System 42 (2014) 93–104

Contents lists available at ScienceDirect

System
journal homepage: www.elsevier.com/locate/system

Native speaker–non-native speaker study abroad

conversations: Do they provide feedback and opportunities
for pushed output?
Marisol Fernández-García a,1, Asunción Martínez-Arbelaiz b, *
a
Emmanuel College/Tufts University, Department of Foreign Languages, 400 The Fenway, Boston, MA 02115-5000, USA
b
University Studies Abroad Consortium, Edificio Ignacio M. Barriola, Universidad del País Vasco/EHU, Plaza Elhuyar 1,
20018 Donostia-San Sebastián, Spain

a r t i c l e i n f o a b s t r a c t

Article history: An important body of literature indicates that, for language development to occur, learners
Received 9 November 2012 must engage in exchanges that provide feedback and chances to incorporate it in their
Received in revised form 21 October 2013 production. This research has been carried out mainly in classroom and experimental
Accepted 31 October 2013
settings (Gass, Mackey, & Ross-Feldman, 2005; Sheen, 2004). In the current study, we
examine whether opportunities for feedback and uptake emerge in the context of semi-
Keywords:
informal study abroad conversations. Oral data were collected from eight dyads of
Interaction
Native Speaker–Non-Native Speaker (NS–NNS) of Spanish, who voluntarily participated in
Negative feedback
Uptake
Spanish–English language exchange sessions. Some feedback moves previously identified
Pushed output in the literature were also found in this study abroad context. In addition, learner-initiated
Study abroad moves played a crucial role in discourse development and led to the highest levels of
uptake. We discuss how this active involvement of the learner of Spanish in the analyzed
semi-informal conversations is facilitated by the fact that both interactants share an
identity as language learners (Benson, Barkhuizen, Bodycott, & Brown, 2012).
Ó 2013 Elsevier Ltd. All rights reserved.

Introduction

In a recent review of study abroad research, Collentine (2009) notes a commonly underlying assumption in most studies,
namely, that learners benefit from study abroad because it provides a wide range of opportunities to comprehend and
produce the L2 in communicative contexts. He cautions, however, that there is a lack of primary data that document the
nature of the interaction between learners and native speakers. In particular, he observes that although there are studies that
have examined opportunities for negotiation from a sociocultural perspective (DuFon & Churchill, 2006; Kinginger, 2008), not
much research has addressed this issue form the SLA interactionist view.
The interactionist perspective (Gass, 1997; Iwashita, 2003; Long, 2007; Pica, 1994) maintains that, through conversations,
language learners are exposed to two types of evidence crucial for language learning: positive evidence and negative evi-
dence. Positive evidence has been defined as “the set of well-formed sentences to which learners are exposed” (Gass, 1997, p.
36). Via this type of evidence learners gather data about possible and acceptable utterances in the target language. Negative
evidence, on the contrary, provides information about what is neither possible nor acceptable.

* Corresponding author. Tel.: þ34 943 015 352; fax: þ34 943 311 623.
E-mail addresses: [email protected] (M. Fernández-García), [email protected] (A. Martínez-Arbelaiz).
1
Tel.: þ1 617 489 5578.

0346-251X/$ – see front matter Ó 2013 Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.system.2013.10.020
94 M. Fernández-García, A. Martínez-Arbelaiz / System 42 (2014) 93–104

Learners typically obtain this type of evidence by means of corrective feedback, that is, an interlocutor’s turn containing
some sort of indication that the learner’s utterance is not target-like.
There has been considerable debate in the literature with respect to whether both types of evidence are necessary or if
positive evidence alone is sufficient for L2 development. On the one hand, some researchers (Krashen, 1981; Schwartz, 1993)
have argued that positive evidence is all that is needed for language acquisition to take place, and that negative evidence
might even be counterproductive. The implication for language teaching, including study abroad, would be that all efforts
should aim at providing exposure to input rather than feedback on linguistic accuracy. Other researchers (Gass, 1997; Long,
2007; Pica, 1988; Swain, 1985; White, 1991), however, have put forward a role for negative evidence in SLA. Swain (1985), for
example, has argued that through negative feedback learners can be pushed to reformulate their utterances and make them
more accurate by resorting to their own linguistic resources.
Interactionists (Gass, 1997; Long, 2007; Pica, 1988, 1994) have posited a major role for both positive and negative evidence
in SLA. They argue that, through the provision of negative feedback, learners are given opportunities to consider problems in
their output, as indicated by their conversational partner. This negative feedback, in turn, may help them make the appro-
priate modifications. In other words, this process is said to allow learners to realize that there are discrepancies between their
output and the target norm. Underlying this claim is Schmidt’s noticing hypothesis (1990, 2001), which argues that some
degree of consciousness is necessary for language acquisition to proceed.
As Ellis, Basturkmen, & Loewen (2001a) contend, there are theoretical bases for considering that uptake may facilitate
acquisition. Uptake has been primarily studied in classroom contexts, and it has been commonly defined as ‘‘what the
student attempts to do with the teacher’s feedback’’ (Lyster & Ranta, 1997, p. 49). Ellis et al. (2001a) and Ellis, Basturkmen,
and Loewen (2001b) point out that uptake may also occur in the absence of learners’ errors. That is, they broaden the
notion of uptake to include not only reactive but also preemptive focus on form. Uptake would then be any learner’s
attempt to incorporate teacher’s feedback that: a) follows a learner’s error or b) follows a learner’s doubt or question
about a language form. Furthermore, Ellis et al. (2001a) specify that uptake is an optional move in which the learner tries
to incorporate the teacher’s feedback into her own output. This feedback is provided in response to an evident gap in the
learner’s knowledge and can vary on a number of dimensions that, in turn, may affect the likelihood of feedback
incorporation (Loewen, 2004).
One of the most frequently studied characteristics of feedback in relation to uptake is explicitness/implicitness. There
is some evidence that suggests that explicit feedback affords more opportunities for noticing than other types of more
implicit feedback. For example, clarification requests and prompts are two types of explicit negative feedback that are
likely to elicit learners’ modifications of their previous ill-formed utterances. That is so because these types of explicit
negative feedback hold back correct forms and instead offer hints to encourage learners to adjust their output and thus
stretch their L2 knowledge. Clarification requests and prompts are then more often followed by modified responses, that
is, by uptake, than implicit types of feedback, such as recasts (e.g., Lyster, 1998; Lyster & Ranta, 1997; Oliver, 1995, 2000).
A recast is a reformulation of an interlocutor’s previous utterance that corrects it in some way (Lyster & Ranta, 1997)
within the context of a communicative exchange (Long, 2007; Sheen, 2006). Since the recast already provides the ex-
pected modification of the learner’s output, she may not perceive an immediate pressure or need to respond to it. In fact,
studies have shown that prompts are in general more effective than recasts at eliciting modified responses in the
classroom (Ammar & Spada, 2006; Ellis, Loewen, & Erlam, 2006; Lyster, 2004; Lyster & Saito, 2010; Yang & Lyster, 2010).
This is in part due, according to Yang & Lyster (2010), to prompts being more consistent in terms of discourse saliency as
well as in terms of requiring self-correction.
Despite their advantages for acquisition, prompts are not as frequent as recasts, which are claimed to be the most recurrent
feedback type in typical classroom discourse (Lyster & Ranta, 1997; Sheen, 2004). Teachers often prefer to use recasts because
they are an implicit form of feedback, less disruptive of conversational interaction than explicit feedback types such as
prompts, explicit correction or metalinguistic clues. Recasts can be quite comfortably inserted in a conversation to provide
feedback because they function similarly to naturally occurring conversational continuants. For this reason, it has been argued
(Ellis & Sheen, 2006) that recasts have the potential to provide positive and negative evidence. On the one hand, they can act
as models for acquisition and thus constitute positive evidence; on the other, if the learner is able to perceive the contrast
between her turn and the recast, then, recasts can contribute as negative evidence for acquisition.
Together with recasts, another frequently studied type of feedback-including interaction has been the negotiation of
meaning (e.g., Iwashita, 2003; Mackey, 2006; Mackey & Philp, 1998; Oliver, 2000; Varonis & Gass, 1985). As noted by Ellis
(2006), negotiation for meaning can offer feedback types that are input-providing and output-pushing. A negotiation is
input-providing when the learner indicates non-understanding to the native speaker, thus prompting her to give the
learner additional input. Conversely, a negotiation is mainly output-pushing when the native speaker is the one signaling
non-understanding to the learner. However, depending on the extent of the non-understanding, the native speaker may
choose to use a specific indicator such as a confirmation check, which is mainly of the input-providing type, or a less
specific one such as a clarification request, which is mainly an output-pushing type. It has been pointed out (Loewen &
Philp, 2006; Lyster, 1998) that, in the former case, there may be an overlap with recasts, when the confirmation check
repeats the learner’s previous utterance without the error. While that is possible, there are also confirmation checks that
are unambiguous recasts. They do not qualify as indicators of non-understanding given that they repeat the learner’s
utterance while correcting a form-error that has no bearing on meaning (e.g., NNS: He work in the office, NS: He works in
the office? NNS: Yes).
 M. Fernández-García, A. Martínez-Arbelaiz / System 42 (2014) 93–104 95

