7 Types of Language Input that can facilitate Language Acquisition according to research

Introduction: in search of language input to drive second language acquisition

One key question that always occupies the minds of language educators will be “What languaging activities do I want to organise or facilitate in my class today to achieve the best language learning outcomes?”. Essentially, we are actually considering the most optimal design of language exposure in our classrooms, and making decisions on what learners should be listening to, speaking about, reading or writing. In other words, we are considering the types of language input to activate language learning in our learners.

I remembered vividly of my earlier experience as a pre-service teacher. I had a strong disposition that language learning should be communicative in nature, where the core of classroom activities should be interactive. Learners should be actively using the target language, with the teacher facilitating conversations.

While such a disposition can be valid, I do recall the few trial lessons I had with the teenage learners in my classroom. They were rather heterogeneous, with varying language backgrounds and proficiency levels. A number of them could only cope with the language at elementary level.

Insensitive to these conditions, I tried to facilitate a discussion on their experience with different sports in one of the lessons. Well, the keyword is “tried”, because there was hardly any discussion, but more of a monologue of instructions and guiding questions. Only a few learners were responsive, while a number were struggling to even find suitable sentence starters. It was a big flop.

Now there can be many learning points from this experience. One important point, which I liked to emphasise in this post though, is that I did little to consider the prior knowledge of my learners to provide a suitable amount of input, both content and linguistic, to scaffold the learning process. My ignorance of the key types of input that are important to activate language learning contributed to the flop.

The nature of “Language Input” for Language Acquisition

INPUT is a vital concept within language acquisition, especially within second language acquisition. Theoretical models in research on language acquisition typically require the explanation of how INPUT is treated by the learners and the impact on language acquisition. In technical terms, there must be some discussion, supported by empirical evidence, on how learners form or change their mental representation of the grammar(s) in the target language.

Generally, LANGUAGE INPUT of a target language refers to everything in the “language which learner is exposed either orally or visually” (Gass, 1997:1). In other words, it is everything that the learner hears or reads in that language. LANGUAGE INPUT provides the necessary positive evidence of how meaning-bearing content can be expressed in a target language.

The presence of INPUT does not guarantee acquisition. This is generally understood. Watching 20 hours of Korean drama does not necessarily lead to acquisition of Korean – if that works, we would have loads of proficient Korean speakers around the world now!

Notwithstanding that, we also understand that the absence of target language INPUT also implies no language acquisition will take place for the target language. This should also be easily understood. It cannot be the case that a person who has never listened to a word of Yiddish, to be able to speak Yiddish suddenly.

There are various theories on the psycholinguistic processes of how INPUT generates acquisition. Nevertheless, most scholars agree that INPUT needs to be converted to INTAKE before acquisition can take place. Using the analogy of the Korean drama earlier, we may also have heard of examples of learners who seemed to have acquire new understanding and able to use Korean at a slightly different level after watching numerous Korean dramas. Yet, the same experience does not always hold for all others. In that sense, for those who have improved or level up, INPUT has become INTAKE in their minds.

Technically speaking, INPUT is akin to data (in fact, some scholars define input as primary linguistic data) available in the environment that the learner has opportunity of access, and intake refers to the data that the learner actually takes in and process. Just as in data science where processed raw data then becomes useful information, LANGUAGE INPUT that is processed can then become information for acquisition – the psychological region where “new information is matched against prior knowledge” (Gass, 1997:5), enabling the learner to confirm/reject any previous hypothesis or form new hypothesis about the target language. Note that this can be implicit and is unknown to the learner.

The key question then is: What types of LANGUAGE INPUT can then have a higher opportunity of becoming INTAKE? This is one question of inquiry that research has yet to provide very conclusive answers, although there are many influencing factors. As such, the various types of LANGUAGE INPUT to be discussed contribute in different manners to acquisition. We should take note of these types to construct relevant languaging experiences for our learners.

1. Naturalistic Input

NATURALISTIC INPUT is probably the default among the various types of language input. NATURALISTIC INPUT refers to language input received under naturalistic conditions. Basically, it is language to which learners are exposed in the “real world”, where the language is usually unfiltered, meaning-focused and can be highly variable, subject to the diversity of interlocutors the learners come in contact with. Such input is not primarily produced with the intent to support language development, but to perform communicative functions in the actual lives of the parties involved.

