Input, Interaction and Output

An Overview

Susan Gass and Alison Mackey

Michigan State University / Georgetown University

This paper presents an overview of what has come to be known as the Interaction
Hypothesis, the basic tenet of which is that through input and interaction with
interlocutors, language learners have opportunities to notice differences between
their own formulations of the target language and the language of their conver-
sational partners. They also receive feedback which both modifies the linguistic
input they receive and pushes them to modify their output during conversation.
This paper focuses on the major constructs of this approach to SLA, namely,
input, interaction, feedback and output, and discusses recent literature that ad-
dresses these issues.

Introduction

This paper presents an overview of what has come to be known as the Interaction
Hypothesis, the basic tenet of which is that through input and interaction with inter-
locutors, language learners have opportunities to notice differences between their own
formulations of the target language and the language of their conversational partners.
They also receive feedback which both modifies the linguistic input they receive and
pushes them to modify their output during conversation. This paper focuses on the
major constructs of this approach to SLA, namely, input, interaction, feedback and
output, and discusses recent literature that addresses these issues.
We begin by noting that the Interaction Hypothesis subsumes aspects of the In-
put Hypothesis (Krashen 1982, 1985) and the original Output Hypothesis (Swain 1985,
1995). As we explain in Gass and Mackey (in press), the Interaction Hypothesis has
been characterized and referred to in various ways, evolving over the years to the point
that current research often refers to it as the interaction ‘approach’ or as a ‘model’ (see,
for example, Block’s 2003 discussion of the input, interaction, output model). We re-
turn to these various characterizations at the end of this paper.
In simple terms, the interaction approach considers exposure to language (input),
production of language (output), and feedback on production (through interaction) as
constructs that are important for understanding how second language learning takes

AILA Review 19 (2006), 3–7.

issn 1461–0213 / e-issn 1570–5595 © John Benjamins Publishing Company
4 Susan Gass and Alison Mackey

Negotiation

Attention
Learning

Noticing

Feedback Recasts

Figure 1. Model of Interaction and Learning

place. Many researchers also make reference to learners’ attentional processes. Gass
(2003), for example, argues that interaction research “takes as its starting point the
assumption that language learning is stimulated by communicative pressure and ex-
amines the relationship between communication and acquisition and the mechanisms
(e.g., noticing, attention) that mediate between them” (p. 224). Long (1996) makes a
similar claim, proposing that “environmental contributions to acquisition are medi-
ated by selective attention and the learner’s developing L2 processing capacity, and that
these resources are brought together most usefully… during negotiation for meaning.
Negative feedback obtained… may be facilitative of L2 development” (p. 414). Swain’s
(1993) emphasis is on output: “Learners need to be pushed to make use of their re-
sources; they need to have their linguistic abilities stretched to their fullest; they need
to reflect on their output and consider ways of modifying it to enhance comprehensi-
bility, appropriateness, and accuracy” (pp. 160–161).
Figure 1 characterizes the components of interaction. During interaction, there
are instances in which the conversational participants negotiate meaning due to a lack
of understanding. As part of this negotiation, learners receive feedback on their pro-
duction, thereby potentially drawing attention to linguistic problems and leading them
to notice gaps between their production and the target language. Long (1996) defines
negotiation for meaning in the following way:
The process in which, in an effort to communicate, learners and competent speak-
ers provide and interpret signals of their own and their interlocutor’s perceived
comprehension, thus provoking adjustments to linguistic form, conversational
structure, message content, or all three, until an acceptable level of understanding
is achieved (p. 418).
 Input, Interaction and Output 5

To explain the mechanisms involved in learning language as a result of this, Long notes
that negotiation for meaning connects “input, internal learner capacities, particularly
selective attention, and output in productive ways” (pp. 451–452).
Having outlined the basic concepts involved in the interaction approach to second
language acquisition, we turn now to a discussion of each of the specific constructs
necessary for understanding how interaction and learning are linked.