Most studies have examined issues related to the provision of positive and negative evidence in interactions between
learners and NSs in the second and foreign language classroom, in laboratory settings as well as in immersion programs. As
noted by Collentine (2009), not much research has investigated learners’ opportunities to obtain positive and negative ev-
idence in study abroad interactions. Thus, despite the claim that study abroad provides an ideal context for language
acquisition because it combines both formal classroom instruction and access to a range of naturally occurring interactions
with native speakers, little is known about the nature of these conversations.
To the present authors’ knowledge, only two studies have adopted the interactionist perspective to examine the
conversations of L2 learners with NS interlocutors in a study abroad context (Fernández García & Martínez Arbelaiz,
2007; McMeekin, 2006). In both studies, the participants were college-age students (18–21 years old) with a level of
proficiency that ranged from intermediate to low advanced, as measured by the placement test of the host universities.
McMeekin’s study of five female native speakers of English learning Japanese found that these learners moved between
two different sub-contexts. In the host family environment, learners often engaged in negotiations of meaning of the
input providing type. Native speakers responded to most of students’ requests for clarification, helping them to under-
stand their utterances. Thus, while students’ need for comprehensible input was addressed in the host family environ-
ment, opportunities for uptake were rather limited. McMeekin found that negotiations of meaning of the type that Ellis
calls output-pushing (2006) were rare, mainly because native speakers rephrased most of the learners’ potentially
troubled utterances, leaving no room for learners to modify their output.
In contrast, in the classroom context, learners found fewer opportunities to obtain native-like comprehensible input
mainly because instructors favor a traditional focus on learning. Rather than clarifying their own utterances in response to
students’ request, instructors often opened the floor to other students. This resulted in some input that was comprehensible
but less than optimal due to learners’ limited command of the L2 and the frequent reliance on English in the classroom.
Moreover, in the classroom context instructors provided negative feedback to push the learners to correct their utterances
even in the absence of communication breakdowns. To sum up, in McMeekin’s study, learners seemed to move from a family
context that favored communication and the provision of positive evidence to a classroom context where linguistic accuracy
and the provision of negative evidence predominated.
In a previous study, we examined (Fernández García & Martínez Arbelaiz, 2007) study abroad NS–NNS interaction in
what we called, following Collentine and Freed (2004), “a hybrid context”: the out-of-class semi-informal conversations
between English L1 learners of Spanish and their Spanish native speaker conversational partners. The study focused on
how these conversations afforded opportunities for comprehensible input through input-providing negotiations of
meaning. As in McMeekin’s study, results showed that the discourse generated by these dyads contained negotiation
moves that not only helped make the NS’s input comprehensible to the NNS, but also offered a number of syntactic,
semantic and more rarely morphological modifications that provided L2 information on how the target language is
codified and de-codified (Pica, 1994). Moreover, an analysis of the type of indicators used by NNSs to request clarifi-
cation showed that the most effective ones in prompting NSs’ input rich in linguistic modifications were those that
pointed to the specific source of the non-understanding. Therefore, NSs’ modifications helped to make salient for the
learners the form, function and meaning of the linguistic units, thus providing positive evidence for the rules of the
target language. This study, however, did not examine negotiations of the output-pushing type, which we noted were
also frequent in the interactions.
In the current study, we analyze the same conversational data to ascertain whether there were opportunities, other than
negotiations of meaning, for learners to adjust their output toward the L2 norm. It is generally believed that learners in study
abroad programs can benefit from a variety of situations in which they have to use the language with a social and/or
communicative purpose. Smartt & Scudder (2004), for instance, consider study abroad an “L2-rich environment, with
appropriate modeling”. They observe in students with a study abroad experience an increasing disuse of language switch and
a greater effort to locate the most appropriate word compared to those students who only had classroom language experience
(p. 597). On the other hand, Lafford (2004) contends that NNSs abroad learn to communicate within pragmatic constraints
that discourage the use of communication strategies. Not much is known, however, about whether and how study abroad
conversations might help learners to improve language accuracy. As stated above, research indicates that in order to improve
language accuracy learners must engage in exchanges that provide feedback and chances to incorporate it in their production.
Such research, however, has been carried out mainly in classroom and experimental settings (Gass et al., 2005; Sheen, 2004).
In this study, we examined whether opportunities for uptake following feedback emerge in the context of semi-informal
study abroad conversations.

