Long's interaction hypothesis

Theory · All theories
Evidence · 39 claims (13 for, 26 mixed) · 15 studies (8 causal, 3 quant-synthesis, 2 review, 2 associational), q2–q3 · 3 of 15 report an effect size · 37 claims rest on one study

Description

Ellis (eric-ed338037) summarises the interaction hypothesis, which he associates most clearly with Michael Long, as a three-part hierarchy. First, comprehensible input is necessary for L2 acquisition, which is Krashen's input hypothesis. Second, the modifications to conversational structure that happen while speakers negotiate a communication problem make input comprehensible. These are the confirmation checks, clarification requests, comprehension checks and repetitions that Long (1980) found much more often in native/non-native than in native/native conversation. Third, some conditions create more of this negotiation. Two-way tasks that require an exchange of information create more than one-way tasks (Long), and so do partners with equal status despite unequal proficiency (Pica 1987). Long drew teaching proposals from the hypothesis: fewer display questions, more group work, and two-way, closed tasks.

The hypothesis changed over time. By 1989, Ellis reports, Long still held that comprehensible input is necessary but accepted that it may not be sufficient. Drawing on Swain's output hypothesis, he also accepted that interaction helps by pushing learners to modify their output. On this view some moves do more than others: requests for clarification "stretch" the learner, while confirmation checks solve the problem for them. Later accounts put feedback at the centre. Rassaei and Moinzadeh (eric-ej1080713) cite Long (1996) for the view that recasts promote L2 development because they come during meaning-focused activity, giving learners both comprehensible input and a focus on form. Lv and Liu (eric-ej1361438) cite Long (2007) for the view that a recast can lead a learner to notice the gap. Shetye (eric-ej1435677) places the hypothesis inside Gass and Mackey's (2020) "interaction approach", in which negotiation draws learners' attention to input, output and feedback. Altiner (eric-ej1208383) reports Long (1985) as holding that interaction in the L2 is "the obligatory condition" for acquisition.

The articles disagree about how far interaction explains acquisition. Ellis found the most convincing evidence for the third part of the hypothesis, the conditions that produce negotiation. He found weaker evidence that interactionally modified input leads to acquisition, rather than to comprehension. His evidence includes Sato's two learners, who never acquired past-tense marking because interactional support made it unnecessary, and Loschky's comprehension gain with no retention gain. His revision (see Ellis's revised interaction hypothesis) says that interaction makes acquisition possible by helping learners notice features and compare them with their own output, but does not cause it. Ponniah (eric-ej1071042) reports Krashen's opposite view, that negotiating conversation "won't help us to acquire the language".

Implications

Context

• Two-way information-exchange tasks, and closed tasks with a limited set of correct solutions, are the task conditions the hypothesis favours (Ellis's account of Long).
• In EFL peer interaction the amount of negotiation depends on task type and proficiency pairing (eric-ej1269664). Unfocused communicative tasks drew more negotiation than focused ones in one Iranian study (eric-ej1348282).
• Ellis warns that the quality of modifications may matter more than their number, and that teacher-dominated discourse is not always free of negotiation.

Target Learners

• Second language learners interacting with native speakers, teachers or peers. The fetched studies cover adult and university EFL learners (eric-ej1080713, eric-ej1348282) and primary-age EFL children (eric-ej1269664, eric-ej1361438).
• Ellis reports that negotiation helped the comprehension of learners rated lower in comprehension ability more than it helped others (Pica 1989).

Target Learning Objectives

• Comprehension of L2 input, which is the best-supported link.
• Grammatical development through interactional feedback such as recasts. Here the fetched studies are quasi-experiments with short delays.