Input

Input refers to language that is available to the learner through any medium (listen-
ing, reading, or gestural in the case of sign language). All theories of second language
learning recognize the significance of input as a basic component in the acquisition
process. Language is not learned in a vacuum; learners need “raw data” to serve as
linguistic evidence which they can use to formulate hypotheses about the second lan-
guage system.
Even though all approaches to SLA recognize the significance of input, it is not the
case that they all attribute the same weight to it. For example, researchers of Universal
Grammar view input as a trigger that interacts with an innate system and/or the native
language to promote language learning. Researchers who take a different perspective
include those who consider the role of frequency of the input (see Ellis 2002), for
example, the Associate-Cognitive Creed (see the chapter by N. Ellis), which suggests
that language constructions are learned “through using language, engaging in com-
munication” (p. 101) — in other words, through language use as well as exposure to
input. Ellis further argues that “an individual’s creative linguistic competence emerges
from the collaboration of the memories of all the utterances in their entire history of
language use and from the frequency-biased abstraction of regularities within them”
(p. 101, this volume). With frequency constituting a basis for understanding language
learning, input quite clearly has an essential and central position.
Since input forms the positive evidence that learners use as they construct their
second language grammars, many researchers have attempted to characterize and
describe the input that learners receive. One of the earliest studies in this regard is
that of Ferguson (1971), who investigated what came to be known as foreigner talk,
the language addressed to non-native speakers. Ferguson noted common phenom-
ena amongst a range of speech types, although, it should be noted that his data were
based on asking people what they thought they would say in certain situations, rather
than what they actually did say. Specifically, in language directed toward linguistically
less competent individuals (e.g., young children, learners of a language), he noted nu-
merous adjustments that native speakers make to their speech in all areas, including
pronunciation, grammar, and lexis. With regard to input, the following are examples
taken from Ferguson’s original work, exemplifying language addressed to learners of
Spanish and Arabic.
6 Susan Gass and Alison Mackey

(1) Language addressed to non-native speakers of Spanish

Standard Spanish Learner-directed
yo veo al soldado mi ver soldado
I I see DO soldier me to see soldier
(DO = direct object marker)
(2) Language addressed to non-native speakers of Arabic
Standard Arabic Learner-directed
ya’rif ya’rif
he knows he/she/I/you know
In example (1), in speech directed at a non-native speaker of Spanish, an infinitive is
used in place of a fully conjugated verb form, and the direct object marker is not used.
In example (2), the form ya’rif, which is generally restricted to the third-person mascu-
line singular, is extended to many forms. Further examples of modified speech can be
found in overviews provided in Gass and Selinker (2001) and Hatch (1983).
Besides simplifications such as those shown above, there are other forms of input
modification. For example, speech can be elaborated, providing learners with more
detailed semantic information. An illustration of this is shown in (3) below. In this ex-
change, the non-native speaker (NNS) indicates a lack of understanding (Pardon me?),
to which the native speaker (NS) replies by elaborating upon her original response and
by changing the pronoun, them, to a full noun phrase, my eating habits.
(3) Elaboration (Gass and Varonis 1985; emphasis in original)
NNS: How have increasing food costs changed your eating habits?
NS: Well, I don’t know that it’s changed THEM. I try to adjust.
NNS: Pardon me?
NS: I don’t think it’s changed MY EATING HABITS.

Interaction

Conversational interaction, along with input, is central to the interaction approach to

understanding SLA. Interaction is believed to be helpful in exchanges between/among
learners or between/among learners and native speakers (or fluent speakers) of the
language being learned where language is negotiated, or feedback is provided. Exam-
ple (4) illustrates a typical negotiation routine. In this example, the two participants
(a child NNS and an adult NS) are performing a spot-the-difference task in which
they are describing similar pictures to one another with the goal of finding differences
between their pictures.
(4) Negotiation (Mackey and Oliver 2002)
NS: I have a girl with a hand up like that.
NNS: But nothing?
NS: But there’s nothing in her hand?
 Input, Interaction and Output 7

NNS: I have a bang in her hand.