Research questions

1 Which discourse moves do NSs use to help learners of Spanish “repair” or “push” their output when engaged in semi-
informal conversations in a study abroad context?
2 What is the relative frequency of such discourse moves?
3 Which moves are successful at eliciting modified output by learners of Spanish engaged in semi-informal conversations
with NSs in a study abroad context?
96 M. Fernández-García, A. Martínez-Arbelaiz / System 42 (2014) 93–104

The study

Participants
Eight NS–NNS dyads participated in this study. The NNSs were learners of Spanish that were enrolled in a semester abroad
program in San Sebastián, in the Basque Country, Spain. They were undergraduates from different North American univer-
sities, and all had English as their L1. While in Spain, they were taking third- or fourth-year Spanish language and culture
courses for foreigners at the University Studies Abroad Consortium within the University of the Basque Country. Based on the
results of the placement test, these students were enrolled in intermediate to low advanced Spanish courses.2 The NSs were
also undergraduate Spanish students. Both the NSs and NSSs were volunteers in a language exchange program. The Study
Abroad Office provides support in initiating these exchanges by placing advertisements around campus to announce the
availability of these language exchange sessions. Through these ads a group of Spanish NS college students were recruited.
Study abroad American students were then offered the opportunity of participating in these exchanges. Those interested
were paired up with a Spanish partner, and individual preferences (e.g., gender preference, age, hobbies) were honored
whenever possible. The Study Abroad Office recommends alternating Spanish exchange sessions and English exchange
sessions so that both learners have equal opportunity to practice their L2. At the moment that data were collected, the
participating dyads had been meeting for language sessions for approximately a month. This study focuses on the Spanish
language exchange sessions only.
While arguably these semi-informal out-of-class exchange conversations could take place in any context other than study
abroad,3 a cursory look at the transcripts of the conversations revealed that the fact that the Spanish exchange sessions took
place in the country were the language was spoken facilitated the active involvement of both participants in the conversa-
tions. We noted that topics that related to the local culture (e.g., visits to small towns, using public transportation, restaurants
and night life, local food, etc.) predominated, and that both participants contributed to topic initiation in a balanced manner.
That is, despite the fact that the Spanish NS acted as the culture and language "expert", the learner’s experiences abroad
allowed her to participate on an equal footing in the conversation.

Data collection

The data used for this study came from recordings of exchange sessions, one per dyad. After the dyads had met for a
month, one conversation was taped as a representative sample of these language exchange interactions. Prior to the data
collection, dyads were asked for their consent to participate in the recording session. In order to not disrupt the regular
conditions of the conversations, recordings were made at the location where dyads had scheduled their meeting. One of the
researchers left a portable recorder on, and after thirty minutes she returned to the meeting point and turned the recorder off.

Analysis of the data

The recordings were first transcribed following a simple convention system as specified in Appendix A. In order to address
the research questions, all instances in which learners received feedback from the NSs were identified. Given the scarcity of
research on the topic of feedback in the context of study abroad conversations, the researchers did not confine the codification
to those moves identified in previous research.4 As expressed above, much of previous research on feedback has been carried
out in classroom contexts where certain moves, such as prompts, meta-linguistic feedback and explicit correction, may be a
result of an overt pedagogical goal. Based on our own observations, we expected that this type of overt pedagogical feedback
would be less frequent or even absent in the context of a semi-informal out-of-class conversation. Furthermore, there existed
the possibility that dyads would use other feedback moves congruent with the social and communicative nature of these out-
of-class conversations. Following this reasoning, we approached the data looking for moves in which the NS offered any sort
of feedback or answer to learner’s moves that might help the NNS to make her output more comprehensible and/or accurate.

Results

The goal of the first research question was to find out which discourse moves pushed the learner to repair or modify her
output. With respect to the type of moves, an analysis of the data showed that dyads used several interactive moves that

2
In broad terms the proficiency of these students ranged from a B1 to a B2 level according to the Common European Framework of Reference for
Languages (CEFR).
3
We raise this point here in response to one of our anonymous reviewer’s concern about how the NS–NNS conversations in a study abroad context may
differ from any other NS–NNS conversations.
4
The term "move" has been used rather loosely in the interactionist literature, sometimes referring to a series of turns in which participants temporarily
step aside from the main line of the conversation to resolve a language difficulty, and sometimes referring to a single turn within such a sequence of turns.
In this paper we use the term "move" to refer to a series of turns, and we focus on analyzing how these asides serve as feedback for the non-native speaker.
The terms "routine" and "episode" have been used in the interactionist literature to refer to these departures from the main line of conversation. "Routine"
has been used to refer to a feedback type known as negotiation of meaning, whereas "episode" has been used to refer to other feedback types such as
recasts, prompts and explicit correction. We do not exclude the possibility of finding new types of feedback in the particular study abroad context of this
study, and therefore we prefer the broad term "move".
 M. Fernández-García, A. Martínez-Arbelaiz / System 42 (2014) 93–104 97

sometimes were initiated by the NS and sometimes by the NNS. Five moves were identified and codified. Two of them,
negotiation of meaning and recast, have been extensively researched in classroom and laboratory-based studies that suggest
that they have a facilitative role in the acquisition process (de la Fuente, 2002; Mackey, 1999; Smith, 2004). The other three
moves, though sometimes reported in the literature, have not received much attention, and thus no specific label or name has
been consistently used to refer to them. In the present study, we assigned the following names to these three moves: lexical
assistance, form assistance and completion. In what follows, we offer a description of the five moves together with repre-
sentative samples from our data.

