Session 3B

Research suggests that longer temporal spacing between repetitions of a given target item facilitates retention more than no or shorter spacing, a phenomenon known as the distributed practice effect (e.g., Cepeda et al., 2006). Although existing studies have produced somewhat inconsistent results regarding the effects of spacing on the learning of grammar (e.g., Suzuki & DeKeyser, 2017) or pronunciation (e.g., Li & DeKeyser, 2019), most studies have suggested that distributed practice facilitates second language (L2) vocabulary learning (e.g., Bahrick & Phelps, 1987; Nakata & Webb, 2016). Despite the increasing recognition that spacing facilitates learning, distributed practice is not always implemented effectively in actual classrooms (e.g., Carpenter et al., 2012; Sobel et al., 2011). One reason for this may be that research has failed to provide clear guidelines regarding how spacing should be incorporated into vocabulary instruction. For instance, even though instructors are aware that spacing facilitates vocabulary learning, they may have difficulty deciding how much spacing should actually be used to enhance learning. One way to incorporate distributed practice into the curriculum may be to use cumulative tests, where not only recently studied but also previously studied materials are tested (e.g., Carpenter et al., 2012). By including cumulative questions in vocabulary quizzes, students may be encouraged to distribute practice opportunities over a long period of time, which may lead to better retention. Positive effects of cumulative tests over non-cumulative tests have been reported for the learning of a number of materials such as mathematics and psychology (e.g., Beagley & Capaldi, 2016; Lawrence, 2013). However, since none of the earlier studies have examined the effects of cumulative tests on L2 vocabulary learning, it is not clear whether the positive effects of cumulative tests also extend to L2 vocabulary learning. The present study was designed to compare the effects of cumulative and non-cumulative quizzes on L2 vocabulary learning. In this study, 72 Japanese university students enrolled in a medical English course studied 80 medical English words over nine weekly classes. Participants were randomly assigned to the cumulative or non-cumulative group. In both groups, participants were introduced to 10 target items each week, and took a vocabulary quiz in the following week. In the non-cumulative group, the 10 items introduced in the most recent week appeared in vocabulary quizzes in the following week. In the cumulative group, not only target items introduced in the previous week but also target items introduced in earlier weeks were tested. Learning was measured by cued form-recall and meaning-recall posttests administered three weeks after the last weekly quiz. The analysis showed that the non-cumulative group led to significantly higher scores than the cumulative group on weekly quizzes. On the posttest, in contrast, the results were reversed, and the cumulative group significantly outperformed the non-cumulative group, demonstrating the relative advantage of 106% (meaning-recall) to 238% (form-recall). The findings suggest that non-cumulative group participants perhaps crammed before each weekly quiz but did not review materials after, resulting in higher weekly quiz scores but lower posttest scores. Cumulative group participants, in contrast, probably distributed practice opportunities more evenly throughout the semester, which led to lower weekly quiz performance but better retention over time. Pedagogically, the findings indicate that vocabulary learning can be improved significantly by making simple changes to vocabulary quizzes (i.e., incorporating not only recently introduced but also previously introduced items). The findings have direct application to pedagogy because they provide practical guidelines regarding how to incorporate distributed practice into classrooms. Theoretically, the present study demonstrates that positive effects of cumulative tests reported in earlier non-L2 research also extend to L2 vocabulary learning.

            Research suggests that collocational knowledge facilitates accurate and fluent second language (L2) performance and is essential for successful communication (e.g., Bahns & Eldaw, 1993; Nesselhauf, 2005; Pawley & Syder, 1983). Given the importance of collocational knowledge for L2 learning, how collocational learning can be facilitated is an important question for researchers, instructors, and learners. One potential way to facilitate collocational learning may be to introduce temporal spacing between encounters of a given collocation. Existing research suggests that spaced learning, which involves spacing between encounters for a given target item, facilitates L2 vocabulary learning in comparison to massed learning, which does not involve any spacing (e.g., Karpicke & Bauernschmidt, 2011; Nakata, 2015; Nakata & Elgort, 2019). Considering the benefits of spacing for vocabulary learning, the present study examined whether positive effects of spacing also extend to L2 collocational learning. 