In simpler terms, NATURALISTIC INPUT comes through languaging in the daily lives of the learners, where the target language is used for all sort of purposes beyond education – be it transactions at the grocery stores, listening passively to radio programmes, watching online videos, part of religious rituals or engaging in casual conversations with family and friends. The learners, and us as language educators, have little control over what those situations be like and the features of actual language that are used.

NATURALISTIC INPUT is somewhat linked to the notion of authenticity (although this is a highly contested construct). The value of authenticity can mean different things to different people. For language educators and learners who defines language generally as a tool for communication, learning characterised by authenticity makes a lot of sense as that is the real litmus test of language learning outcomes, where the learners ascertain whether they are able to transfer what is taught in controlled conditions, such as the classrooms, to actual situations where they will use the language organically. From the perspective of linguistic input, NATURALISTIC INPUT challenges learners to see if they really can listen and read in the target language in the real world.

However, given the features as described, NATURALISTIC INPUT may or may not support the learning of the target language – at least not necessarily aligned with the intended learning outcomes of a standardised language in an institutional language classroom. Taking reference again from the Korean drama example earlier, the NATURALISTIC INPUT may be packed with age-inappropriate features (e.g. profanities, adult vocabulary), high variability (e.g. different people using different varieties of the same language and thus not modelling the target features of the standardised language) and limited contexts (e.g. only in informal casual use). We cannot assume that desired language learning outcomes are the same for all of us, as different learners can have different needs and motivations for language learning. As such, despite the benefits of naturalistic input, we do need to be cognisant of means and ways to deal with challenges from NATURALISTIC INPUT.

For us, it would be good to have a good sensing of the NATURALISTIC INPUT to which our language learners are exposed. To what extent do they have NATURALISTIC INPUT of the target language? What observations might they have of the target language used in their daily lives?  How similar or different is the naturalistic input from the desired outcomes of the target language learning? These are just a few good guiding questions for us to consider with reference to NATURALISTIC INPUT.

One last point to be highlighted is that NATURALISTIC INPUT can be present in our language classrooms, such as during breakout discussions and peer learning activities. NATURALISTIC INPUT can also be “engineered” into our lesson design and enactment by institutionalising some form of authentic learning. So, we can and should include this concept in our repertoire when doing lesson planning.

2. Comprehensible Input

The term “COMPREHENSIBLE INPUT” was coined by Stephen Krashen (1981, 1982, 1985) as part the influential “INPUT HYPOTHESIS” (later re-termed as Comprehension Hypothesis). Among the different types of language input, this is also probably the one with the most influence on language acquisition research. Generally, COMPREHENSIBLE INPUT refers to the subset of language input received by the learner where the content or meaning is “readily understood” (Piske & Young-Scholten, 2009b).

The main premise of the INPUT HYPOTHESIS is that input can only be converted to intake when it is comprehensible, based on the argument that learners are unable to process data which is incomprehensible. Unsurprisingly, if we listen to strings of words with no idea of what is spoken/written, we are unable to map these linguistic forms (across the different components of the language system) to anything useful for transfer to other contexts of language use.

Another important point is that there is potential progress in competence (e.g. thus something linguistic is acquired), when the language input is pitched at a level slightly higher than the current level of the learner. This is represented by the formula “i = i + 1”, where the first i refers to the acquired interlanguage and “+1” the intended increment that is possible. This is somewhat like the idea of Vygotsky’s Zone of Proximal Development. The paradox that lies in the comprehension of language input that is slightly beyond the current level of the learner is resolved through using “context, knowledge of the world and other extralinguistic information” (Krashen, 1982).

The INPUT HYPOTHESIS and the notion of COMPREHENSIBLE INPUT are not without their criticisms. The “i” in “i + 1” is hardly measurable – there have been no convincing proposals to that. Neither is “+1” ascertainable, and is more of a post-intervention conclusion (e.g. whether some input fulfils that requirement) rather than a pre-intervention prediction. Without proper measures in place, the scientific claims can hardly be validated or falsified. In other words, it is neither useful to inform research nor practical to guide language instruction.