Negotiation of meaning

Given that the aim of the present study is to identify interactional moves that help the learner to modify her output, we
analyze here output-pushing negotiation moves, that is, those that emerge when the NS is the one that signals that she does
not understand the NNS’s prior turn. Central to the construct of negotiation of meaning is the existence of “an overt indication
that understanding between participants has not been complete” (Varonis & Gass, 1985, p. 73).5
In identifying negotiation moves, following Varonis and Gass’ model (1985) we distinguished three obligatory primes
(Trigger, Indicator, Response) and an optional one (Reaction to Response) in a negotiation sequence. The example that follows
shows how the information provided in the NNS’s first turn in this sequence is not enough for the NS to interpret its meaning.
This prompts a series of turns until mutual understanding is achieved, marked by the last utterance that closes the negotiation
routine (the Reaction to Response).
Example of Negotiation of Meaning

T NNS: en esosta en Estados Unidos veintiuno es un grande cumpleaños

in the Unit – in the United States twenty one is a big (ill-formed) birthday
I NS: es un gran[ (.) cumpleaños[
is it a big[ birthday[
R NNS: grande cumpleaños
a big (ill-formed) birthday
I NS: aniversario o así[
anniversary or something like that[
R NNS: aniversario
anniversary
I NS: XX qué se celebra[
XX what do you celebrate[
R NNS: porque uh es el primero año que es legal
because uh it is the first year that is legal
NS: legal sí
legal yes
R NNS: para tomar
to drink (ill-form)
RR NS: ¡AAAAH los veintiún años de edad! Vale (.) vale (.) no es no es aniversario
AAAAH twenty one years of age! Ok (.) ok (.) it is not it is not an anniversary
(Dyad 2)

This example illustrates how both participants contribute to the interaction with turns that aim at achieving mutual
understanding. Thus, although the second turn, the NS’s first indicator appears to be a recast in the form of a verification
check that corrects the NNS’s error in the noun phrase “un grande cumpleaños”, subsequent indicators, “aniversario o
así[”, “X qué se celebra[”, reveal that the focus of the interaction is on meaning. The NS was not able to interpret the
NNS’s first turn, and kept trying to clarify it with the use of the latter mentioned indicators. Such indicators helped the NS
to keep the focus on the problematic item “un grande cumpleaños”, and to push the NNS to expand her output with key
information. That is, the meaning of the adjective “grande” was successfully clarified through a series of negotiated turns
that made explicit that the NNS was referring to her twenty-first birthday, the legal age for drinking alcohol in the United
States.
This exchange shows how negotiation of meaning addresses output issues exclusively in terms of problems with the
meaning of the NNS’s trigger utterance. There were, however, instances in the data in which verification checks were counted
as recasts because a contextualized analysis of the interaction made it clear that the NS’s feedback was addressing a form
problem rather than a comprehension problem.

5
This point of view is akin to the concept of repair, defined in the Conversation Analysis literature (see review in Hall, 2007) as “a fundamental unit of
talk that allows for catching and resolving troubles in speaking and hearing as they arise” in a conversation (p. 515).
98 M. Fernández-García, A. Martínez-Arbelaiz / System 42 (2014) 93–104

Recast

As in previous studies (Lyster, 1998; Lyster & Ranta, 1997), we codified as recasts the NS’s implicit provision of a correct
reformulation of all or part of a learner’s ill-formed utterance. The following examples show how the NS corrects the NNS by
extracting the non-target forms and providing the corresponding target model. In the first one, the NS extracts the prob-
lematic verb form “subó” and replaces it with the target one “subió”. In the subsequent turn, the NNS reacts by repeating the
last part of her sentence and incorporating the form “subió” provided by the NS.
Example of Recast 1

NNS y de allí al aeropuerto y él bajó también en Saint Saint Jean de Luz[ y uhm subó en nuestro bus
and from there to the airport and he also got off at Saint Saint Jean de Luz[ and uhm he got in (ill-formed) our bus
NS subió
he got in
NNS subió en nuestro bus a-
he got in our bus to-
NS ¡otra vez![
again![
(Dyad 8)

A similar correct reformulation of part of the learner’s erroneous turn is found in Example 2 below. In this case, the NS uses
a verification check to provide the targeted morpho-syntactic structure. This verification check, in contrast with the one
shown in the negotiation of meaning example, was counted as a recast and was assumed to be addressing a form problem.
Though the turn in itself is ambiguous, we took into account the surrounding context. Because the speakers continued the
regular line of the conversation without communication breakdowns once the correction was provided, we were more in-
clined to classify this type of feedback as a recast rather than an indicator.
Example of Recast 2

NNS: y yo quiero encontrar (.) lo

and I want to meet (.) him (ill-formed)
NS: encontrarte con él[
to meet him[
NNS: sí
yes
NS: en Florencia[
in Florence[
NNS: sí (.) pero es un poco difícil porque es-
yes (.) but it is a little difficult because it is-
NS: tienes que estudiar (.) tienes que estudiar aquí
you have to study (.) you have to study here
NNS: sí
yes
(Dyad 4)

Lexical assistance

We used the label Lexical Assistance to refer to a move that allows the NS to assist the NNS when she shows uncertainty or
explicitly states that she does not know how to express something, ranging from a lexical item to a whole phrase. This move is
initiated by the NNS with a turn that, with a varying degree of implicitness/explicitness, attempts to involve the NS in the
construction of her own output. The examples that follow illustrate how this move is shaped by the use of different linguistic
resources.
In the first example, the rising intonation used by the NNS with the word “tiempo”, preceded by the elongation of the last
vowel in the word “primero”, implicitly marks doubt and makes the NS aware of the point in discourse at which help may be
needed. In the following turn, the NS correctly interprets this request for help by providing the target word “vez”, which
would be the one needed in this context.
Example of Lexical Assistance 1

NNS sí primero[ tiempo[

yes first [ time (ill-formed)[
NS vez
time
 M. Fernández-García, A. Martínez-Arbelaiz / System 42 (2014) 93–104 99

NNS primera vez en Europa

first time in Europe
(Dyad 3)

In Example 2, the NNS initiates the move in a more explicit way, with a turn in which she uses the communicative strategy
of approximation (Faerch & Kasper, 1983; Liskin Gasparro, 1996)d“(.) la (.) no la cabeza pero-”, preceded and followed by
pausesdwhich allows the NS to interpret it as an appeal for help.
Example of lexical assistance 2

NNS: uh euskera castellano inglés (.) mucho más fácil para ella (.) yo gustaba aprender y ella entiende más (.) rápidamente
porque es (.) la (.) no la cabeza pero-
uh Basque Castilian English (.) much easier for her (.) I liked (ill-formed) to learn and she understands more (.) quickly
because it is (.) the (.) not the head but-
NS: la mente
the mind
NNS: la mente más abierto (.) sin preconcepciones yo pienso (.) porque cuando yo luegro luego aprender es como siempre
estoy pensando como qué es en inglés (.) bla bla bla
the mind more open (ill-formed) (.) without preconceptions I think (.) because when I then (ill-formed) then learn (ill-
formed) it is like I am always thinking like what it is in English (.) bla bla bla
(Dyad 6)

As the third example shows, in some cases the NNS opts for a very overt form of request for help (“cómo se dice “Ím used
to”[”), and directly asks how to translate an English word or phrase into Spanish.
Example of Lexical Assistance 3

