (with or without written lyrics) have an effect on vocabulary learning through songs over time? The participants in this study were grade 8 Italian learners of English (N = 60) in three intact classes, who listened to two songs repeatedly in two different classroom sessions (one song per session). Group 1 (N = 20) and Group 2 (N = 18) listened to the songs while reading the lyrics. For each group there was a direct focus on vocabulary for one of the songs only, in a counterbalanced fashion. Group 3 (N = 22) only received oral input; otherwise, they followed the same treatment as Group 1. There were 16 target words, on which all the students performed a productive (C-test) and a receptive (multiple choice) vocabulary test before and after the treatment. The results of the GLMMs suggest that, overall, the treatment was quite positive and students experienced significant vocabulary gains after two sessions listening repeatedly to two songs. Moreover, including a direct focus on vocabulary led to more vocabulary gains, but only in terms of productive knowledge. As for modality, having access to the written lyrics did not promote more vocabulary learning than just oral input. References Murphey, T., & Alber, J. L. (1895). A pop song register: The motherese of adolescents as affective foreigner talk. TESOL Quarterly, 19, 793-795. Pujadas, G., & Muñoz, C. (2019). Extensive viewing of captioned and subtitled TV series: a study of L2 vocabulary learning by adolescents. The Language Learning Journal,47, 479-496. Webb, S., & Chang, A. (2012). Vocabulary learning through assisted and unassisted repeated reading. Canadian Modern Language Review, 68(3), 267–290.

Research in second language (L2) phonological and lexical acquisition has shown that the L1 lexicon is activated during L2 speech processing (e.g., Costa, 2006). Greater level of L1 activation is in fact found when processing cognate words, that is, L2 words that share semantic, phonological and orthographic properties with their L1 counterparts, resulting in faster vocabulary learning and access to lexical meaning (Tessel et. al. 2018). By contrast, studies have reported less accurate pronunciation in cognate than in non-cognate words (e.g., Flege and Munro, 1994; Amengual, 2016), presumably due to a greater influence from the L1 with cognates. Still, while some studies argue that cognate effects may disappear with increasing L2 experience (e.g., Jacobs et al., 2016), other studies have failed to find a relationship between L2 experience and cognate effects (Amengual, 2012). Finally, research on the relationship between cognate status and L2 perception is scarce (cf. Tessel et al.'s (2018) EEG study). The goal of this paper is to examine the relationship between cognate status and proficiency in L2 perception and production. In particular, the study evaluates if mispronunciations are more difficulty to detect in cognate than in non-cognate words (i.e., cognate effect), due to a greater L1 influence with cognates. In a first study three groups of L1 Spanish/Catalan L2 English speakers differing in L2 proficiency were asked to determine if instances of two English words were equally accurately produced or if one sounded more native-like than the other. Stimuli consisted of target-like as well as mispronounced cognate and non-cognate words involving several target English features (aspiration, intervocalic voiced stops and the non-L1 phoneme /v/). Results showed cognate effects for all groups except the lowest proficiency group, which showed the most errors regardless of cognate status. These results will be contrasted with the results of a second study underway, in which similar groups of participants perform a forced choice goodness task including additional phonological variables. In addition, participants' production of the words used in the perception task is also elicited by means of a story reading task. Preliminary results reveal greater individual variation than in the first study and a consequent less clear pattern of overall cognate effects in terms of accuracy of responses. The results of the two studies will be discussed in terms of the relationship between cognate status in L2 perception and production and the roles of L2 proficiency and type of phonological variable. Amengual, M. (2012). Interlingual influence in bilingual speech: Cognate status effect in a continuum of bilingualism. Bilingualism, 15(3), 517–530. Amengual, M. (2016). Cross-linguistic influence in the bilingual mental lexicon: Evidence of cognate effects in the phonetic production and processing of a vowel contrast. Frontiers in Psychology, Apr;7:617. Costa, A. (2006). Speech Production in Bilinguals (Chapter 8). The Handbook of Bilingualism, 201–223. Flege, J. E., & Munro, M. J. (1994). the Word Unit in Second Language Speech. Studies in Second Language Acquisition, 381–411. Jacobs, A., Fricke, M., and Kroll, J. (2016). Cross-language activation begins during speech planning and extends into second language speech. Language Learning, 66, 324–353. Tessel, C. A., Levy, E. S., Gitterman, M., & Shafer, V. L. (2018). Neurophysiological indices of the effect of cognates on vowel perception in late Spanish-English bilinguals. Journal of Phonetics, 68, 117–137.

On-line processing of idioms by native and non-native speakers of English: Evidence from eye-tracking Abstract Mastering formulaic language is widely accepted as a mark of proficiency as it poses a challenge to even the most proficient L2 learners. Siyanova-Chanturia, Conklin, and Schmitt (2011) found a clear evidence that formulaic language holds a privileged processing status for native speakers, but this is clearly not the case for non-native speakers of English. There is also no conclusive evidence on what drives the idiom advantage that is apparent among native speakers. Previous research suggests that L2 proficiency level, familiarity with the idiomatic expression, and L1 knowledge all figure prominently in the processing of idioms by non-natives. Using eye-tracking, this study investigates the on-line processing of idioms by native and non-native speakers of English. Thirty-one native English speakers and 42 Arabic-speaking L2 learners of English from various proficiency levels (their IELTS scores ranged between 4.0 and 7.5) participated in the experiment. The study looks at whether participants, who vary in language proficiency levels, show a processing advantage for idiomatic phrases over control phrases, and the extent to which this effect is modulated by familiarity and idiom congruency (i.e., the degree to which the idiom is similar to the L1 of the non-native participants). Participants were presented with a set of English idioms (66 stimuli) and control phrases (66 stimuli) embedded in short, context-neutral sentences. In control phrases, the first word of the idiom was replaced by a carefully matched plausible alternative (e.g., break/crack the ice). The selection of idioms was initially based on the level of their familiarity and congruency, and so were varied along these dimensions. Nevertheless, all participants were asked to complete a rating task to indicate subjective familiarity with the idioms (administered after conducting the experiment). Additionally, non-native speakers were asked to rate the congruency level of all the idioms. Following Carrol, Conklin and Gyllstad (2016), the study employed a range of early and late measures at both the phrase level and the sentence level involving analysis of the number and duration of fixations during natural reading. Results were analysed using an omnibus linear mixed-effects model which shows clear effects of Group, whereby English native speakers were faster readers than the non-natives. Native speakers’ results also indicate a processing advantage for idioms over control phrases. Contrary to this, non-native findings suggest that L2 learners process idioms at a similar speed to control phrases, particularly at the lower proficiency level. Results also support the view that L1 knowledge is utilised during on-line processing of idioms, but advanced proficiency does not lead to native-like processing in the L2. References 1- Carrol, G., Conklin, K., & Gyllstad, H. (2016). Found in translation: The influence of the L1 on the reading of idioms in a L2. Studies in Second Language Acquisition, 38(3), 403–443. 2- Siyanova-Chanturia, A., Conklin, K., & Schmitt, N. (2011). Adding more fuel to the fire: An eye-tracking study of idiom processing by native and non-native speakers. Second Language Research, 27(2), 251–272.