📘 Lecture 19 — Input and Interaction in Second Language Acquisition
📖 Overview: This lecture examines the role of input and interaction in second language acquisition, building on foundational theories such as Krashen's Input Hypothesis and Long's Interaction Hypothesis. It critically evaluates these models and explores how social interaction and comprehensible input facilitate language learning, making it essential for understanding modern SLA pedagogy.
🗂️ Topics Covered
The lecture covers an overview of input in SLA, Krashen's Input Hypothesis and its critique, the Interaction Hypothesis and other alternative psycholinguistic theories, and detailed discussion of interaction in SLA including Long's 'Interaction Hypothesis'. It also addresses cross-cultural perspectives on child-directed speech and the role of negotiation of meaning in language acquisition.
📝 Lecture Summary
An Overview of the Field; Input in Second Language Acquisition: Krashen's Input Hypothesis
In cultures where child directed speech (CDS) of the Western type is rare or absent, children's early utterances frequently include partial imitation and rote-learned segments. Cross-cultural study of child language development supports the idea that children will only learn to talk in an environment which they can make some sense of. On the other hand, they can learn to talk in a much wider variety of environments (Lieven, 1994). The normally developing child is well buffered against variation in the input. Buffering implies either a small amount of social support or any of several different environmental events (Snow, 1994). Researchers agree that multi-dimensional models of acquisition are necessary, and the way forward in clarifying how input and interaction may be facilitating language acquisition lies at present in close analysis.
🔑 Definition — Comprehensible input: Input that is both necessary and sufficient for second language learning to take place; language that the learner understands. 📐 Formula: Input → Understanding → Acquisition. Humans acquire language in only one way – by understanding or by receiving 'comprehensive input.' Speaking is a result of acquisition, not its cause. If input is understood and there is enough of it, the necessary grammar is automatically provided.
💡 Why this matters: This hypothesis fundamentally shapes how language teachers think about providing accessible yet challenging language material to learners.
Critique of Krashen’s Input Hypothesis
The Input hypothesis is supported by rather little empirical evidence and is not easily testable. The concepts of 'understanding' and 'noticing gap' are not clearly operationalized and constantly proposed. The processes whereby language in social context is analyzed and new elements are identified and processed by language acquisition device are imprecise.
🔑 Definition — Noticing gap: The cognitive awareness of the difference between one's current language production and target language forms, a concept criticized for lack of clear operationalization.
Research Traditions Inspired by Krashen
Among the research traditions that took inspiration from Krashen's proposals are:
- The Interaction hypothesis
- The Output hypothesis
Some alternative psycholinguistic theories and claims are:
- Noticing hypothesis – emphasizes conscious awareness of linguistic features in input
- Input processing hypothesis – focuses on how learners process input based on limited processing capacity
- Autonomous induction hypothesis – proposes that acquisition is driven by innate mechanisms independent of explicit instruction
🔑 Definition — Output hypothesis: The proposal that producing language (speaking/writing) is not just a result of acquisition but an active part of it, pushing learners to process language more deeply.
Interaction in Second Language Acquisition
Various studies during the 1970s and 1980s demonstrated that the talk addressed to learners was typically grammatically regular but somewhat simplified linguistically. The degree of simplification reported in many descriptive studies was puzzlingly variable. Also, these studies stopped short at description of distinctive features of 'Foreigner Talk Discourse'. Long proposed a more systematic approach to linking features of environmental language and learners' second language. Long proposed his Interaction hypothesis as an extension of Krashen's original Input hypothesis.
Following on Long's original studies, many others used a similar taxonomy and taught us a good deal about the types of task that are likely to promote extensive negotiation of meaning, inside and outside the classroom. They also demonstrated that negotiation of meaning occurs between non-native speaker peers as well as between more fluent and less fluent speakers, given the right task conditions.
🔑 Definition — Foreigner Talk Discourse: Modified speech (simplified vocabulary, slower rate, clearer articulation) that native speakers use when addressing non-native speakers. 🔑 Definition — Negotiation of meaning: Interactive modification of speech between speakers to achieve mutual understanding, involving checks for comprehension, clarification requests, and repetition.
⭐ Key Takeaways
The Input Hypothesis claims comprehensible input is necessary and sufficient for acquisition, but empirical support is weak and concepts like 'understanding' and 'noticing gap' lack clear definition. Alternative models including the Interaction Hypothesis, Output Hypothesis, Noticing Hypothesis, and Input Processing Hypothesis offer more nuanced accounts. Interaction, especially negotiation of meaning, emerges as critical for driving acquisition in both NS-NNS and NNS-NNS dyads. Child-directed speech shows that learners are buffered against input variation, suggesting multiple paths to acquisition exist across cultures. Task type (e.g., requiring information exchange) crucially determines how much negotiation of meaning occurs.