NNS: sí (,) un poco raro pero (.) cómo se dice “Ím used to”[
yes (,) a little strange but (.) how do you say “Ím used to”[
NS: acostumbrada
acostumbrada
NNS: acostumbrada con lengua diferente de castellano (.) con inglés (.) bueno pero creo que sí (.) pero hay un festival de
Cataluña con los baladores <<dancers>>[
used to a language different from Castilian (ill-formed) (.) to English (ill-formed) (.) well but I think so (.) but there is a festival
in Catalonia with dancers (ill-formed) <<dancers>>[
(Dyad 5)

In Example 4, the Lexical Assistance move takes a very explicit form throughout several turns. The NNS resorts to the
shared knowledge of the L1 and L2 to make her appeal clear and to obtain the needed linguistic support from her conversation
partner. Similar to Example 3, in Example 4 we find the use of the direct question (“Cómo se dice þ English word/
expression[”) and, in addition, a circumlocution, that is, a description in the target language of the word being searched for.
Example of Lexical Assistance 4

NNS: cuando ella empiece la universidad ella quería ser una- cómo se dice una persona que quiere ser como
<<translator>>[ (.) una persona que quiere trabajar con gente que no puede hablar inglés[
when she began (ill-formed) at the university she wanted to be a- how do you say a person who wants to be like
<<translator>>[ (.) a person that wants to work with people that cannot speak English[
NS: mmm.
mmm.
NNS: entiendes[
do you understand[
NS: sí (.) pero no sé como
yes (.) but I do not know how
NNS: pero en inglés <<translator>> (.) sabes[
but in English <<translator>> (.) you know[
NS: Ah (.) traductor
Ah (.) traductor
NNS: traductor (.) y ella asistió una universidad en no sé en <<Mid West>> como <<Iowa>> (.) <<or>> un lugar (.)
sabes[ <<Iowa>>
translator (.) and she attended a university in I don’t know in <<Mid West>> like <<Iowa>> (.) <<or>> a place (.) you
know[ <<Iowa>>
(Dyad 7)
100 M. Fernández-García, A. Martínez-Arbelaiz / System 42 (2014) 93–104

Table 1
Frequency of discourse moves.

Dyads

Type of moves 1 2 3 4 5 6 7 8 Total %

Negotiation 2 12 4 4 2 5 6 1 36 15.8
Recast 2 34 16 18 2 3 7 1 83 36.4
Lexical assistance 0 7 13 3 18 10 10 2 63 27.6
Form ass. 1 3 5 4 3 0 2 1 19 8.3
Completion 3 5 2 2 1 2 6 6 27 11.8
Total 8 61 40 31 26 20 31 11 228 100
% 3.5 26.8 17.5 13.6 11.4 8.8 13.6 4.8 100

Form assistance

The label Form Assistance was given to an exchange in which the NS assists the NNS when she shows that she is searching
for a grammatical form, usually related to verb morphology. In the example that follows, the NNS uses twice the verb “estar”
(to be), first in the present tense and then, with rising intonation, in the imperfect tense. The NS interprets these signs as an
appeal for help, and reacts providing the context-appropriate verb tense “estaba”.
Example of Form Assistance

NNS sí (.) sí (.) vi [ está estaba[

yes, yes, I saw [ I is (ill-formed) I was [
NS estaba
estaba
NNS estaba en Las Ramblas XX de Barcelona (.) y (.) y fui al zoo[(.) <<zoo>>
I was in the Ramblas XX of Barcelona (.) and (.) I went to the zoo[ (.) <<zoo>>
(Dyad 5)

Completion

Though previously identified (Fernández García & Martínez Arbelaiz, 2007; García Mayo & Pica, 2000; Pica, Lincoln-Porter,
Paninos, & Linell, 1995; Swain, 1995), not much research has examined this move in the interactionist literature. In a
Completion, the NS predicts what the NNS is going to say and completes her turn. By anticipating the NNS’s output, the NS
provides a model of the target language that the NNS can then either ignore, confirm or incorporate in her subsequent turns.
As the example below shows, the NNS turn (“no conoz-no conozco los-”) is interrupted by the NS, who takes the floor to
make a guess at what she thinks the NNS is trying to convey (“los equipos”). In this particular example, the NNS acknowledges
that the NS’s guess was appropriate in the immediately subsequent turn (“los equipos sí”). The type of cues found in the
Lexical Assistance and Form Assistance moves (pauses, hesitation, rising intonation, etc.) are not necessarily present in the
Completion move. Furthermore, in the Completion move there is no specific request for help on a particular language item.
Rather, the NS finishes the NNS’s previous turn based on what she thinks her intended meaning is.
Example of Completion

NS sí (.) prefieres que gane la Real[ (laugh)

yes (.) do you prefer that la Real wins [ (laugh)
NNS uh no
uh no
NS te da igual[ (laugh)
you don’t care[ (laugh)
NNS no conoz- no conozco los-
I don’t kno- know the-
NS los equipos
the teams
NNS los equipos sí
the teams yes
(Dyad 2)

Once the discourse moves used to help learners “repair” or “push” their output were identified and described, their
relative frequency in our data was calculated in order to address our second research question. Table 1 shows the number of
moves that each dyad used as well as their relative frequency.
 M. Fernández-García, A. Martínez-Arbelaiz / System 42 (2014) 93–104 101

Table 2
Amount of modified output/uptake per move.

Moves Modified output

Correct N (%) Partially correct N (%) Incorrect N (%)

Negotiation 15 (41.6) 5 (13.9) 5 (13.9)
Recast 47 (56.6) 0 0
Lexical Ass. 39 (61.9) 0 0
Form Ass. 14 (73.7) 0 0
Completion 9 (33.3) 0 0

