What Is the Interaction Hypothesis?
The Interaction Hypothesis states that second language acquisition occurs most effectively when learners engage in conversational interaction with negotiation of meaning to make input comprehensible. Proposed by psycholinguist Michael Long in 1981 and revised in 1996, this theory emphasizes that learners acquire language not through passive exposure but through active conversational exchanges that include clarification requests, confirmation checks, and corrective feedback. The hypothesis argues that comprehensible input alone is necessary but not sufficient—interaction and negotiation processes connect input, cognitive processes, and output to facilitate language development (Long, 1981, 1996; Tran-Hoang-Thu, ERIC 2009).
Skip to Your Section
What Is the Interaction Hypothesis in Second Language Acquisition?
The Interaction Hypothesis is a theory of second language acquisition stating that language development is promoted through face-to-face conversational interaction where learners and interlocutors negotiate meaning to make input comprehensible. Developed by Michael Long in his 1981 paper “Input, interaction, and second-language acquisition” published in Annals of the New York Academy of Sciences, the hypothesis posits that comprehensible input is necessary but not sufficient for acquisition (Long, 1981; Tran-Hoang-Thu, ERIC 2009).
According to Rod Ellis’s 1991 critical evaluation published in ERIC, the Interaction Hypothesis advances two main claims:
- Comprehensible input is necessary for second language acquisition
- Modifications to the interactional structure of conversations taking place in the process of negotiating a communication problem help to make input comprehensible to a second language learner
(Ellis, 1991, ERIC; Tran-Hoang-Thu, ERIC 2009, p. 3)
Long’s 1996 revision in “The role of linguistic environment in second language acquisition” (published in Handbook of Second Language Acquisition, edited by W. Ritchie and T. K. Bhatia, pp. 413-468) emphasized cognitive factors. Long wrote:
“…it is proposed that environmental contributions to acquisition are mediated by selective attention and the learners’ developing L2 processing capacity, and that these resources are brought together most usefully, although not exclusively, during negotiation for meaning. Negative feedback obtained during negotiation work or elsewhere may be facilitative of L2 development, at least for vocabulary, morphology, and language-specific syntax, and essential for learning certain specifiable L1-L2 contrasts.”
(Long, 1996, p. 414; cited in Tran-Hoang-Thu, ERIC 2009, pp. 5-6)
Three core components define the Interaction Hypothesis according to Ellis (1991):
- Comprehensible input is necessary for second language acquisition
- Negotiation of meaning makes input comprehensible through conversational modifications
- Negative feedback during interaction facilitates development of vocabulary, morphology, and syntax
Comprehensible input forms the foundation—learners must be exposed to language at a level they can understand (Krashen’s i+1 concept, where i represents current proficiency and +1 represents the next developmental stage). However, unlike Krashen’s Input Hypothesis which viewed this as both necessary and sufficient, Long argued that input alone doesn’t guarantee acquisition (Tran-Hoang-Thu, ERIC 2009, p. 3).
The critical difference lies in how that input becomes comprehensible. According to Ellis (1994), Krashen claimed input becomes comprehensible through simplification and contextual clues, whereas Long argued that interactive input is more important than non-interactive input (cited in Tran-Hoang-Thu, ERIC 2009, p. 3).
Negotiation of meaning occurs when communication breaks down and interlocutors use strategies to repair understanding. As Long wrote: “modifications to the interactional structure of conversations which take place in the process of negotiating a communication problem help to make input comprehensible to an L2 learner” (Ellis, 1991, p. 4; cited in multilingualpedagogy.lmc.gatech.edu).
Negative feedback through interaction provides learners with information about the incorrectness of their utterances. According to Long’s 1996 revision, this feedback may be facilitative of L2 development, particularly for vocabulary, morphology, and language-specific syntax, and essential for learning certain L1-L2 contrasts (Long, 1996, p. 414).