🧠 Quick Revision Questions
- What is Krashen's Input Hypothesis, and what does it claim about the relationship between comprehensible input and acquisition?
- Why has Krashen's Input Hypothesis been criticized, particularly regarding the concepts of 'understanding' and 'noticing gap'?
- How does Long's Interaction Hypothesis differ from Krashen's original Input Hypothesis?
- What is 'negotiation of meaning,' and what kinds of tasks are most likely to promote it between learners?
- What does cross-cultural research on child-directed speech say about the necessity of Western-style CDS for language acquisition?
📘 Lecture 20 — Consciousness Raising, Negative Feedback and Negative Evidence
📖 Overview: This lecture examines how conscious attention, corrective feedback, and negative evidence contribute to second language acquisition. It explores the reformulated Interaction Hypothesis, the Output Hypothesis, and the role of recasts and negative feedback in both first and second language learning contexts, addressing ongoing debates about whether negative evidence is necessary for language development.
🗂️ Topics Covered
The lecture covers noticing, consciousness raising and attention as learner-internal factors; Rethinking the Interaction Hypothesis with emphasis on selective attention; Output in Second Language Acquisition including Swain's three functions; Feedback, Recasts and Negative Evidence in First Language Acquisition with nativist perspectives; Negative Feedback and Recasts in child-directed speech; and Negative Feedback and Recasts in the Second Language Classroom based on observational studies by Lyster and colleagues.
📝 Lecture Summary
Noticing, Consciousness Raising and Attention; Rethinking the Interaction Hypothesis
Mackey's (1999) study provided the clearest evidence that taking part in interaction can facilitate second language development. However, somewhat contradictory findings showed a need for stronger theoretical models clarifying the claimed link between interaction and acquisition. Researchers appealed to add the ideas of noticing, consciousness-raising, and attention to acquisition models. Earlier interactionist research was criticized for being one-sidedly pre-occupied with functional aspects of SL interaction and neglecting linguistic theory.
Long's eventual reformulation of the Interaction Hypothesis places more emphasis on linking features of input and the linguistic environment with learner-internal factors and explaining how such linkages may facilitate subsequent language development. This new version highlights the possible contribution of negative evidence to SLL and clarifies the processes by which input becomes intake through introducing the notion of selective attention.
💡 Why this matters: This reformulation bridges the gap between external linguistic input and internal cognitive processing, providing a more complete theoretical framework for understanding how interaction leads to acquisition.
🔑 Definition — Interaction Hypothesis (reformulated): A theory proposing that features of input and the linguistic environment linked with learner-internal factors through selective attention facilitate subsequent language development.
🔑 Definition — Selective Attention: The cognitive process by which learners consciously focus on specific features of input, enabling input to become intake.
Output in Second Language Acquisition: Output Hypothesis
Many researchers consider output as necessary to increase fluency and emphasize practice. However, the Output Hypothesis by Swain (1985, 1995) goes beyond this "practice" function and proposes three further functions for learner output:
- The noticing/triggering function (consciousness-raising role)
- The hypothesis-testing function
- The metalinguistic function (reflective role)
Swain believes that the activity of producing target language may push the learner to become aware of gaps and problems in their current SL system, provide them with opportunities to reflect on, discuss and analyze problems explicitly, and offer opportunities to experiment with new structures and forms.
🔑 Definition — Output Hypothesis: A theory proposing that learner output serves not only for practice, but also triggers noticing of gaps, allows hypothesis testing, and enables metalinguistic reflection.
📐 Three Functions of Output:
- Noticing/triggering → Learners become aware of what they cannot say
- Hypothesis-testing → Learners try out new forms and structures
- Metalinguistic → Learners reflect on and analyze language explicitly
Feedback, Recasts, and Negative Evidence in First Language Acquisition
These are increasingly important topics in debates. Theorists arguing for a strongly Innatist Model claimed that language is simply not learnable from normal input that provides positive evidence and lacks negative evidence. Without negative evidence, learners cannot discover the limits and boundaries of the language system. On the other hand, Nativists find the answer in some form of Universal Grammar to eliminate many possible generalizations about language structure that are compatible with the input received but are actually incorrect.
🔑 Definition — Positive Evidence: Input that shows what is grammatical in a language (e.g., actual utterances heard by the learner).