            Although several studies have examined the effects of spacing on L2 collocation learning (Macis et al., 2019; Snoder 2017), they are limited in that they only examined the item learning of collocations, and not system learning. In other words, existing studies only investigated whether spacing facilitates the acquisition of collocations encountered during the treatment, and not whether spacing allows learners to transfer the knowledge of studied collocations to unstudied, novel collocations. For instance, when learners are exposed to collocations such as "draw attention," "draw tears," or "draw laughs," they perhaps not only learn these collocations as unanalyzable chunks (i.e., item learning), but also make generalizations about what kinds of nouns the verb "draw" can take as an object (i.e., system learning). One may assume that massing is particularly beneficial for facilitating system learning. For instance, when learners are exposed to multiple collocations for the node word "draw" on the same day (i.e., massed learning), it may allow learners to notice the core meaning underlying different uses of the node word. In contrast, when multiple collocations for the same node word are introduced on different days (i.e., spaced learning), learners may have difficulty extracting the core, underlying meaning of the node. This suggests that although spacing may facilitate the item learning of collocations according to the spacing effect, massing may be more beneficial for system learning. The present study examined whether the benefits of spacing observed in existing research also extend to collocational learning, in terms of both item and system learning.
            In this study, 90 Japanese high school students learned 27 verb-noun collocations under massed, spaced, or hybrid (a combination of massing and spacing) conditions. The treatment was conducted on nine separate days over three weeks. In the massed group, three collocations with the same node verb (e.g., draw a line, draw a tear, draw a conclusion) were studied on the same day and they were never repeated in subsequent sessions. In the spaced group, three collocations for the same node verb were studied in different weeks (i.e., Week 1: draw a line, Week 2: draw a tear, Week 3: draw a conclusion). In the hybrid group, three collocations for the same node verb were introduced on the same day. Furthermore, practice opportunities for a given collocation were distributed over 3 weeks. Learning was measured by a pretest and posttests. To examine whether participants can transfer the knowledge of studied collocations to unstudied collocations, the pretest and posttests also included collocations that contained the same node verb but were not encountered during the treatment.
            Results showed the hybrid group led to the largest gains for both studied and unstudied items, with the massed group being the least effective, suggesting a combination of massing and spacing facilitates item and system learning of collocations. 

Deictic motion verbs such as come and go seem to be particularly challenging for second language (L2) learners. Previous studies have shown that languages can be classified according to how come verbs are used, only for motion towards the speaker (e.g. Spanish) or for motion towards the addressee and the speaker (e.g. English, Greek) (Gathercole, 1977, 1978). These differences in how deictic verbs are used can lead to different thinking-for-speaking patterns (Slobin, 1996). When acquiring an L2, these contrasts can be the source of conceptual transfer or cross-linguistic influence (Jarvis & Pavlenko, 2008). Greek deictic verbs present specificities (Antoponoulou & Nikiforidou, 2002), such as the implication of the presence/absence of the speaker in accompaniment situation or the use of ??? ‘go’ as ‘take’ when it becomes transitive. Previous research showed how challenging the acquisition of these verbs is in L2 Greek, for monolinguals with different deictic patterns (Spanish) and for bilinguals with closer patterns (Catalan/Spanish) (Authors, 2018). However, pedagogical interventions seem to help learners to improve their use of deictic patterns (Colasacco, 2019). The main aim of this study is to examine whether the explicit teaching of thinking-for-speaking patterns improves the acquisition of these verbs in L2 Greek, a language that presents more specificities than Spanish in terms of deictic motion. Participants were a group of intermediate (B1) L1 Spanish/Catalan learners of Greek (n=15) who were learning the target language in a language school in Barcelona, Spain. Data were elicited by means of (i) four videos that encouraged the use of motion verbs; (ii) a cloze test activity which included the target verbs and (iii) an activity of pedagogical translation. Prior to the intervention, participants’ knowledge of the deictic motion verbs was explored (pre-test). The intervention consisted of a two-hour session of explicit instruction of the verbs under analysis. After the intervention, post-tests were conducted in three different times: post-test 1 (T1) immediately afterwards; post-test 2 (T2) one week later and post-test 3 (T3), one month later. Different data collection times were used in order to explore the long-term effects of the intervention (or lack thereof). Both quantitative