The hypothesis emerged during the late 1970s and early 1980s when second language acquisition research shifted from behaviorist to cognitive approaches. According to Tran-Hoang-Thu’s 2009 ERIC literature review, the origins of the Interaction Hypothesis lie partially in Evelyn Hatch’s 1978 work on discourse analysis and second language acquisition, and partially in Krashen’s Input Hypothesis (Tran-Hoang-Thu, ERIC 2009, p. 6). Hatch demonstrated that learner-native speaker conversations involved specific patterns that guided production and development.
For further understanding of how learners’ internal language systems develop through interaction, explore Interlanguage Theory: Understanding Student Language Development Stages, which examines the systematic nature of learner language at different developmental stages.
Who Created the Interaction Hypothesis and When?
Michael H. Long created the Interaction Hypothesis in his doctoral research at the University of California in the 1980s (eal360.wordpress.com, 2014). His 1980 PhD dissertation at UCLA was titled “Input, interaction, and second language acquisition” (Tran-Hoang-Thu, ERIC 2009, p. 12, References).
Long first published the hypothesis in 1981 in a paper titled “Input, interaction, and second language acquisition” in Annals of the New York Academy of Sciences, Volume 379, pages 259-278 (Tran-Hoang-Thu, ERIC 2009, p. 12). He refined it in 1983 in “Native speaker/non-native speaker conversation and negotiation of comprehensible input” published in Applied Linguistics, Volume 4(2), pages 126-141 (Tran-Hoang-Thu, ERIC 2009, p. 13).
According to eal360 (2014), Long conducted his PhD research pairing native speakers and non-native speakers of English to engage in face-to-face oral tasks, such as giving instructions for games, playing games, or informal conversation, examining how they overcame communication difficulties.
Long published his major 1996 revision in “The role of linguistic environment in second language acquisition,” published in W. Ritchie and T. K. Bhatia’s Handbook of Second Language Acquisition, pages 413-468 (Tran-Hoang-Thu, ERIC 2009, p. 13). This revision integrated cognitive factors including noticing, attention, and awareness into the framework.
The hypothesis evolved through three key stages (Tran-Hoang-Thu, ERIC 2009; Wikipedia, 2024):
- 1981: Initial formulation emphasizing that “participation in conversation with native speakers, which is made possible through the modification of interaction, is the necessary and sufficient condition for second language acquisition” (Tran-Hoang-Thu, ERIC 2009, p. 2)
- 1983: Refinement incorporating negotiation of meaning and emphasizing that “access to comprehensible input is characteristics of successful acquisition in first and second language” (Ellis, 1991, cited in Tran-Hoang-Thu, ERIC 2009, p. 3)
- 1996: Major revision emphasizing that environmental contributions to acquisition are mediated by selective attention and L2 processing capacity, with negative feedback playing a facilitative role (Long, 1996, p. 414)
The 1981-1983 initial period established two fundamental claims according to Ellis (1991): (1) comprehensible input is necessary for SLA, and (2) modifications to interactional structure during negotiation help make input comprehensible (Tran-Hoang-Thu, ERIC 2009, p. 3).
However, according to Tran-Hoang-Thu (2009), there was a clear shift in Long’s position by 1989: “Long (1981) clearly stated that input is both necessary and sufficient for SLA; however, in his paper in 1989 he admitted that although comprehensible input is necessary for SLA, it may not be sufficient” (Tran-Hoang-Thu, ERIC 2009, p. 4).
The 1996 major revision represented the most significant evolution. Long clarified that the Interaction Hypothesis “is certainly not intended as anything like a complete theory of language learning because many aspects of the proposal have barely been investigated in adult second language acquisition” (Long, 1996; cited in Tran-Hoang-Thu, ERIC 2009, p. 6).
According to Gass and Mackey (2007), Long’s 1996 revision means that “through interaction learners’ selective attention is directed to problematic features of knowledge of production. Learners may first recognize what they say differs from what a native speaker does and learners may notice that they cannot express what they wish to express” (cited in Tran-Hoang-Thu, ERIC 2009, p. 6).