Irrespective of that, the constructs afforded by the INPUT HYPOTHESIS and COMPREHENSIBLE INPUT have highlighted the importance of language input, so much so that it becomes a prominent object of inquiry within language acquisition research. The notion of “i + 1” also provides an easily understood frame for researchers and educators to assess the relative value of various types of language input.

On top of this, the greatest takeaway from the concept of “COMPREHENSIBLE INPUT” and its related discussions is the perspective that not all language input is made equal when it comes to their impact on language acquisition. It also foregrounds a possible challenge that learners may be cognitively overloaded when they have to process both form and meaning in the acquisition process.

3. Interactional Input

INTERACTIONAL INPUT, as the name suggests, refers to language input that comes through interactions with others (e.g. daily transactions, casual conversation, group discussion, question and answer, live debates, social media). INTERACTIONAL INPUT can be naturalistic in nature if the language is provided under naturalistic conditions (e.g. not created for the purpose of language learning, no instruction provided); and can also be non-naturalistic if the input is designed for teaching and learning (e.g. planned conversational pairs, scaffolded dialogue in a foreign language classroom).

Regardless of whether it is naturalistic or not, the meaning of INTERACTIONAL INPUT can be better understood as there are opportunities of negotiation and clarification. In other words, the input can be made more comprehensible through more conversational turns. In fact, meaning can be negotiated until all interlocutors fully understand the meaning.

For those who have the privilege of travelling to regions or countries using a target language in which one is less proficient, we probably have no lack of experience engaging in conversations with the natives using a lot more conversational turns just to clarify. In one of my earlier trips to Japan in 2004, I spoke the Japanese language minimally. The use of Google Maps and international mobile data was not such a pervasive thing back then. In addition, I was just a university student trying to cut down on budget. I relied very much on the conventional way of asking natives for directions, and it was not always the case that the natives I chanced upon know English well enough. As such, I had to negotiate meaning (e.g. seek clarification, doing confirmation checks, asking for repetition) to then understand the input I received. My interlocutors also adjusted their language upon knowing that I do not speak Japanese proficiently.

The opportunity for negotiation in interaction, and thus an opportunity to make further sense of the meaning in the incoming language input, provides the chance for learners to notice the linguistic forms to map onto the meaning in the interactional context (e.g. the sound of the directional words onto the directions).

The potential of INTERACTIONAL INPUT in facilitating language acquisition was largely credited to Michael Long (1981) through the “INTERACTION HYPOTHESIS”. In Long’s article, the inspiration came from the observations of how native speakers interact with non-native speakers where they tend to simplify the input to become more comprehensible, thus providing opportunities of learning for the learners. The native speakers provided the INTERACTIONAL INPUT as a model for the non-native learners, and the non-native learners could negotiate the meaning with the natives to then process the input for learning.

However, over the years, different scholars have also criticised this proposal. Firstly, whether INTERACTIONAL INPUT facilitates acquisition is subject to the quality of the interactions. At one end, there are interactions where the native speakers provided even more complex or overwhelming replies in response to clarification requests, so much so that non-native learners cannot process the input at all. On the other end, there are native speakers that made the reply so “comprehensible” that there is no delta that provides the “+1” learning pathway.

Secondly, it is not always the case where the desired model of language is the one used in interactions. I have shared earlier under naturalistic input that interlocutors may use a different variety of the same language which may thus further confuse the learners. Furthermore, interlocutors in our classrooms also involve other learners who have not yet completely mastered the language, and they also provide input in the target language – with characteristics of an interlanguage. Learners may process all these as part of the learning.

The discussion on INTERACTIONAL INPUT is a reminder to us that all language that happens in our lessons (or in any interactions with our learners in the target language) can potentially lead to language acquisition or impede it. This includes instances of us giving instructions, doing classroom management, and providing facilitation; and our learners engaged in interactional practices, classroom presentations and discussions, and online social media interactions. Our main motivation then lies in leveraging opportunities and mitigating challenges that come with it.

4. Modified Input – Simplification

MODIFIED INPUT refers to language input that is altered for a particular purpose (not necessarily to facilitate language acquisition). Modification can go in different directions: simplification or elaboration. The current type we are talking about is SIMPLIFIED INPUT.