NS: Oh, do you mean like a hammer?
NNS: Um, yeah.
NS: Does she have a hammer in her hand?
Here, the NNS asks a reduced question (But nothing?), which the NS attempts to confirm
and expand upon in the following turn, using a targetlike question form. After the NNS
uses the word bang, a negotiation sequence ensues in which the NS questions what sort
of object is being referred to (Oh, do you mean like a hammer?). Throughout the ex-
ample, the NNS receives feedback on her production as well as models of more targetlike
language, and the interlocutors eventually reach an acceptable level of understanding.

Feedback

Feedback is an important feature of interaction which can be manifested in a number

of ways. It is generally considered to be a form of negative evidence, or information that
a particular utterance is deviant vis-à-vis target language norms. The interaction hy-
pothesis is primarily concerned with reactive feedback — that is, feedback that occurs
as a reaction to some linguistic problem. It can come from many individuals, including
not only teachers, but also other learners or native speakers of a language.
Feedback can be viewed on a continuum between explicit and implicit. A straight-
forward example of explicit feedback would be a pedagogical situation in which a
teacher says something like, “No, we don’t say x in English; we say y because x is a
noun.” This instance of feedback provides not only an explicit statement that there is
an error, but also information about what the correct form should be and metalin-
guistic or rule-based information as well (i.e., “x is a noun”). At the other end of the
continuum is implicit negative feedback, which is illustrated through the negotiation
sequence in example (4) above. Through this interaction, the learner recognizes that
there is an error and also recognizes what the correct form should be, but this process
is not explicit.
In defining and operationalizing the components of negotiation, Long (1983) in-
cluded confirmation checks, clarification requests, and comprehension checks. Con-
firmation checks are “any expressions… immediately following an utterance by the
interlocutor which are designed to elicit confirmation that the utterance has been
correctly heard or understood by the speaker” (Long 1983: 137). Example (5) shows
a confirmation check in which NNS 1 confirms that she has heard NNS 2 correctly
regarding the girl’s money.
(5) Confirmation check (Gass, Mackey, and Ross-Feldman 2005)
NNS 1: ¿Ella tiene mucho dinero, or …?
She has a lot of money, or …?
NNS 2: No mucho dinero.
Not much money.
8 Susan Gass and Alison Mackey

→ NNS 1: ¿No mucho dinero?

Not much money?
NNS 2: Sí.
Yes.
A clarification request is “any expression… designed to elicit clarification of the inter-
locutor’s preceding utterance(s)” (Long 1983: 137), as illustrated in example (6). Here,
NNS 2 requests clarification regarding a girl in blue mentioned by NNS 1 by asking
whether she is wearing a blue dress.
(6) Clarification request (Gass et al. 2005)
NNS 1: No veo un pelota pero yo tengo un chica en azul
I don’t see a ball but I have a girl in blue
→ NNS 2: ¿Vestido azul?
Blue dress?
NNS 1: Sí
Yes.
A comprehension check is an attempt “to anticipate and prevent a breakdown in com-
munication” (Long 1983: 136), as in example (7), where NNS 1 includes as part of his
turn a preemptive check as to whether NNS 2 would like him to repeat what he has
just said.
(7) Comprehension check (Gass et al. 2005)
NNS 1: La avenida siete va en una dirección hacia el norte desde la
→ calle siete hasta la calle ocho. ¿Quieres que repita?
Avenue Seven goes in one direction towards the north from Street
Seven to Street Eight. Do you want me to repeat?
NNS 2: Por favor.
Please.
NNS 1: La avenida seven, uh siete, va en una dirección hacia el norte desde
la calle siete hasta la calle ocho.
Avenue Seven, uh Seven, goes in one direction towards the north from
Street Seven to Street Eight.
Another frequent form of implicit feedback is what is known as a recast. This interac-
tional move has been defined in a number of ways. In the literature on child language
acquisition, Bohannon, Padgett and Nelson (1996) define it as an utterance in which
an interlocutor “expands, deletes, permutes, or otherwise changes the [non-native ut-
terance] while maintaining significant overlap in meaning” (p. 551). Similarly, in the
second language literature, what is common to most definitions of recasts is that there is
some modification of the original erroneous utterance without a concomitant change in
meaning, as shown in the definitions below. An example of a recast is provided in (8).
Long (1996: 434): Recasts are utterances that rephrase a child’s utterance by
changing one or more sentence components (subject, verb, or object) while still
referring to its central meanings.
 Input, Interaction and Output 9