Claims

• Interactionally modified input aids comprehension but shows no advantage for retention among beginner learners [~M]
• Interactionally adjusted and premodified input yield statistically comparable comprehension overall, with negotiation benefiting lower-ability learners [~M]
• Comprehensible input alone is insufficient for acquisition of high levels of grammatical proficiency [~M]
• Child EFL learners receiving recasts in task-based lessons improve their production of third person singular -s from pretest to posttest, in a study with no no-recast comparison group [+M]
• Metalinguistic feedback's gains on wh-question performance decline over a ten-day delay while recast gains hold, leaving the two no different [+M]
• Metalinguistic feedback outperforms recasts on the immediate post-test, but the two do not differ significantly on the delayed post-test [~M]
• Metalinguistic feedback and recasts, but not clarification requests, significantly improve Iranian EFL learners' wh-question performance on an immediate post-test [~M]
• Higher task complexity is associated with lower written production of the target form among child learners receiving recasts [~M]
• Mixed proficiency dyads negotiate for meaning more than matched proficiency dyads, irrespective of time and language [~M]
• EFL primary children's meaning negotiation during a spot-the-differences task is low in amount but significantly higher in the L2 than in the L1 [~M]
• Unfocused communicative tasks elicit significantly more meaning negotiation than focused tasks among intermediate EFL learners [~M]
• Task condition (decision-making vs. opinion-exchange) did not significantly moderate the amount of meaning negotiation [~M]
• In high school groups, learners with negotiation of meaning obtained more comprehensible input in the match task than learners with pre-modified input only [+M]
• In both high school and college groups, learners with negotiation of meaning scored higher on the delayed post-vocabulary test than learners with pre-modified input only [+M]
• In high school groups, frequency of negotiation of meaning correlates positively with vocabulary score (r = 0.761) and with comprehensible input (r = 0.310) [+M]
• In college groups, negotiation of meaning made no difference to comprehensible input obtained in the match task [~M]
• In college groups, no correlation was found between frequency of negotiation of meaning and vocabulary score [~M]
• Learner age/level moderates the benefit of negotiation of meaning: high school learners relied on it more than college learners [~M]
• Non-native speaker dyads provided with subtitled videos before tasks produce a significantly higher percentage of negotiated turns in online TBLL than dyads without videos [~M]
• Completing meaning-focused tasks through CMC fosters negotiated interactions in all three modes (discussion board, text chat, videoconferencing) [+M]
• Closed meaning-focused tasks foster more negotiation of meaning and negative feedback than open tasks in CMC [+M]
• Videoconferencing was conducive to more negotiation of meaning than text chat and discussion board, measured as proportion of negotiated turns [~M]
• Discussion board interactions generated no negative feedback at all, suggesting synchrony is required to foster corrective feedback [~M]
• Unprompted self-correction dominates negative feedback in NNS-NNS CMC interaction, interpreted as a face-saving preference [~M]
• Content-related problems trigger most negotiation routines, and syntactic triggers are absent from text chat where turns are short [~M]
• Mixed dyads outperform matched dyads in meaning negotiation only in the L2 (proficiency pairing × language interaction) [+W]
• All meaning negotiation strategies appear in child EFL peer interaction except comprehension checks, which are absent [~W]
• Input-stage, but not output-stage, interaction patterns are significantly associated with interpreting fluency and target language quality, with Intensive Engagers scoring lowest [+W]
• Clarification requests show a three-way interaction of proficiency pairing, time and language, declining over time only in matched dyads and in the L1 [~W]
• Strategy-specific effects: time affects self-repetitions and marginally clarification requests; proficiency pairing affects confirmation checks and negative feedback; language affects other-repetitions and negative feedback [~W]
• Time (two years) shows no significant effect on children's overall amount of meaning negotiation [~W]
• Task condition (decision-making vs. opinion-exchange) did not significantly moderate negotiation strategy use [~W]
• Within-language strategy profiles differ: other-repetition prevails in the L2 while clarification requests prevail in the L1, with negative feedback least used [~W]
• Conversational interaction facilitates second language grammatical development in adult ESL learners [+W] — attached 2026-10-14 from Alison Mackey (1999), which proposed "Long's (1996) updated interactionist hypothesis: implicit negative feedback obtained through negotiated interaction facilitates SLA"; tests this page's relationship.
• Negotiated interaction yields higher moment-to-moment comprehension of L2 input than either noninteraction condition, which do not differ from each other [+W] — attached 2026-10-15 from Lester Loschky (1994), which proposed "Input hypothesis and its interactionist modification as competing accounts of the comprehension-acquisition relationship"; tests this page's relationship.
• The learner's developmental level conditions the outcomes of interactional input [~W]
• Laboratory and classroom studies assess the effects of recasts and different CF types differentially [~W]
• Interaction in both SCMC and FTF contexts has a significant impact on L2 development (meta-analysis of 14 studies, 1990–2012) [+M]