My earlier example of interaction with the Japanese during my trip in Japan illustrates a communicative situation where the native speaker (e.g. the Japanese from whom I sought assistance) simplifies the linguistic form of the message so that the interlocutor can understand. In language research, this is the type of SIMPLIFIED INPUT termed as “FOREIGNER TALK”, where native speaker simplify the language they use when communicating with non-native speakers.

Another group of prominent SIMPLIFIED INPUT belongs to “BABY TALK”. This are situations where caregivers of young children, usually the parents or grandparents, simplify their language to express meaning to infants, toddlers and juniors. Typical examples include the use of a different prosody (e.g. higher and wider pitch) and simplified speech units (e.g. “choo-choo” for trains, “tum-tum” for tummy). The natural tendency to engage in “BABY TALK” in such situations usually arise from a communicative motivation but can be very useful for infants and toddlers as they build stronger phonological awareness of languages in general. On a side note, “BABY TALK” can also happen to pets and between adults for a different purpose (e.g. derogatory, flirtatious).

Unsurprisingly, “TEACHER TALK” is the version of SIMPLIFIED INPUT in learning contexts and can take place between all teacher-learner interactions – just think about how early childhood educators speak to the pre-schoolers. In the context of language education, “TEACHER TALK” can then refer to simplification of language input for the message or content to be understood by the students.

It is important to note that “TEACHER TALK” need not always be directed at younger children, and can also be used for teenage and adult language classrooms. Of course, with older learners, we do not go “choo-choo trains” or “itsy witsy spider” but we can definitely still go at a slower rate or paraphrase with simpler words or use more extralinguistic cues (e.g. body gestures).

The premise of MODIFIED INPUT for simplification is that it makes the language input comprehensible, as described in the Input Hypothesis. This is the case for many language learning situations we imagine, where the authentic texts that we intend to use or the communicative contexts that we hope to emulate (i.e. naturalistic input) surpass the current level of our learners beyond “i + 1”.

What happens then, when we do hope to stretch our learners who are already performing at a certain proficiency level (e.g. B2 or C1 of the Common European Framework of Reference)? At this point, the learners’ linguistic learning targets have become more complex and may be focused on the deeper layers of morphology and syntax, which may not be that frequent in naturalistic input. In this case, the language input may then be modified in the other direction: elaboration.

5. Modified Input – Elaboration

Interestingly, the premise of MODIFIED INPUT for elaboration is also related to the Input Hypothesis with respect to “i + 1”. In these cases, the “i” of learners is already at a certain level, this requires “i + 1” to be pitched even higher, which may not be commonly found during daily communicative acts. An example of such is a fast-moving debate on sophisticated topics with highly complex structures. This is where MODIFIED INPUT in the elaborated direction will be useful to drive language acquisition for our learners.

Something not discussed earlier was the starting point of modification. It usually comes with an actual base speech/text, usually of naturalistic input where the speaker or writer (or signer for sign languages) produce the speech/text centred on a purpose of communication without deliberate modification. In an interactive mode, it can also refer to a natural flowing speech/written communication where the person producing the language input does it without modification in mind.

ELABORATED INPUT thus takes the default speech/text and alter to include more complicated linguistic forms, or nudge the person producing the language input to construct a similar message in more complex ways. While such an approach may be challenged by asking educators to seek out naturalistic input that fulfils the “i + 1” conditions, it is nevertheless useful since naturalistic input is usually beyond our control and may not fulfil the Goldilocks Principle to the optimal degree. Think about an instance when you found a speech/text with interesting content that will entice your learners but with linguistic features that are too simple for your learners. Such input then requires modification in the direction of elaboration.

While ELABORATED INPUT is usually positioned to increase the difficulty level of a speech/text, it can also work the other way to make the speech/text simpler, thus making it more comprehensible. This can make ELABORATED INPUT also an alternative to simplified input. How does that happen?

The argument put forward by some scholars is that elaboration can also be a way of simplification, whereby more details of the naturalistic input are given with the use of simple vocabulary and sentence structures. In comparison, this approach (instead of conventional simplification) maintains a larger space to expose learners to native-like linguistic features. In other words, an elaboration gives more input in quantity which models the target language further (and provides more data) for the learners.