Lyster and Ranta (1997: 46): Recasts involve the teacher’s reformulation of all or
part of a student’s utterance minus the error.
(8) Recast (Philp 2003)
NNS: Why he want this house?
→ NS: Why does he want this house?
Learners’ original utterances may have one or multiple errors, which may be fairly easy
to correct, as with the morphological (agreement) error in (9), or more complex, as in
(10), in which the forms of words as well as the word order are changed. Regardless of
the specific definition, recasts reformulate an error or errors either partially or fully,
which may or may not be taken as a correction by the learner.
(9) Morphological change (Leeman 2003)
NNS: En la mesa hay una taza rojo
on the table there is a cup (f.) red (m.)
There is a red cup on the table.
NS: Um hmmm, una taza roja. ¿Qué mas?
a cup (f) red (f) what more
Um hmmm, a red cup. What else?
(10) Multiple changes (Lyster and Ranta 1997)
St = student; T4 = teacher
St: Parce que il veut juste lui pour être chaud.
Because he wants just him for to be warm.
T4: Oh. Quelqu’un qui veut juste avoir la chaleur pour lui-même.
Oh. Someone who wants just to have the warmth for himself.
There is some debate about the source of the effectiveness of recasts in bringing about
interlanguage change. On the one hand, as several researchers (including Leeman
2003) have discussed, they provide positive evidence by containing a correct target
language model; on the other hand, as recasts occur adjacent to errors, it has also
been argued that they can serve as a form of negative evidence (Saxton 1997, 2005).
Researchers (e.g., Lyster 1998) have also pointed out that regardless of an interlocutor’s
intent, a recast can be interpreted by a learner in at least two different ways: as a correc-
tion (negative evidence) or simply as an alternative way of saying something, with no
implication that the original utterance was incorrect. In addition, learners may or may
not respond to recasts, thus reducing their involvement in that part of the discourse. In
negotiation episodes, by contrast, learners’ participation is more active, increasing the
possibility that they will be aware of problems in need of resolution.
Difficulties in interpreting feedback as it is intended are not limited to recasts. Il-
lustrative of this is the excerpt in (11), in which the first clarification request appears
to fall short of the mark, with the student making no change at first. As the storytelling
continues, however, the student seems to be more sensitive to past tense forms and
self-corrects in the last turn following another clarification request.
0 Susan Gass and Alison Mackey

(11) Clarification request unrecognized as negative evidence (Takashima 1995)

S = student; T = teacher
S: One day, the fairy, sting the magic wand to Cinderalla.
T: Sorry?
S: One day, the fairy sting the magic wand to Cinderalla.
T: OK.
S: Cinde, ah, Cinderaella changed into, the beautiful girl. (Laugh) Ah, and,
the, Cin, Cinderella wen Cinderella went to the palace by coach. The, the
prince fall in love at a first glance.
T: Sorry?
S: Ah, the prince fall in, falled falled in love Cinderella at a first glance.
And they dance, they danced … Ah, Cin, Cinderella have, Cinderella
have to go home.
In this example, repeated clarification requests may have alerted the learner to a problem,
possibly even leading him to change his original output “fall in love” to “falled in love.”
As well as negotiation and recasts, the interaction literature has also focused on
interactions in which learners attend to language in some way, as in language-related
episodes (LREs) and focus on form episodes (FFEs). Language-related episodes have
been the focus of much recent research by Swain and her colleagues, and are defined
by Swain and Lapkin (1998) as “any part of a dialogue in which students talk about the
language they are producing, question their language use, or other- or self-correct” (p.
70). An example of this is provided in (12), which involves a discussion of the Spanish
words for pairs and bird.
(12) Language-related episode (Gass et al. 2005)
NNS 1: ¿Trece? Tengo uh diecisiete… ¿Cuántos parejas?
Thirteen? I have uh seventeen… How many pairs?
NNS 2: ¿Parejas de amores?
Pairs of lovers?
→ NNS 1: ¿Qué es parejas?
What is ‘parejas’?
NNS 2: Pairs.
NNS 1: Oh.
NNS 2: No tiene el merry-go-round.
It doesn’t have the merry-go-round.
→ NNS 1: Hmm. ¿Cómo se dice bird?
Hmm. How do you say bird?
NNS 2: Pájaro
Bird
NNS 1: Oh! That’s what I was trying to say. ¿Cuántos pájaros?
How many birds?
 Input, Interaction and Output 