How Does Negotiation of Meaning Work in the Interaction Hypothesis?
Negotiation of meaning occurs when communication breakdowns trigger interlocutors to use clarification requests, confirmation checks, and comprehension checks to resolve misunderstandings and make input comprehensible. According to Long’s research at University of California in the 1980s, when native and non-native speakers faced communication difficulties, pairs would negotiate meaning through various methods (eal360, 2014).
According to eal360 (2014), Long identified these negotiation strategies:
- Repetition of words and phrases
- Confirmation checks: when one speaker sought confirmation of the other’s preceding utterance through repetition or rising intonation
- Comprehension checks: when one speaker—usually the native speaker—checks if the other has understood
- Clarification requests: when one speaker—usually the non-native speaker—asks the other for help in understanding
This negotiation process has also been termed “conversational repair” (eal360, 2014).
According to Teresa Pica’s 1987 research “Second-language acquisition, social interaction, and the classroom” published in Applied Linguistics, negotiation signals indicate comprehension difficulties. Gass and Mackey (2007) explain that feedback constitutes two broad types:
- Explicit feedback: includes corrections and metalinguistic explanation
- Implicit feedback: includes negotiation strategies such as confirmation checks, requests for clarification, comprehension checks, and recasts (a rephrasing of an incorrect utterance using a corrected version while maintaining the original meaning)
(Gass & Mackey, 2007; cited in Tran-Hoang-Thu, ERIC 2009, p. 8)
Three primary negotiation strategies according to Pica (1987):
| Strategy | Definition | Function | Source |
|---|---|---|---|
| Clarification Request | Questions or statements seeking explanation | Signals non-understanding; prompts reformulation | Pica, 1987 |
| Confirmation Check | Repetition with rising intonation to verify understanding | Verifies correct interpretation of utterance | eal360, 2014 |
| Comprehension Check | Speaker verifies listener has understood | Tests comprehension; ensures mutual understanding | eal360, 2014 |
Example of negotiation from research: Mitchell and Myles (2004, p. 168) provide an example citing Pica et al. (1987), documented on eal360 (2014):
NS: And right on the roof of the truck place the duck. The duck
NNS: I to take it? Dog?
NS: Duck.
NNS: Duck.
NS: It’s yellow and it’s a small animal. It has two feet.
NNS: I put where it?
NS: You take the duck and put it on top of the truck. Do you see the duck?
NNS: Duck?
NS: Yeah. Quack, quack, quack. That one. The one that makes that sound.
NNS: Ah yes, I see in the…in the head of him.
(Mitchell & Myles, 2004, p. 168, citing Pica et al. 1987; documented in eal360, 2014)
This example demonstrates multiple negotiation strategies: clarification (“Duck?” seeking confirmation), repetition (“Duck” repeated), comprehension checks (“Do you see the duck?”), and elaboration (“It’s yellow and it’s a small animal. It has two feet” providing additional descriptive information).
According to eal360 (2014), Vivian Cook suggests that “second language acquisition depends on interaction that involves resolving comprehension problems.” The adjustments from this interaction connect input, intake, and output.
According to Long’s 1996 revision, “negotiation for meaning, especially negotiation work that triggers interactional adjustment by the native speakers or more competent interlocutors, facilitates acquisition as it connects input, internal learner capacities, particularly selective attention, and output in productive ways” (Long, 1996; cited in Tran-Hoang-Thu, ERIC 2009, p. 6).
Research by Long and Porter (1985) found that “learners of English talked more with other learners of English (especially advanced learners) than they did with native speakers” (eal360, 2014; Wikipedia, 2024). Yule and MacDonald (1990) found this was especially the case when advanced learners were given a less dominant role—when advanced learners dominated group or paired tasks, less advanced learners became passive (eal360, 2014).
What Is the Difference Between the Interaction Hypothesis and Input Hypothesis?