🔑 Definition — Negative Evidence: Information about what is ungrammatical or not possible in a language, either explicit (corrections) or implicit (e.g., recasts).
We previously saw that negative evidence is much more prevalent in child-directed speech. Caretakers' recasts offer implicit negative evidence about children's interim grammatical hypotheses. It remains still less clear whether negative evidence is necessary for the acquisition of core aspects of language or not.
🔑 Definition — Recast: A reformulation of a learner's erroneous utterance that preserves the original meaning while correcting the error (e.g., Child: "He goed there." Adult: "Yes, he went there.")
In light of the first language debate, related questions can be asked about the role of negative evidence in SLL, such as: To what extent does indirect negative evidence influence SL learners? And to what extent do they notice and make use of this evidence?
Oliver's (1995) study provided significant findings on these questions. It indicated that more than 60% of errors by non-native speaker children received negative feedback from their native speaker partners; the most frequent were negotiation of some kind. Recasts usually occurred in response to utterances containing single errors in association with particular types of grammar mistakes.
📌 Example — Oliver's (1995) findings: In interactions between native speaker and non-native speaker children, over 60% of errors triggered some form of negative feedback, with negotiation being the most common type. Recasts were specifically associated with utterances containing single grammatical errors.
Negative Feedback and Recasts in the Second Language Classroom
Further observational studies examined the occurrence and apparent effects of negative feedback in SL classrooms. For instance, Lyster and colleagues (1997) noted that recasts were the most common type of feedback. However, recasts were much less likely to lead to immediate self-correction by the students.
A further analysis reported that feedback provided by the teachers varied according to the type of error that had been made:
- Lexical errors → Teachers responded with negotiation (e.g., clarification requests)
- Grammatical errors → Teachers responded with recasts
- Phonological errors → Teachers responded with recasts
Recasts were seen as an effective strategy for phonological errors. However, recasting was much less effective for repair of grammatical mistakes. The study suggested that more interactive feedback modes would be more effective in pushing learners to amend their hypotheses about SL grammar as well as vocabulary.
📌 Example — Lyster (1997) classroom findings: A teacher might respond to a phonological error like "I tink" with a recast "I think" which effectively leads to repair. However, responding to a grammatical error like "He go to school" with the recast "He goes to school" was less likely to lead to student self-correction than using negotiation strategies like clarification requests.
⭐ Key Takeaways
The reformulated Interaction Hypothesis highlights selective attention as the mechanism linking input features to learner-internal factors, explaining how input becomes intake. Swain's Output Hypothesis proposes three crucial functions of output beyond practice: noticing gaps, testing hypotheses, and metalinguistic reflection. The debate over negative evidence remains unresolved in first language acquisition, though caretaker recasts provide implicit negative evidence. Oliver's study showed over 60% of child L2 errors receive negative feedback, primarily through negotiation. Lyster's classroom research revealed that while recasts are the most common feedback type, they are less effective for grammatical repair than more interactive modes like negotiation, suggesting teachers should vary feedback strategies according to error type.
🧠 Quick Revision Questions
-
What are the three functions of output proposed by Swain's Output Hypothesis, and how does each contribute to second language acquisition?
-
How does Long's reformulated Interaction Hypothesis differ from earlier versions, and what role does selective attention play in this reformulation?
-
According to the nativist perspective, why might negative evidence be necessary for language acquisition, and how does Universal Grammar offer an alternative explanation?
-
What did Oliver's (1995) study reveal about the frequency and types of negative feedback received by non-native speaker children during interaction?
-
Based on Lyster and colleagues' (1997) classroom research, why are recasts more effective for phonological errors than for grammatical errors, and what alternative feedback strategies might be more effective for grammar?
📘 Lecture 21 — Input Processing and Autonomous Induction Theory
📖 Overview: This lecture examines two major theoretical models that explain how second language input is converted into intake for learning: Bill Van Patten’s Input Processing theory and Suzanne Carroll’s Autonomous Induction theory. It also discusses the role of attention, noticing, and consciousness-raising in second language acquisition, providing a critical lens on interactionist research.
🗂️ Topics Covered
This lecture covers attention, consciousness-raising, and ‘Focus on Form’; theorizing input and interaction research; Input Processing theory including its principles and critique; and Autonomous Induction theory along with its critique. It explores how researchers have linked attention types—especially noticing—to the conversion of input to intake, and presents two contrasting models for explaining internal linguistic processing in SLA.