input

Input type positive evidence

(input & models) negative evidence

Feedback types authentic modified preemptive reactive

simplified elaborated grammar rules explicit implicit

(before incorrect use)

overt error correction recasts

communication
breakdown

simple complex

Figure 2. Adapted from Long and Robinson (1998: 19). Schematization of types of input
and feedback

The features of interaction that we have presented here are, of course, not mutually
exclusive. In fact, within a single interactional sequence or exchange, more than one
is often present, as in example (12) above, in which the LREs seem to result from
NNS 2’s clarification request (¿Parejas de amores?). Figure 2 (from Long & Robinson
1998) represents an overview of input and feedback types. Figure 3 (from Gass 1997)
represents a schematization of the function of negative evidence, which can occur in
the form of negotiation (including, for example, LREs and recasts) or other types of
correction. These forms of negative evidence can lead learners to notice errors in their

Negative Evidence (feedback)

Negotiation Other types of correction

Notice error

Search input

Input available Input not available

(Confirmatory/nonconfirmatory)
Figure 3. Function of negative evidence (Source: Gass 1997).
2 Susan Gass and Alison Mackey

own speech, but they do not necessarily provide information on how to correct the er-
rors. In general, the argument is that interaction provides a forum for feedback, which
serves to alert learners to problems providing them with opportunities to focus their
attention on language. That is, interaction may prime learners to “search” for more in-
formation, to be more sensitive to future input (e.g., uses of a word, structure, pronun-
ciation, spelling), or to be more aware of their hypotheses about language. Subsequent
input is either available or not; if available, it might serve to confirm an hypothesis or
disconfirm one, in which case the learner may form another, with the cycle repeating
until an hypothesis is confirmed. During this process, learners sometimes also modify
their output as a way of testing their hypotheses.

Perceptions about feedback

We view feedback as double-pronged in the sense that the intent of the provider of
feedback and the interpretation of the receiver of feedback are both important. From
the perspective of second language acquisition, interpretation is a crucial piece of the
puzzle in understanding interaction-driven learning. A major assumption, which may
be warranted in some circumstances more than in others, is that feedback is perceived
by the learner in the way it was intended by the provider. For example, if feedback is
given on a lexical error, it is desirable for it to be understood as lexical feedback by the
learner. That is, the learner should understand the target of the feedback. In relation
to this, as suggested above in the discussion of recasts, it is important to note that the
intent of a feedback provider can be multifaceted; for instance, an interlocutor may
intend to correct an error and to continue the conversation at the same time.
In investigating the relationship between intent and interpretation of feedback,
Mackey, Gass, and McDonough (2000) found that learners’ perceptions differed ac-
cording to feedback type and focus, with lexical and phonological feedback tending to
be perceived more accurately than feedback on morphosyntax. The study employed a
stimulated recall procedure (see Gass & Mackey 2000) involving two parts: a video-
taped interaction task between a language learner and a more competent speaker of
the target language, followed immediately by a viewing of the video with questions re-
garding the thought processes of the learner during the interaction. Examples (13) and
(14), taken from Mackey et al. (2000), indicate a match and mismatch, respectively,
between the intended targets of the feedback and the learners’ perceptions.
(13) Match between intent and interpretation: Morphosyntactic feedback
perceived as such (Mackey et al. 2000)
NNS: Three key.
NS: Three?
NNS: Key er keys.
Recall: After “key” again, I make a little effort to say “keys” because you
have three, I was thinking try a little better English.
 Input, Interaction and Output 3