The Interaction Hypothesis differs from Krashen’s Input Hypothesis by emphasizing how input becomes comprehensible through active negotiation rather than merely receiving comprehensible input passively. While Krashen viewed comprehensible input (i+1) as necessary and sufficient for acquisition, Long argued it is necessary but not sufficient—learners must also engage in negotiation for meaning and receive negative feedback during interaction (Long, 1994, 1996; multilingualpedagogy.lmc.gatech.edu).
According to multilingualpedagogy.lmc.gatech.edu, Michael Long wrote:
“Whereas Krashen views comprehensible input (CI) one step ahead of the learner’s current level as necessary and sufficient for acquisition, I have long argued for the interaction hypothesis….I maintain that CI is necessary but not sufficient for SLA…I have further argued for the importance of negotiation for meaning and negative feedback in orienting learners’ attention to form in this way.”
(Long, 1994, p. 788; cited in multilingualpedagogy.lmc.gatech.edu)
According to Ellis (1994), there is a difference in how Krashen and Long viewed comprehensible input: “Krashen claimed that input becomes comprehensible thanks to simplification and with the help of contextual and extralinguistic clues, whereas Long argued that interactive input is more important than non-interactive input” (cited in Tran-Hoang-Thu, ERIC 2009, p. 3).
Key differences according to research:
| Aspect | Input Hypothesis (Krashen) | Interaction Hypothesis (Long) | Source |
|---|---|---|---|
| Role of Input | Necessary and sufficient | Necessary but not sufficient | Tran-Hoang-Thu, ERIC 2009, p. 4 |
| How Input Becomes Comprehensible | Simplification and contextual clues | Interactive negotiation | Ellis, 1994; ERIC 2009, p. 3 |
| Key Mechanism | Comprehensible input (i+1) | Negotiation of meaning | Ellis, 1991 |
| Feedback | Not emphasized | Essential for development | Long, 1996 |
According to Tran-Hoang-Thu (2009), “Long (1981) clearly stated that input is both necessary and sufficient for SLA; however, in his paper in 1989 he admitted that although comprehensible input is necessary for SLA, it may not be sufficient” (ERIC 2009, p. 4). This represented a significant evolution in Long’s thinking.
The Interaction Hypothesis positions itself as filling perceived gaps in Krashen’s Input Hypothesis rather than refuting it entirely. Both frameworks agree that comprehensible input is essential. According to multilingualpedagogy.lmc.gatech.edu, “Long’s interaction hypothesis, which does not refute but rather fills in perceived gaps in Krashen’s Input Hypothesis, suggests that comprehensible input is important, but the negotiations created by interactions between speaker and audience are an essential component in promoting language acquisition.”
To understand how errors in learner language can become permanent without proper attention and feedback mechanisms, read What is Fossilization in Language Learning and How Can You Prevent It?, which explores why some linguistic features remain incorrect even after years of study.
How Does the Interaction Hypothesis Apply to Language Teaching?
The Interaction Hypothesis applies to language teaching through Task-Based Language Teaching (TBLT), where communicative tasks naturally elicit negotiation of meaning during pair and group work. According to a 2018 review by Loewen and Sato published in Language Teaching (Cambridge Core), “it is this theory and related empirical findings that provide the main psycholinguistic underpinnings for TBLT [task-based language teaching]” (Cambridge Core, Volume 51, Issue 3, July 2018, pp. 285-329).
Long’s 2015 book Second Language Acquisition and Task-Based Language Teaching emphasized that TBLT creates optimal conditions for the negotiation sequences found beneficial for L2 development (cited in Cambridge Core, 2018).
According to Tran-Hoang-Thu (2009), the Interaction Hypothesis has pushed pedagogical research into new frontiers. Brown (2007) stated it “centers educators on the language classroom not just as a place in which learners of varying abilities, styles, and backgrounds mingle, but as a place in which contexts for interaction are carefully designed” (cited in ERIC 2009, p. 9-10).