The evidence by question

• Does negotiated interaction help learners acquire the second language, beyond helping them understand the input at the moment? Conversational interaction facilitates second language grammatical development in adult ESL learners [+M]
• Does corrective feedback on second-language errors improve accuracy, and which form of correction (recast, prompt, metalinguistic clue, explicit correction with or without a rule) works for which target and learner? Corrective feedback on second-language errors has a significant effect on L2 development that is maintained on delayed tests [+S]
• In meaning-focused second-language tasks done through text chat and other computer-mediated modes, which task and mode features produce negotiation of meaning and corrective feedback? Completing meaning-focused tasks through CMC fosters negotiated interactions in all three modes (discussion board, text chat, videoconferencing) [+M]

Related Theories

• Ellis's revised interaction hypothesis: noticing, comparison, and integration as the basic acquisitional procedures — Ellis's revision of this hypothesis
• Krashen's input hypothesis and Monitor Model of second language acquisition — the source of the hypothesis's first claim
• Swain's output hypothesis — which Long and Pica later took in
• Schmidt's noticing hypothesis
• Lyster and Ranta's taxonomy of corrective feedback types
• Skehan's limited attention capacity model applied to child learners' uptake of recasts
• The Comprehension Hypothesis: comprehensible input as the crucial ingredient of second language acquisition
• Ellis's revised Interaction Hypothesis: acquisition via noticing, comparison and integration
• Interaction hypothesis (Long): negotiation of meaning during interaction generates comprehensible input that drives L2 acquisition
• Varonis and Gass four-phase model of negotiation of meaning routines
• Negotiation of meaning sequences model (Varonis & Gass, revised by Smith) with trigger and resolution parts

Examples

• For child beginner language learners, keep task complexity low or moderate so attention remains available for noticing recasts
• Provide incidental corrective feedback through clarification requests, recasts, and a P.S.
• Promote task-based peer interaction in primary EFL classes and scaffold L2 meaning negotiation through formulaic language and explicit training in negotiation strategies

Key Sources

• Long, M. H. (1983). Native speaker/non-native speaker conversation and the negotiation of comprehensible input. Applied Linguistics, 4(2), 126–141. doi:10.1093/applin/4.2.126
• Long, M. H. (1983). Native speaker/non-native speaker conversation in the second language classroom. In M. Clarke & J. Handscombe (Eds.), On TESOL '82: Pacific perspectives on language and teaching. TESOL.
• Long, M. H. (1996). The role of the linguistic environment in second language acquisition. In W. C. Ritchie & T. K. Bhatia (Eds.), Handbook of second language acquisition (pp. 413–468). San Diego, CA: Academic Press. doi:10.1016/B978-012589042-7/50015-3
• Pica, T. (1987). Second-language acquisition, social interaction, and the classroom. Applied Linguistics, 8(1), 3–21. doi:10.1093/applin/8.1.3
• Pica, T., Young, R., & Doughty, C. (1987). The impact of interaction on comprehension. TESOL Quarterly, 21(4), 737–758. doi:10.2307/3586992
• Ellis, R. (1991). The interaction hypothesis: A critical evaluation. ERIC Document ED 338 037. https://eric.ed.gov/?id=ED338037

Page metadata

Field	Value
type	theory
title	Long's interaction hypothesis
description	Long's claim that comprehensible input drives second language acquisition and that negotiating communication problems in conversation, through clarification requests, confirmation checks and recasts, is what makes input comprehensible and draws attention to form.
status	review
canonical	True
generated	by claude/unspecified on 2026-09-30
trust	unverified — no one has confirmed this page's content

Frontmatter is the page's machine-readable record — see the Schema & Guide.