(14) Mismatch between intent and interpretation: Morphosyntactic feedback

perceived as being about lexis (Mackey et al. 2000)
NNS: C’è due tazzi.
There is two cups (m-pl).
NS: Due tazz — come?
Two cup — what?
NNS: Tazzi, dove si puó mettere tè, come se dice questo?
Cups (m-pl), where one can put tea, how do you say this?
NS: Tazze?
Cups (f-pl)?
NNS: Okay, tazze.
Okay, cups (f-pl).
Recall: I wasn’t sure if I learned the proper word at the beginning.
Empirical research needs to further address the issue of the role of learners’ percep-
tions in interaction-driven learning. As noted above, Mackey et al. (2000) showed that
perceptions are not always straightforward; nevertheless, nascent research by Egi (in
press) has suggested that learners can benefit from recasts even without perceiving
them as such. Gass and Lewis (in press) have extended the database of studies on
learners’ perceptions by looking at heritage learners (i.e., those with a home-language
background different from the language of the external environment) and non-heri-
tage learners, and finding that these two subgroups differed in their perceptions about
feedback although this may in part have been due to the different types of feedback re-
ceived. In their study, the heritage learners perceived most feedback about semantics.

Output

Output refers to the language that learners produce. Interaction research often focuses
on output that is modified following feedback (Ellis and He 1999; McDonough 2005;
McDonough and Mackey in press.; Muranoi 2000; Shehadeh 2002; Swain 1985, 1995,
2005). This sort of modified output has been argued to promote learning since it stim-
ulates learners to reflect on their original language. In example (15) below, the learner
modifies her utterance following morphosyntactic feedback.
(15) Modified output (Mackey et al. 2000) (INT=interviewer)
NNS: Ecco il forno e i fornelle
Here is the oven and the (m-pl) little oven (f-pl)
INT: I fornelli?
The (m-pl) little ovens (m-pl)?
NNS: I fornelli
The little ovens (m-pl)
Recall: I was thinking, um, the stovetop, um, was that the right word for it
— il — i fornelli?
4 Susan Gass and Alison Mackey

Swain, in particular, has argued that production “may force the learner to move from
semantic processing to syntactic processing” (1985: 249), and many researchers have
claimed that output provides a forum for receiving feedback, which pushes learners
to produce more accurate, appropriate, complex, and comprehensible forms (Swain
1993, 2005; Swain and Lapkin 1995; see also Gass 1988, 1997; Long 1996; Pica 1994).
Swain (1995) has also suggested that output provides an opportunity for learners to
test hypotheses about the target language, and modify them where necessary.
For modified output to be useful, most interaction researchers suggest that it is
necessary for learners to notice the relationships between their initially erroneous
forms, the feedback they receive, and their output. Additionally, even if feedback is
perceived accurately, if the correct form is not supplied as part of the feedback, learn-
ers must be able to figure out how to correct their original erroneous forms. Thus, an
intervening variable in the interaction-learning relationship is developmental level, in
the sense that if a learner is not sufficiently advanced to be able to make appropriate
changes or to be sensitized to future patterns in the input, the feedback may not be
developmentally useful.

Conclusions

In concluding this description of the interaction framework, it is important to point

out that recent research is now considering the mechanisms involved in interaction
and is shifting its focus to an understanding of how interaction works. In other words,
what are the components of interaction that can help explain the positive effects, and
what learner-internal (cognitive) factors can help to explain differential benefits? It is
a truism that learners differ greatly in their ability to learn a second language, in terms
of both rate and ultimate attainment. If individual differences in cognitive capacities
affect learners’ engagement in interactional processes, then their developmental out-
comes may be impacted as a result. Robinson (2002), for instance, has highlighted the
importance of attentional resource allocation for the noticing of mismatches between
input and output, and, as noted above, many researchers have claimed that learners’
attentional processes mediate the interaction-learning relationship. Other individu-
al differences currently under investigation include motivation, learning strategies,
working memory, language aptitude, cognitive styles, and social context.
When considering these claims, it is important to remember, as Mackey (in press)
notes, that “the Interaction Hypothesis was not intended or claimed to be a complete
theory of SLA, despite the fact that it is occasionally characterized this way in the
literature” (p. 30). However, evaluated in terms of VanPatten and Williams’ (in press)
statements regarding theories, models, and hypotheses, the interaction hypothesis in-
cludes more than simply the elements of an hypothesis, or an idea that needs to be
tested about a single phenomenon. Arguably, it also contains elements of a model,
describing a set of related processes, and even elements of a theory, explaining why
attested phenomena occur in the way they do (using cognitive concepts derived from
 Input, Interaction and Output 5