6. Enhanced Input

ENHANCED INPUT is very much related to modified input, though it takes a very different perspective. The concept of INPUT ENHANCEMENT was introduced by Michael Sharwood Smith (1991, 1993) to denote all alterations of the language input to make targeted linguistic features more salient to the learners. The assumption is that learners will have a higher probability of noticing those features and the increased attention can then activate the input-intake processing. ENHANCED INPUT is thus the product(s) of this process.

Sharwood Smith’s conceptualisation of INPUT ENHANCEMENT was to encompass many different forms of enhancement techniques. However, two major forms eventually came to dominate research and practice: INPUT FLOOD and PERCEPTUAL ENHANCEMENT.

INPUT FLOOD refers to the technique where learners are exposed to the targeted linguistic features in high frequency by re-designing written texts to use targeted linguistic forms repeatedly or embedding many instances of the targeted linguistic form in speech. The rationale is that the high repetition increases the likelihood that learners may notice the forms to then acquire them. Such an approach can be commonly observed in books targeted at pre-schoolers, even if the writers are not consciously using the technique. For example, Dr Seuss’s books tend to reuse certain phonological patterns and word orders over and over again.

PERCEPTUAL ENHANCEMENT, on the other hand, makes use of phonological and typographical cues in attempt to shift attention to the targeted forms. Manipulation examples include bolding, using italics, underlining, highlighting, changing the font size or style, or any combination of these for written texts; or varying intonation, stress and loudness for speech/utterances. It is hoped that the enhanced perceptual cues can draw learners’ attention to the linguistic forms underpinning the manipulated words (or strings of words) by increasing their salience.

Unlike other types of language input which generally converges on providing positive evidence, INPUT ENHANCEMENT may also be implemented to provide negative evidence, usually in the form of feedback, thus drawing learners’ attention to incorrect forms and signalling to them that certain means of expression has deviated from the linguistic norms in the target language.

Despite the theoretical soundness of INPUT ENHANCEMENT as an approach to expedite language acquisition, empirical research has found either mixed effects or small-sized effects from the approaches. As concluded by Benati (2016), there are other factors that may constrain these effects: “proficiency level, the developmental stage and the degree of readiness of the learner, the type of linguistic feature chosen, and the intensity of the treatment”.

Nevertheless, if we are keen to explore input enhancement techniques systematically in our language lessons, we should be mindful of the science underpinning them and possibly the best ways to incorporate them into the design of instructional materials that provide exposure to language.

7. Structured Input

STRUCTURED INPUT is a type of input that is embedded within PROCESSING INSTRUCTION (VanPatten, 1993, 2004, 2009). The concept of PROCESSING INSTRUCTION is predicated on INPUT PROCESSING (VanPatten, 1993) which makes two major claims:

As such, PROCESSING INSTRUCTION seeks to help learners develop optimal language processing habits in the target language by exposing them to “STRUCTURED INPUT” – the type of language input which has been artificially weaved into language tasks where the learner is conditioned to do processing of linguistic forms while still attending to meaning (e.g. rely on word order to interpret meaning). In that sense, learners are forced to establish connections between form and meaning, usually through repeated examples. This will then then increase the likelihood of acquiring the target forms.

Unlike other types of language input, STRUCTURED INPUT is constructed in addressing pre-determined forms. Therefore, the first step in generating STRUCTURED INPUT is to identify the processing problem (e.g. learners’ tendency to rely on lexical items, learners’ confusion due to word order) of the linguistic form in question.  As such, Wong (2004) proposed 5 guidelines in creating STRUCTURED INPUT activities for language teaching:

Conclusion: Charting the linguistic journey of our learners through input management

This article has provided a taxonomy of language input, particularly 7 key types of language input that are highly influential in language education and represent the range of languaging experiences that language teachers can help facilitate or engineer for our learners. Of course, to take this further, we can also examine the characteristics of language input that we can manipulate to achieve the most optimal designs.

Thank you for reading! If you like what you are reading, do subscribe to our mailing list to receive updated resources and tips for language educators. Please also feel free to provide us any feedback or suggestions on content that you would like covered.