Interaction Hypothesis
The Interaction Hypothesis, proposed by Michael Long, claims that conversational interaction—especially when speakers negotiate meaning—promotes second language acquisition. It extends the Input Hypothesis by explaining how input becomes comprehensible.
Core Claim
"Negotiation for meaning, and especially negotiation work that triggers interactional adjustments by the NS or more competent interlocutor, facilitates acquisition because it connects input, internal learner capacities, particularly selective attention, and output in productive ways." — Long (1996)
Development of the Theory
Long (1983) - Original Version
- Comprehensible [[Input|Comprehensible input]] is necessary for acquisition
- Interaction provides comprehensible input
- Native speakers modify their speech when talking to learners
Long (1996) - Revised Version
Added:
- Role of output (from Swain)
- Role of attention and noticing (from Schmidt)
- Importance of negative feedback, especially recasts
How Interaction Helps
When communication breaks down, speakers negotiate:
Communication problem → Negotiation → Modified [input](/terms/input)/output → AcquisitionNegotiation Strategies
| Strategy | Definition | Example |
|---|---|---|
| Clarification request | Asking for explanation | "Sorry, what do you mean?" |
| Confirmation check | Verifying understanding | "You said Thursday?" |
| Comprehension check | Checking listener understood | "Does that make sense?" |
| Repetition | Repeating for clarity | "The RED one, red." |
| Recast | Reformulating learner error | L: "He goed." NS: "Yes, he went." |
Why Negotiation Promotes Learning
- Makes input comprehensible: Adjustments help learners understand
- Draws attention to form: Breakdowns highlight problematic structures
- Provides negative evidence: Recasts show what's ungrammatical
- Pushes output: Learners must clarify their own production
The Role of Recasts
Long's revised hypothesis emphasizes recasts—reformulations that correct while maintaining meaning:
Learner: "Yesterday I go to the store." Native: "Oh, you went to the store?"
Recasts provide:
- Positive evidence (correct form)
- Negative evidence (implicit correction)
- Without disrupting communication
Research Evidence
Long (1983) compared:
- Native-native speaker conversations
- Native-nonnative speaker conversations
NS-NNS pairs showed significantly more:
- Confirmation checks
- Comprehension checks
- Clarification requests
- Topic shifts and repetitions
This "negotiation work" makes input accessible.
Interaction vs. Input Hypothesis
| Input Hypothesis (Krashen) | Interaction Hypothesis (Long) |
|---|---|
| Input alone is sufficient | Interaction makes input work |
| Output not necessary | Output plays a role |
| Mechanism unexplained | Negotiation is the mechanism |
| Individual processing | Social exchange |
Long doesn't reject input's importance—he fills in the gap about how it works.
Criticisms
- Lab vs. classroom: Much research in controlled settings
- Silent learners succeed: Some acquire through listening alone
- Cultural variation: Negotiation patterns differ across cultures
- Quality over quantity: Not all interaction is equally helpful
Classroom Applications
Design tasks that require genuine communication:
| Task Feature | Why It Helps |
|---|---|
| Information gap | Forces exchange of information |
| Two-way requirement | Both partners must participate |
| Convergent goal | Must reach agreement |
| Closed outcome | Specific answer required |
Example Tasks
- Spot-the-difference pictures
- Jigsaw readings
- Map directions
- Problem-solving with distributed information
Related Notes
- Michael Long - Developer of the Interaction Hypothesis
- Interactionist Theory - Broader theoretical framework
- Input Hypothesis - Krashen's foundational theory
- Output Hypothesis - Swain's complementary hypothesis
- Noticing Hypothesis - Schmidt's attention account
- Task-Based Language Teaching - Methodology aligned with interaction