psychology, such as attention and working memory). Substantial changes have oc-
curred since the early 1980s when the interaction hypothesis was first proposed. As
we have noted throughout this paper, the links between interaction and learning have
been clearly demonstrated through a good deal of empirical research, and various as-
pects of the hypothesis have been tested separately. In light of this progress, such de-
velopments suggest that a change in terminology may be warranted. Elsewhere (see
Gass and Mackey in press), we have provisionally referred to this area of research as
the interaction approach; as more research is carried out in the coming years, we look
forward to further advances.

References

Block, D. 2003. The Social Turn in Second Language Acquisition. Edinburgh: Edinburgh Univer-
sity Press.
Bohannon, J. N., Padgett, M. and Nelson, K. 1996. Useful evidence on negative evidence. Devel-
opmental Psychology 32: 551–555.
Egi, T. In press. Recasts, learners interpretations, and L2 development. In Conversational In-
teraction in Second Language Acquisition: A series of empirical studies, A. Mackey (ed.).
Oxford: Oxford University Press.
Ellis, R. and He, X. 1999. The roles of modified input and output in the incidental acquisition of
word meanings. Studies in Second Language Acquisition 21: 285–301.
Ellis, N. 2002. Frequency effects in language processing. Studies in Second Language Acquisition
24:143–188.
Ferguson, C. 1971. Absence of copula and the notion of simplicity: A study of normal speech,
baby talk, foreigner talk and pidgins. In Pidginization and Creolization of Languages, D.
Hymes (ed.), 141–150. Cambridge: Cambridge University Press.
Gass, S. 1988. Integrating research areas: A framework for second language studies. Applied
Linguistics 9: 198–217.
Gass, S. 1997. Input, Interaction, and the Second Language Learner. Mahwah NJ: Lawrence Erl-
baum Associates.
Gass, S. 2003. Input and interaction. In Handbook of Second Language Acquisition, C. Doughty
and M. Long (eds), 224–255. Oxford: Blackwell.
Gass, S. and Lewis, K. In press. Perceptions of interactional feedback: Differences between heri-
tage language learners and non-heritage language learners. In Conversational Interaction in
Second Language Acquisition: A series of empirical studies, A. Mackey (ed.). Oxford: Oxford
University Press.
Gass, S. and Mackey, A. 2000. Stimulated Recall Methodology in Second Language Research. Mah-
wah NJ: Lawrence Erlbaum Associates.
Gass, S. and Mackey, A. In press. In Theories in Second Language Acquisition, J. Williams and B.
VanPatten (eds). Mahwah NJ: Lawrence Erlbaum Associates.
Gass, S. Mackey, A. and Ross-Feldman, L. 2005. Task-based interactions in classroom and labo-
ratory settings. Language Learning 55: 575–611.
Gass, S. and Selinker, L. 2001. Second Language Acquisition: An introductory course. Mahwah NJ:
Lawrence Erlbaum Associates
6 Susan Gass and Alison Mackey