and qualitative analyses were performed. Results showed that, prior to the intervention, deictic motion verbs constituted a problematic area for the participants. However, the analysis of the post-tests revealed that the intervention seemed to have a positive effect on the acquisition of these patterns. The study concludes by discussing the pedagogical implications for L2 learning and teaching, with a potential long-term effect of the explicit instruction of thinking-for-speaking patterns. REFERENCES Antonopoulou, E. & Nikiforidou, K. (2002). Deictic motion and the adoption of perspective in Greek. Pragmatics, 12(3), 273-295. Colasacco, M. (2019). A Cognitive Approach to teaching deictic motion verbs to German and Italian students of Spanish. International Review of Applied Linguistics to Language Teaching (IRAL), 57(1), 71-96. Gathercole, V. (1977). Study of the comings and goings of the speakers of four languages: Spanish, Japanese, English and Turkish. Kansas Working Papers in Linguistics, 2, 61-94. Gathercole, V. (1978). Towards a universal for deictic verbs of motion. Kansas Working Papers in Linguistics, 3, 72-88. Jarvis, S., & Pavlenko, A. (2008). Crosslinguistic influence in language and cognition. New York, NY: Routledge. Slobin, D. I. (1996). From “thought and language” to “thinking for speaking”. In J. J. Gumperz & S. C. Levinson (?ds.), Rethinking linguistic relativity (pp. 70-96). Cambridge, UK: Cambridge University Press. Stam, G. (2010). Can L2 Speaker’s Patterns of Thinking-for-Speaking Change? In Z.H. Han & T. Cadierno (Eds.), Linguistic Relativity in SLA. Thinking for speaking (pp. 59-83). Bristol, UK: Multilingual Matters.

The role of attention has been one of the main issues studied in SLA and, following VanPatten's (2004) proposal of the primacy of meaning, a number of researchers (e.g. Leow et al. 2008; Morgan-Short et al. 2012) have explored whether attention to meaning takes precedence over attention to form when second language (L2) learners engage in the process of comprehension. Continuing this line of research and building on a large multi-site replication project (authors xxxx), this paper reports on two studies which investigated the effects of attention to form during listening comprehension by 81 L1-Polish learners of Spanish. However, rather than using English as in the original design, we included Polish, our participants' mother tongue (Study 1; N=37,) and Spanish, their dominant foreign language (Study 2; N=44), to measure their L2 listening comprehension. By including this modification, we were able to examine the effects of the language in which L2 performance was measured as well as explore the role of proficiency in the process of L2 comprehension. In terms of treatment, the learners listened to a short text in Spanish and answered ten multiple-choice comprehension questions. Importantly, during the listening process, the learners were also asked to pay attention to L2 forms and make check marks upon hearing different linguistic features: a lexical item ('sol'), an article ('la'), a verb-ending suffix ('-n'), and a focus only on the listening process (a control group). The aim of this procedure was to check whether directing L2 learners' attention to grammatical and lexical forms affected their comprehension scores. Results revealed a fairly high comprehension level (over 40%), but the students' scores varied depending on which language they were tested in and which feature they were asked to attend to. While in Study 1 there were no differences in the comprehension scores between the four conditions, in Study 2, where the comprehension was measured in Spanish, there was a statistical difference between the control condition and the 'sol' and '-n' conditions, with the –n suffix appearing to be particularly problematic for the learners and therefore negatively affecting their comprehension. Such results suggest the difficulty of paying simultaneous attention to meaning and non-salient L2 forms such as verb-ending suffix -n if the main goal of the language task is text comprehension. Finally, we also ran correlation analyses between the learners' L2 proficiency and comprehension, with only Study 2 revealing a significant relationship (r=.56). This result is in line with the original study (authors xxxx), where we also found a significant correlation between participants' Spanish proficiency level and their comprehension. The results of these two studies will be discussed in relation to our previous findings in this area as well as the findings of other researchers. We will particularly focus on the role of L2 proficiency and multilingualism as potential factors that might have influenced the behaviour of our participants. We will also consider the benefits of conducting replication research, offering further ideas for studies into the role of attention in L2 comprehension. References Leow, R. P., Hsieh, H. and Moreno, N. (2008). Attention to form and meaning revisited. Language Learning, 58, 3, 665-695. Morgan-Short, K., Heil, J., Botero-Moriarty, A. and Ebert, S. (2012). Allocation of attention to second language form and meaning. Studies in Second Language Acquisition, 34, 4, 659-685. VanPatten, B. (2004). Input processing in SLA. In B. VanPatten (Ed.) Processing instruction: theory, research and commentary (pp. 5-31). Mahwah: Erlbaum.