Practical applications include:
- Information-gap tasks: where learners must exchange unknown information, prompting natural negotiation
- Jigsaw activities: requiring collaborative reconstruction through sustained interaction
- Focus-on-form instruction: brief attention to linguistic features during meaning-focused tasks
According to Ellis (2008), the later version of the Interaction Hypothesis has been closely related to another construct, Focus-on-Form, which is “a type of instruction where attention to form arises out of meaning-centered activity derived from the performance of a communicative task” (cited in Tran-Hoang-Thu, ERIC 2009, p. 6).
Long distinguished focus-on-form from focus-on-forms (traditional grammar instruction). According to research documented on multilingualpedagogy.lmc.gatech.edu, Long explained that “during an otherwise meaning-focused lesson, and using a variety of pedagogic procedures, learners’ attention is briefly shifted to linguistic code features, in context, when students experience problems as they work on communicative tasks.”
According to eal360 (2014), Long and Porter’s (1985) and Yule and MacDonald’s (1990) research has “considerable implications for grouping students, suggesting that mixed ability groupings facilitate learning more than grouping students of similar ability” because when advanced learners dominated tasks, less advanced learners became passive.
Note on classroom application: Gass and Mackey (2007) cautioned that “it is also wrongly believed that the IH can be directly applicable to classroom methodology because the IH is focused primarily on how languages are learned. Therefore…it is likely to be premature to see direct application to the classroom” (cited in Tran-Hoang-Thu, ERIC 2009, p. 9).
What Are the Limitations of the Interaction Hypothesis?
The Interaction Hypothesis has documented limitations according to research. Ellis (2008) identified several caveats with the updated version of the hypothesis (cited in Tran-Hoang-Thu, ERIC 2009, p. 7):
Three main limitations identified by Ellis (2008):
- Limited scope: “A theory of language acquisition based on one type of interaction (negotiation sequence) or a single interactional strategy such as recasts would seem relatively limited as it constitutes only a small portion of the total interaction a learner experiences”
- Proficiency level effects: “Negotiation may work best with intermediate learners, whereas it may not be very effective for a beginner or an advanced learner as the former does not have enough linguistic resources to negotiate effectively and the latter tends to attend to opinion and interpretation rather than comprehension and linguistic clarity”
- Individual differences: “Other factors to be considered include learners’ ability or their preparedness to negotiate and differences in interactants’ negotiation styles”
(Ellis, 2008; cited in Tran-Hoang-Thu, ERIC 2009, p. 7)
According to Tran-Hoang-Thu’s 2009 ERIC literature review, “the role of negative feedback in facilitating SLA may seem more ideal than realistic in real life communication and interaction. The rationale behind the usefulness of negative feedback for SLA is convincing and seems to be well grounded, but negative feedback is not often found in daily communication and interaction” (ERIC 2009, p. 10).
Research support is mixed. According to Tran-Hoang-Thu (2009), “there are a number of research studies that lend support to the IH such as Pica, Young, and Doughty (1987), Gass and Varonis (1994), Loschky (1994), and Ellis, Tanaka, and Yamazaki (1994). However, there are other studies that fail to support the IH (Derwing, 1996; Erlich, Avery, & Yorio, 1989; Sato, 1986)” (ERIC 2009, p. 7).
Long himself acknowledged limitations. According to the 1996 revision, Long “clearly pointed out that the IH is certainly not intended as anything like a complete theory of language learning because many aspects of the proposal have barely been investigated in adult second language acquisition” (cited in Tran-Hoang-Thu, ERIC 2009, p. 5-6).
Despite limitations, Ellis (2008) concluded that “the interaction hypothesis has generated considerable interest in the field of SLA, has spawned a large number of research studies, and has demonstrated enough explanatory power. No theory of SLA…can be complete without an account of the role played by interaction and the interaction hypothesis is still one of the most convincing statements of this role to date” (cited in Tran-Hoang-Thu, ERIC 2009, p. 7).