Gass, S. and Varonis, E. 1985. Variation in native speaker speech modification to non-native
speakers. Studies in Second Language Acquisition 7: 37–57.
Hatch, E. 1983. Psycholinguistics: A second language perspective. Rowley MA: Newbury House.
Krashen, S. 1982. Principles and Practice in Second Language Acquisition. London: Pergamon.
Krashen, S. 1985. The Input Hypothesis: Issues and implications. New York NY: Longman.
Leeman, J. 2003. Recasts and second language development: Beyond negative evidence Studies
in Second Language Acquisition 25: 37–63.
Long, M. H. 1983. Linguistic and conversational adjustments to non–native speakers Studies in
Second Language Acquisition 5: 177–193.
Long, M. H. 1996. The role of the linguistic environment in second language acquisition. In
Handbook of Language Acquisition, Vol. 2, W.C. Ritchie and T. K. Bhatia (eds), 413–468.
New York NY: Academic Press.
Long, M. H. and Robinson, P. 1998. Focus on form: Theory, research, and practice. In Focus on
Form in Classroom Second Language Acquisition, C. Doughty and J. Williams (eds), 15–41.
New York NY: Cambridge University Press.
Lyster, R. 1998. Recasts, repetition, and ambiguity in L2 classroom discourse. Studies in Second
Language Acquisition 20: 51–81.
Lyster, R. and Ranta, L. 1997. Corrective feedback and learner uptake: Negotiation of form in
communicative classrooms Studies in Second Language Acquisition 20: 37–66.
Mackey, A, (ed.) In press. Conversational interaction and second language acquisition: A collec-
tion of emperical studies. Oxford: Oxford University Press.
Mackey, A. and Oliver, R. 2002. Interactional feedback and children’s L2 development. System
30: 460–477.
Mackey, A., Gass, S. and McDonough, K. 2000. How do learners perceive interactional feed-
back? Studies in Second Language Acquisition 22: 471–497.
McDonough, K. 2005. Identifying the impact of negative feedback and learners’ responses on
ESL question development. Studies in Second Language Acquisition 27: 79–103.
McDonough, K. and Mackey, A. In press. Responses to recasts: Repetitions, primed production
and linguistic development. Language Learning.
Muranoi, H. 2000. Focus on form through interactional enhancement: Integrating formal in-
struction with a communicative task in EFL classrooms. Language Learning 50: 617–673.
Philp, J. 2003. Constraints on noticing the gap: Nonnative speakers’ noticing of recasts in NS-
NNS interaction. Studies in Second Language Acquisition 25: 99–126.
Pica, T. 1994. Research on negotiation: What does it reveal about second-language learning
conditions, processes, and outcomes? Language Learning 44: 493–527.
Robinson, P. (ed.). 2002. Individual Differences and Instructed Language Learning. Amsterdam:
John Benjamins.
Saxton, M. 1997. The contrast theory of negative input. Journal of Child Language 24: 139–161.
Saxton, M. 2005. Recast in a new light: Insights for practice from typical language studies. Child
Language Teaching and Therapy 21: 23–38.
Shehadeh, A. 2002. Comprehensible output, from occurrence to acquisition: An agenda for ac-
quisitional research. Language Learning 52: 597–647.
Swain, M. and Lapkin, S. 1995. Problems in output and the cognitive processes they generate: A
step toward second language learning. Applied Linguistics 16: 371–391.
Swain, M. and Lapkin, S. 1998. Interaction and second language learning: Two adolescent French
immersion students working together. The Modern Language Journal 82: 320–337.
 Input, Interaction and Output 7

Swain, M. 1985. Communicative competence: Some roles of comprehensible input and compre-
hensible output in its development. In Input in Second Language Acquisition, S. Gass and C.
Madden (eds), 235–253. Rowley MA: Newbury House.
Swain, M. 1993. The output hypothesis: Just speaking and writing aren’t enough. Canadian Mod-
ern Language Review 50: 158–164.
Swain, M. 1995. Three functions of output in second language learning. In Principle and Practice
in Applied Linguistics: Studies in honour of H. G. Widdowson, G. Cook and B. Seidlhofer
(eds), 125–144. Oxford: Oxford University Press.
Swain, M. 2005. The output hypothesis: Theory and research. In Handbook on Research in Sec-
ond Language Learning and Teaching, E. Hinkel (ed.), 471–483. Mahwah NJ: Lawrence Erl-
baum.
Takashima, H. 1995. A study of focused feedback, or output enhancement, in promoting accu-
racy in communicative activities. Ed.D. dissertation, Temple University, Japan.
VanPatten, B. and Williams, J. (eds). In press. Theories in Second Language Acquisition. Mahwah
NJ: Lawrence Erlbaum Associates.