Recast, a move initiated by the NS that provides feedback on form accuracy, was most frequent (36.4%). However, a
look at the individual performance reveals that this is due to the high number of recasts in dyads 2, 3, and 4. The second
most frequently used move was Lexical Assistance (27.6%), a move that, in contrast to Recast, is initiated by the learner
and focuses on getting help while searching for the appropriate lexical item to encode a message. It can be observed that
there is a certain degree of variability of use across dyads, with one dyad not using any Lexical Assistance moves at all,
and half of them using at least 10 or more. The same focus on meaning is central to the next most frequent move,
Negotiation of Meaning (15.8%), which is present in all dyads and ranges in use from 1 to 12 instances. Recall that in this
study we only examine those moves that offer the NNS an opportunity to improve her output. For this reason,
only those negotiations in which the NS overtly signals that there is a problem with understanding the learner’s pre-
ceding utterance, that is, only those negotiations that address issues with the learner’s output, were included in the
data.6
The least frequently used moves were Completion (11.8%) and Form Assistance (8.3%). As the data for the Completion move
show, all the NSs stepped in to complete the learner’s turn (ranging from 1 to 6 times). Finally, all learners except for one
initiated a Form Assistance move in which they appealed for help with a target morphological form.
The third research question considers whether learners do actually notice the feedback provided by NSs. According to
research, chances are that when learners incorporate this feedback into their output, some degree of noticing takes place. We
have therefore assumed that when the learner incorporates the NSs’ feedback, she has paid attention to the linguistic features
that the NS used to encode particular target meanings. As the data show (see Table 2), all moves elicited modified output,
although not to the same extent. For example, if we look at the feedback for Form Assistance (73.7%) and Lexical Assistance
(61.9%), the rate of incorporation was well above 50%.
All the examples of Form Assistance and Lexical Assistance provided above to address the first research question show
incorporation of feedback. In the example of Lexical Assistance, presented again below, the NNS starts by showing hesitation
with respect to the word "tiempo" in the first turn of this move. The NS responds to this uncertainty by providing the needed
target word in the second turn. Finally, the last turn of this move shows how the NNS integrates the word "vez" in her
utterance
Example of Incorporation of Feedback

NNS sí primero[ tiempo[

yes first [ time (ill-formed)[
NS vez
time
NNS primera vez en Europa
first time in Europe
(Dyad 3)

The rate of incorporation of Recast, although a more frequent move in the data, was lower (56.6%) than that of Form
Assistance (73.7%) and Lexical Assistance (61.9%). As we will discuss in the next section, the precise nature of these three
moves (all involving the encoding of an already shared meaning), may have accounted for their high rate of incorpo-
ration (or uptake). In contrast, correct uptake was less frequent with Negotiation of Meaning (41.6%) and with
Completion (33.3%), probably because of their discursive nature, which encompasses the encoding of form through the
clarification and co-construction of an intended meaning. In some cases, then, the participants may be engaged in
clarifying the message to be conveyed without necessarily focusing on its grammatical-lexical accuracy. If that is the
case, there would be fewer opportunities to notice particular language forms and to incorporate them in a subsequent
turn. Other instances of Negotiation of Meaning and Completion, however, are more likely to result in noticing and the
incorporation of particular forms because the message to be conveyed by the NNS relies heavily on the accurate use of
such forms.

6
For a complete analysis of the negotiations of meaning that address learners’ comprehension of the NSs’ output, see Fernández García and Martínez
Arbelaiz (2007).
102 M. Fernández-García, A. Martínez-Arbelaiz / System 42 (2014) 93–104

Discussion and conclusions

The results of the first research question show that dyads used five types of moves to help modify or repair the learner’s
output. The data from this study also show that, in the specific context of these study abroad conversations, both the NS and
the learner (the NNS) initiated moves in an attempt to improve the learner’s output. Three of the moves, Lexical Assistance,
Form Assistance and Completion, which were rather frequent in our data, had not received much attention in previous
literature.
As the results of our second research question reveal, Recast, which by definition addresses form accuracy and is NS-
initiated, was the most frequent move. It is not surprising that the NSs in this study favored Recasts, a type of unobtrusive
feedback that blends easily into a conversation. This finding is in line with results from research on classroom feedback.
Unlike in those studies, however, we found that a large portion of moves was learner-initiated. In fact, one such learner-
initiated move, Lexical Assistance, was the second most frequent move and shows how the learner got actively involved in
gaining the NS’s cooperation to provide the needed lexical item. The remaining three moves were less frequent and ranged in
focus from grammatical accuracy (Form Assistance) to meaning-oriented difficulties (Negotiation of Meaning and
Completion).
The results of the first and second research questions indicate that the conversations in our study made it possible to use
moves that were initiated by the NS as well as by the NNS. In contrast to the typical classroom setting, in which the teacher
initiates the provision of feedback, in these dyadic conversations the learner actively seeks for ways to improve her output. In
what follows, we turn our attention to several interrelated factors that we believe favored the involvement of both partici-
pants and, in particular, of the learner in pushing her output toward the target norm.
Most of the literature on feedback has examined how the teacher reacts to a learner’s error, but as Ellis et al. (2001b) point
out, not much attention has been paid to moves that anticipate or identify a possible area of concern in a learner’s output
(preemptive feedback). While Ellis et al. feel that there is a need for research in this area, we also think that, in particular,
learner-initiated moves may be absent in the literature on feedback simply because they may not emerge as readily in
classroom contexts. In their study, as in ours, learner-initiated feedback played an important role in the interactions and led to
a high level of successful uptake. It is interesting to note that while theirs is a classroom study and ours is not, there are certain
commonalities in terms of the type of NNS participants that may explain their active role in getting help while producing
pushed output. We believe that, as in Ellis et al., the adults in our study “were more motivated and more cognitively able to
attend to form than Lyster’s immersion students” (pp. 311–312). In other words, the learners’ willingness to participate in
these language exchanges would explain their proactiveness in seeking feedback.
A second factor that may have favored learner-initiated moves relates to the particular context of these conversations. The
role of such conversations can be better explained if we view study abroad, as suggested by Collentine and Freed (2004), as “a
hybrid communicative learning context” (p. 156) that offers learners academic opportunities to study the L2 in a formal
setting while immersed in the target culture. Furthermore, we believe that study abroad can be conceptualized as a con-
tinuum of contexts that ranges from most formal to most informal. That is, study abroad most likely encompasses a variety of
opportunities for learners to converse with NSs, from classroom interactions, which may involve different degrees of focus on
form, to out-of-class interactions of a predominantly social/informational nature, such as the “service encounters” described
in Shively (2011). The conversations analyzed in this study would fall somewhere in the middle of the continuum given that
they are shaped by a learning purpose (i.e., to improve language skills) but also by roles that are more egalitarian than those
played by teacher and students in the classroom.
Furthermore, the self-perception of the participants in this hybrid context can explain the relative abundance of learner-
initiated feedback. Given that interactants volunteered to exchange conversation sessions that alternated between English
and Spanish, the role of the “expert” was played by a different partner each time the target language was changed. Thus, the
two conversational partners had an opportunity to display their second language identity (Benson et al., 2012) as well as their
language “expert” identity. Unlike classroom interactions or service encounters where the study abroad student is perma-
nently positioned as a second language learner or user, in these exchanges the role of learner is particularly fluid since it is
alternatively played by the dyad members. In this sense, it is likely that this shared identity as language learners allowed for a
more equally distributed management of the conversation than that found in the classroom, with both the learner (the NNS)
and the “expert” (the NS) taking the lead in assigning the topic and focusing on form. Additional social and contextual factors
such as the informal out-of-class setting and the age of the participants may have also contributed to the particular nature of
the discourse analyzed.
Not only was there a high rate of learner-initiated moves but also, as the results for the third research question indicate, a
high rate of incorporation of the feedback. As Ellis et al. (2002) point out, it is logical to assume that when the learner draws
attention to a gap in her knowledge, such a gap must be significant to her and so feedback becomes salient. In fact, Ellis et al.
cite the study by Slimani (1989) that showed that learners had an advantage in remembering new items in episodes that they
themselves had initiated. In our study, learners actively asked for help in a variety of ways: through rising intonation, using
approximations, asking for translations, etc. Moreover, we observed that these moves led to immediate incorporation, since
learners were highly involved in completing their output.
Referring to the classroom context, Ellis et al. (2002) also underscore the importance of teacher feedback, assuming focus
on form is as crucial in language development as indicated in the literature. An advantage of the conversations described in
this study is that they offer access to a language expert that does not play the role of the teacher in the traditional sense. This
 M. Fernández-García, A. Martínez-Arbelaiz / System 42 (2014) 93–104 103