📝 Lecture Summary
Attention, Consciousness-raising and 'Focus on Form'
Researchers have developed the idea that the amount of attention may influence the extent to which SL input and interaction produce SL intake. Schmidt (1990, 1994, 2001) has been most influential in promoting this view. He distinguishes among different types of attention. He uses the term ‘noticing’ to refer to the process of bringing some stimulus into focal attention and registering its simple occurrence, whether voluntarily or involuntarily. He reserves the terms ‘understanding’ and ‘awareness’ for explicit knowledge: ‘awareness of a rule or generalization.’ Schmidt is optimistic about the contribution of both kinds of attention. His main evidence supports the significance of noticing. He originally argued that noticing is the necessary and sufficient condition for conversion of input to intake for learning; later, he modified this view to the claim that: more noticing leads to more learning.
The accuracy of recast reproduction depends on the following factors:
- Learners’ language level
- Length of the recast
- Number of corrections
Theorizing Input and Interaction Research
Survey of input, output, and interaction research showed the descriptive nature of research; attempts to link different types of second language usage with SLL have produced mixed results. Many commentators have argued that stronger theorizing is required for interaction studies to progress. Interactionist researchers are increasingly interested in modelling internal linguistic and psycholinguistic factors. However, no detailed models of language processing have been proposed by any of interactionist researchers. Here, we will briefly discuss two models: Input processing theory and Autonomous induction theory.
Input Processing
Input Processing theory by Bill Van Patten and associates (1996, 2002) is well known because of an associated research program on language pedagogy, ‘processing instruction.’ It attempts to explain how environmental second language input becomes converted into intake. It does not offer a complete model of these processes; instead, it offers a set of ‘principles’ to explain apparent failure of second language to completely process the linguistic forms encountered in SL input.
Input Processing Principles:
Principle No.1: Learners have preferences for semantic processing over morphological processing. 📌 Example: ‘We travelled by train yesterday.’
Principle No. 2: The learners have a preference for beginning and ending words. 📌 Example: Learners give importance to the beginning and ending words of recast/feedback.
This approach has led to a series of pedagogical experiments. Input processing theory is primarily focused on explaining the shortcuts and restricted processing strategies which learners seem to use.
Input Processing: Critique
Input processing theory is primarily focused on explaining the shortcuts and restricted processing strategies which learners seem to use.
Intake in Light of Input Processing Theory by Bill Van Patten:
- Linguistic data is actually processed from the input.
- It is held in working memory for further processing.
- Input processing theory does not offer a complete model of successful processing input.
- It offers a set of principles.
- It explains the apparent failures of learners to process the input.
Autonomous Induction Theory
Autonomous Induction theory by Suzanne Carroll (2000) offers a much more complete and ambitious model of the above mentioned processes. According to Carroll, understanding of SLA processes requires an adequate theory of the following phenomena:
- Representation of language in the mind
- How the language is processed both receptively and productively
- How our mental representations of language can be changed to process the environmental language we encounter
Carroll rejects parameter (re)setting as a totally inadequate metaphor. Instead, she proposes inductive learning (i-learning). Inductive learning refers to learning by generalization from examples. Carroll challenges the claim that increased comprehension can lead to identification and acquisition in a sequential manner. According to her, it is logically impossible. Unless enough formal analysis is done so that elements are identified in the speech stimulus, there is no way of generating interpretation of its meaning.
Autonomous Induction Theory: Critique
Carroll’s model is complex. It’s been criticized for its neglect of the detail of language processing which converts language stimuli into interpretable input.
💡 Why this matters: Carroll’s critique challenges the widely held assumption that comprehension automatically leads to acquisition, arguing instead that formal analysis of input elements must occur first.
⭐ Key Takeaways
This lecture establishes that attention—specifically Schmidt’s concept of noticing—is critical for converting input to intake, though the claim was later softened from “necessary and sufficient” to “more noticing leads to more learning.” Van Patten’s Input Processing theory offers a set of principles (semantic over morphological processing, preference for word beginnings and endings) to explain learner processing failures but does not provide a complete processing model. Carroll’s Autonomous Induction theory provides a more ambitious model, rejecting parameter resetting in favor of inductive learning from examples, and arguing that formal analysis must precede comprehension. Both models are criticized—Input Processing for its limited scope, and Autonomous Induction for neglecting details of converting stimuli into interpretable input. Students must remember that neither model fully explains SLA processing but each offers valuable insights into learner strategies and cognitive constraints.
🧠 Quick Revision Questions
- According to Schmidt, what is the difference between “noticing” and “understanding,” and how did his claim about noticing’s role in SLA change over time?
- What are the three factors that influence the accuracy of recast reproduction according to Philip (2003)?