According to Tran-Hoang-Thu (2009), “the updated version is indeed broader in scope and is seemingly more tenable in comparison with the earlier versions. However, like any other theory or hypothesis, the IH is certainly not complete and is unable to account for all aspects in the process of SLA and Long (1996) did mention this point” (ERIC 2009, p. 10).
Frequently Asked Questions About Interaction Hypothesis
What is the main idea of the Interaction Hypothesis?
The main idea of the Interaction Hypothesis is that second language acquisition occurs most effectively when learners engage in conversational interaction that involves negotiation of meaning to make input comprehensible. According to Ellis (1991), the hypothesis advances two main claims: (1) comprehensible input is necessary for SLA, and (2) modifications to the interactional structure of conversations during negotiation help make input comprehensible to learners (Tran-Hoang-Thu, ERIC 2009, p. 3). Long argued that comprehensible input is necessary but not sufficient—learners must actively negotiate meaning through clarification requests, confirmation checks, and receive feedback during interaction (Long, 1996).
How does the Interaction Hypothesis differ from the Output Hypothesis?
The Interaction Hypothesis differs from Swain’s Output Hypothesis by emphasizing the negotiation processes during interaction rather than primarily the act of producing language. According to Tran-Hoang-Thu (2009), “the current form of the IH includes some features of the input hypothesis proposed by Krashen as well as those of the output hypothesis advanced by Swain (1985, 1995, 2005)” (ERIC 2009, p. 6). Gass and Mackey (2007) explain that after producing a problematic utterance and receiving feedback, “learners can realize that what they have just said is not understood, so they have to force themselves to reformulate the initial utterance so as to make themselves understood by producing more target-like output” (cited in ERIC 2009, p. 8-9). Both theories are considered complementary—interaction provides the context for output, and output opportunities emerge from negotiation.
What are examples of negotiation of meaning strategies?
Examples of negotiation of meaning strategies according to Long’s research and Pica (1987) include: clarification requests (when speakers ask for help understanding, such as questions or “I don’t understand”), confirmation checks (when speakers seek confirmation through repetition with rising intonation), comprehension checks (when speakers verify the listener has understood), repetition of words and phrases, and recasts (rephrasing an incorrect utterance using a corrected version while maintaining original meaning). According to eal360 (2014), Long found these strategies in his research at University of California when native and non-native speakers overcame communication difficulties, and this process has been termed “conversational repair”.
Why is the Interaction Hypothesis important for language teachers?
The Interaction Hypothesis is important for language teachers because it provides the theoretical foundation for Task-Based Language Teaching (TBLT) according to Cambridge Core’s 2018 review. Brown (2007) stated it “centers educators on the language classroom not just as a place in which learners of varying abilities, styles, and backgrounds mingle, but as a place in which contexts for interaction are carefully designed” and “focuses material and curriculum developers on creating the optimal environments and tasks for input and interaction” (cited in Tran-Hoang-Thu, ERIC 2009, p. 9-10). However, Gass and Mackey (2007) cautioned that “it is likely to be premature to see direct application to the classroom” because the hypothesis is focused primarily on how languages are learned (cited in ERIC 2009, p. 9).
Can the Interaction Hypothesis work in online language learning?
The original Interaction Hypothesis research (Long 1980-1996) focused on face-to-face oral interaction, and the ERIC 2009 literature review does not address computer-mediated communication. According to eal360 (2014), Long’s PhD research examined “face-to-face oral tasks” between native and non-native speakers. While subsequent research has examined online contexts, the original hypothesis and its empirical support come from face-to-face conversational interaction. Contemporary applications to online learning would need verification through recent research beyond the scope of the foundational sources consulted here.
Explore More Language Acquisition Theories
The Interaction Hypothesis represents one crucial perspective in understanding how learners develop second language proficiency. For comprehensive insights into other influential frameworks and practical teaching applications, explore more language acquisition theories and research-based strategies in our Language Acquisition & Learning Theories section.