expert is able to provide focus on form tailored to the conversational needs of the learner in an environment where the
anxiety associated with speaking in a foreign language is minimized. In sum, we suggest that the NS identity, both as learner
and as expert, brought about learner-initiated moves that led to uptake.
If one of the advantages of study abroad is to facilitate interaction with a variety of NSs in different situations and with
different purposes, in this study we have shown that out-of-class conversations that still retain a language learning goal
include many of the discourse features identified in the SLA literature as conducive to acquisition. In addition, these ex-
changes may be a way of ensuring that the experience abroad entails some meaningful contact with the host community,
since according to Kinginger (2010: 225), “our students [U.S. students] are at increasing risk of failing to notice their own
ignorance of the communities they join through study abroad”. To this regard, specific course requirements, such as the ones
described in Cadd (2012), that include exchanges with members of the host community may help to overcome the limitation
noted by Kinginger. Furthermore, students may benefit not only from courses that incorporate "micro-tasks" that involve
interactions with members of the host community but also from exchanges with NSs that are sustained over an extended
period of time, such as the ones analyzed in our study. As we have already pointed out, an advantage of the specific in-
teractions of this study is that the conversations kept a pedagogical focus and at the same time allowed for a balanced
contribution in turn taking, topic initiation and utterance construction.
This study explored how learners may get opportunities to improve their output in a specific study abroad conversation
context. It aimed to contribute data to a scarce body of literature on study abroad conversations from an interactionist
perspective. Obviously, this is but one type of a whole array of possibilities of conversations abroad, and to a certain extent still
linked to an academic setting and goal. The results cannot be generalized to contexts in which the NNS speaks with other NSs,
i.e., people who are not involved in language learning, such as the neighbors, the landlady, the local bar attendant, etc. There
is, therefore, a need for primary data from other conversations that learners engage in while abroad. Although there are a few
studies based on students abroad interacting with their families (McMeekin, 2006; Wilkinson, 2002), during service en-
counters (Shively, 2011), etc. we still need more research on the constellation of relationships or the social networks students
develop abroad (such as the work that is currently being carried out by the LANGSNAP project, Mitchell et al., 2013) in order to
ascertain their particular impact on language acquisition. The analysis of data coming from a wide variety of settings and
interactants will allow us to draw a more complete picture of how study abroad conversations can contribute to the process of
L2 acquisition.

Appendix A

Transcription key

Symbol Meaning
NS Native speaker
NNS Non-native speaker
CAPITALS Emphasis
(laugh) Extra information
XX Inaudible or unclear
(.) Micropause
[ Rising intonation
- Interrupted speech
<< >> English pronunciation

References

Ammar, A., & Spada, N. (2006). One size fits all? Recasts, prompts, and L2 learning. Studies in Second Language Acquisition, 28, 543–574.
Benson, P., Barkhuizen, G., Bodycott, P., & Brown, J. (2012). Study abroad and the development of second language identities. Applied Linguistics Review, 3-1,
173–193.
Cadd, M. (2012). Encouraging students to engage with native speakers during study abroad. Foreign Language Annals, 45(2), 229–245.
Collentine, J. (2009). Study abroad research: finding, implications, and future directions. In M. H. Long, & C. J. Doughty (Eds.), The handbook of language
teaching (pp. 218–233). Oxford: Wiley-Blackwell.
Collentine, J., & Freed, B. (2004). Learning context and its effect on second language acquisition. Studies in Second Language Acquisition, 26, 153–171.
de la Fuente, M. J. (2002). Negotiation and oral acquisition of Spanish L2 vocabulary: the roles of input and output in the receptive and productive
acquisition of words. Studies in Second Language Acquisition, 24, 81–112.
DuFon, M., & Churchill, M. (2006). Language learners in study abroad contexts. Second language acquisition. Clevedon, U.K.: Multilingual Matters.
Ellis, R. (2006). Researching the effects of form-focused instruction in L2 acquisition. AILA Review, 19(1), 18–41.
Ellis, R., & Sheen, Y. (2006). Reexamining the role of recasts in second language acquisition. Studies in Second Language Acquisition, 28, 575–600.
Ellis, R., Basturkmen, H., & Loewen, S. (2001). Learner uptake in communicative ESL lessons. Language Learning, 51(2), 281–318.
Ellis, R., Basturkmen, H., & Loewen, S. (2001). Preemptive focus on form in the ESL Classroom. TESOL Quarterly, 35(3), 407–432.
Ellis, R., Basturkmen, H., & Loewen, S. (2002). Doing focus-on-form. System, 30, 419–432.
Ellis, R., Loewen, S., & Erlam, R. (2006). Implicit and explicit corrective feedback and the acquisition of L2 grammar. Studies in Second Language Acquisition,
28, 339–368.
Faerch, C., & Kasper, G. (1983). Strategies in interlanguage communication. London, UK: Longman.
Fernández, García M., & Martínez Arbelaiz, A. (2007). Negotiated interaction in the study abroad context: gauging the opportunities for interlanguage
development. In F. Boers, J. Darquennes, & R. Temmerman (Eds.), Multilingualism and applied comparative linguistics, vol. 1: Comparative considerations in
second and foreign language instruction (pp. 41–66). London: Cambridge Scholars Press.
104 M. Fernández-García, A. Martínez-Arbelaiz / System 42 (2014) 93–104