- State the two principles of Van Patten’s Input Processing theory and provide an example for each.
- According to Carroll’s Autonomous Induction theory, what three phenomena must an adequate theory of SLA explain, and what learning mechanism does she propose instead of parameter resetting?
- What is the main criticism leveled against Carroll’s Autonomous Induction theory regarding language processing?
📘 Lecture 22 — Interactionist Approach and Sociocultural Theory
📖 Overview: This lecture evaluates the Interactionist Approach to Second Language Acquisition, detailing its achievements and limitations. It then introduces Sociocultural Theory, a perspective that views language learning as a fundamentally social and mediated process, drawing on the work of Vygotsky and his followers. Understanding these contrasting views is crucial for grasping how environmental factors and social interaction shape language development.
🗂️ Topics Covered
This lecture evaluates the scope of the Interactionist Approach, covering its achievements in documenting meaning negotiation and relating environmental factors to linguistic theory. It then lists the key limitations of this tradition, such as its Western-centric focus and incomplete modeling. The final section introduces Sociocultural Perspectives, explaining Vygotsky’s theory that learning is a mediated process involving symbolic tools and social interaction, and lists his leading supporters.
📝 Lecture Summary
Evaluation: The Scope of Interactionist Approach
The Input, Output, and Interaction hypotheses have generated significant empirical research. The first research phase focused on documenting the phenomenon of meaning negotiation. The second phase developed by a) relating environmental factors to Universal Grammar, b) increasing interest in negative evidence, and c) giving more attention to information processing theory and the complexities of converting environmental language first into input and then into intake.
Achievements of Interactionist Research
Research in this tradition has demonstrated several key findings. First, native speakers and non-native speakers can and will work actively to achieve mutual understanding. Second, negotiations involve both linguistic and interactional modifications, creating opportunities to 'notice' aspects of the target language form. Third, non-native speakers can attend to, take up, and use language items made available to them. Finally, learners who receive negative feedback on particular structures are significantly advantaged when later tested on those structures.
Limitations of Interactionist Approach
Despite its achievements, this tradition has several important limitations. First, most research has been conducted in Western or Anglophone settings, so more cross-cultural studies are needed to verify that 'negotiation for meaning' is a universal phenomenon. Second, attempts at modeling the interaction between environmental stimuli, the learner's internal language system, and language-specific capabilities remain fragmentary and incomplete. Third, there are few studies that track particular language structures through the processes of instruction, negotiation, and output. Fourth, there is a lack of studies documenting learners' long-term use and control of these language items. Consequently, the approach cannot yet generalize or make powerful predictions about the usefulness of interaction.
Socio-cultural Perspectives on Second Language Learning
🔑 Definition — Sociocultural Theory: A theory originally presented by Soviet psychologist Vygotsky, arguing that humans do not act directly on the physical world but rely on tools and labour activity, using symbolic tools or signs to mediate and regulate their relationships with others. Theorists in this perspective view language learning in essentially social terms, claiming that target language interaction is not simply a source of 'input' but plays a more central role in learning.
From this viewpoint, learning is a mediated process. It is mediated partly through learners' developing use and control of mental tools. It is also seen as socially mediated, meaning it is dependent on face-to-face interaction and shared processes. 💡 Why this matters: This contrasts sharply with the Interactionist view, which treats interaction primarily as a source of data for an internal learning mechanism; Sociocultural Theory places the social interaction itself at the heart of the learning process.
Vygotsky’s Leading Supporters: The leading supporters of this theory include Jerome Bruner, James Wertsch, Barbara Rogoff, James Lantolf, Mercer, and Wells.
⭐ Key Takeaways
The Interactionist Approach has successfully shown that negotiated interaction, including negative feedback, provides crucial opportunities for learners to notice and acquire language forms. However, its limitations include a Western-centric research base and an inability to make powerful predictions about acquisition. In contrast, Sociocultural Theory, rooted in Vygotsky's work, reframes learning as a socially and tool-mediated process, where interaction is not just input but the primary driver of development. The key distinction is that Interactionism sees the learner's internal mechanisms as central, while Sociocultural Theory sees the social context and collaborative process as the core of learning.
🧠 Quick Revision Questions
- What were the two main phases of research within the Input, Output, and Interaction tradition?
- Name three specific achievements of the Interactionist Approach in SLA research.
- What are the four major limitations of the Interactionist Approach listed in the lecture?
- What is the central argument of Vygotsky’s Sociocultural Theory regarding how humans act on the physical world?
- According to Sociocultural Theory, in what two ways is learning a "mediated process"?