García-Mayo, M. P., & Pica, T. (2000). L2 learner interaction in a foreign language setting: are learning needs addressed? International Review of Applied
Linguistics, 38, 35–58.
Gass, S. M. (1997). Input, interaction and the second language learner. Mahwah, NJ: Erlbaum.
Gass, S. M., Mackey, A., & Ross-Feldman, L. (2005). Task-based interactions in classroom and laboratory-settings. Language Learning, 55, 575–611.
Hall, J. K. (2007). Redressing the roles of correction and repair in research on second and foreign language learning. The Modern Language Journal, 91(4), 511–
526.
Iwashita, N. (2003). Negative feedback and positive evidence in task-based interaction. Differential effects on L2 development. Studies in Second Language
Acquisition, 25, 1–36.
Kinginger, C. (2008). Language learning in study abroad: case histories of Americans in France. In Modern Language Journal Monograph Series (Vol. 1). Oxford:
Blackwell.
Kinginger, C. (2010). American students abroad: negotiation of difference? Language Teaching, 43(2), 216–227.
Krashen, S. (1981). Second language acquisition and second language learning. Oxford: Pergamon.
Lafford, B. A. (2004). The effect of context of learning on the use of communication strategies by learners of Spanish as a second language. Studies in Second
Language Acquisition, 26, 201–226.
Liskin-Gasparro, J. E. (1996). Circumlocution, communication strategies, and the ACTFL Proficiency Guidelines: an analysis of student discourse. Foreign
Language Annals, 29, 317–330.
Loewen, S. (2004). Uptake in incidental focus on form in meaning-focused ESL lessons. Language Learning, 54(1), 153–158.
Loewen, S., & Philp, J. (2006). Recasts in the adult English L2 classroom: characteristics, explicitness and, effectiveness. The Modern Language Journal, 90(4),
536–556.
Long, M. H. (2007). Problems in SLA. Mahwah, NJ: Erlbaum.
Lyster, R. (1998). Recasts, repetition, and ambiguity in L2 classroom discourse. Studies in Second Language Acquisition, 20(1), 55–81.
Lyster, R. (2004). Differential effects of prompts and recasts in form-focused instruction. Studies in Second Language Acquisition, 26, 399–432.
Lyster, R., & Ranta, L. (1997). Corrective feedback and learner uptake: negotiation of form in communicative classrooms. Studies in Second Language
Acquisition, 19, 37–66.
Lyster, R., & Saito, K. (2010). Oral feedback in classroom SLA. A meta-analysis. Studies in Second Language Acquisition, 32, 265–302.
Mackey, A. (1999). Input, interaction, and second language development: an empirical study of input, interaction, and second language development.
Studies in Second Language Acquisition, 21, 557–588.
Mackey, A. (2006). Feedback, noticing and instructed second language learning. Applied Linguistics, 27(3), 405–430.
Mackey, A., & Philp, J. (1998). Conversational interaction and second language development: recasts, responses and red herrings. The Modern Language
Journal, 82(3), 338–356.
McMeekin, A. (2006). Negotiation in a Japanese study abroad setting. In M. DuFon, & M. Churchill (Eds.), Language learners in study abroad contexts. Second
language acquisition (pp. 177–202). Clevedon, UK: Multilingual Matters.
Mitchell, R., Tracy-Ventura, N., McManus, K., Richard, L., Romero de Mills, P., & Dewaele, J. (April 10–12, 2013). The influence of social networks, personality
and placement type on language learning during residence abroad: Preliminary findings of the LANGSNAP project. Paper delivered at the Conference on
Residence Abroad, Social Networks, and Language Learning, University of Southampton.
Oliver, R. (1995). Negative feedback in child NS/NNS conversation. Studies in Second Language Acquisition, 17, 459–483.
Oliver, R. (2000). Age differences in negotiation and feedback in classroom and pairwork. Language Learning, 50, 119–151.
Pica, T. (1988). Interlanguage: adjustments as an outcome of NS-NNS negotiated interaction. Language Learning, 38, 45–73.
Pica, T. (1994). Research on negotiation: what does it reveal about second-language learning conditions, processes, and outcomes? Language Learning, 44,
493–527.
Pica, T., Lincoln-Porter, F., Paninos, F., & Linnell, J. (1995). What can second language leaners learn from each other? Only their researcher knows for sure.
Working Papers in Educational Linguistics, 11(1), 1–36.
Schmidt, R. (1990). The role of consciousness in second language learning. Applied Linguistics, 11, 129–158.
Schmidt, R. (2001). Attention. In P. Robinson (Ed.), Cognition and second language instruction (pp. 3–33). New York: Cambridge University Press.
Schwartz, B. D. (1993). On explicit and negative data effecting and affecting competence and linguistic behaviour. Studies in Second Language Acquisition, 15,
147–164.
Sheen, Y. (2004). Corrective feedback and learner uptake in communicative classrooms across instructional settings. Language Teaching Research, 8(3), 263–
300.
Sheen, Y. (2006). Exploring the relationship between characteristics of recasts and learner uptake. Language Teaching Research, 10(4), 361–392.
Shively, R. L. (2011). L2 pragmatic development in study abroad: a longitudinal study of Spanish service encounters. Journal of Pragmatics, 43, 1818–1835.
Slimani, A. (1989). The role of topicalization in classroom language learning. System, 17, 223–234.
Smartt, J. T., & Scudder, R. R. (2004). Immersion study abroad in Mexico: using repair behaviors to assess proficiency changes. Foreign Language Annals, 37(4),
592–599.
Smith, B. (2004). Computer-mediated negotiated interaction and lexical acquisition. Studies in Second Language Acquisition, 26, 365–398.
Swain, M. (1985). Communicative competence: some roles of comprehensible input and comprehensible output in its development. In S. M. Gass, & C. G.
Madden (Eds.), Input in second language acquisition (pp. 235–253). Rowley, MA: Newbury House.
Swain, M. (1995). Three functions of output in second language learning. In G. Cook, & B. Seidelhofer (Eds.), Principle and practice in applied linguistics:
Studies in honor of H.G. Widdowson (pp. 125–144). Oxford: Oxford University Press.
Varonis, E. M., & Gass, S. M. (1985). Non-native/non-native conversations: a model for negotiation of meaning. Applied Linguistics, 6(1), 71–90.
White, L. (1991). Adverb placement in second language acquisition: some effects of positive and negative evidence in the classroom. Second Language
Research, 7, 133–161.
Wilkinson, S. (2002). The omnipresence classroom during summer study abroad: American students in conversation with their French hosts. The Modern
Language Journal, 86(2), 157–173.
Yang, Y., & Lyster, R. (2010). Effects of form-focused practice and feedback on Chinese EFL learners’ acquisition of regular and irregular past tense forms.
Studies in Second Language Acquisition, 32, 235–263.