Implicit and Explicit Learning

Definition:

Implicit learning is the acquisition of patterns and regularities through passive exposure, without deliberate intention and without awareness of what has been learned; explicit learning is the deliberate, conscious acquisition of knowledge through processes that involve attention, hypothesis-testing, and metalinguistic awareness. In SLA, the implicit/explicit distinction is foundational because it underpins debate about whether adult L2 learners can acquire language through the same unconscious, statistical learning mechanisms that characterize L1 acquisition in children, or whether adults must rely disproportionately on explicit, declarative routes that may or may not convert into usable implicit knowledge. Reber’s (1967) artificial grammar learning paradigm first demonstrated implicit learning in cognitive psychology; DeKeyser (2003) introduced it to SLA with important theorization of the interface.

In-Depth Explanation

Fundamental definitions:

Implicit knowledge: Knowledge that cannot be verbalized, is used automatically in real-time performance, and was acquired without conscious intention to learn. It is procedural in character — knowing how without necessarily knowing that.
Explicit knowledge: Knowledge that can be verbalized (the learner can state a rule), is accessed through deliberate controlled processing, and was acquired through intentional, attended learning. It is declarative — knowing that (e.g., “In Japanese, the verb comes at the end of the clause“).

The distinction maps roughly (though not perfectly) onto Krashen’s (1982) acquisition vs. learning distinction — but the implicit/explicit framework is more cognitively elaborated and does not commit to Krashen’s non-interface position.

The interface debate:

The central theoretical question is whether explicit and implicit knowledge are:

Non-interface (Krashen): Completely separate; explicit (learned) knowledge cannot convert to implicit (acquired) knowledge; the Monitor can only monitor, not feed the acquisition system.
Weak interface (Ellis 1994): Explicit knowledge can facilitate implicit acquisition under certain conditions — specifically, explicit knowledge can direct attention (noticing), and noticed input can then be processed implicitly.
Strong interface (DeKeyser 1995): Explicit knowledge can convert to implicit knowledge through practice and automatization; skill-learning applies to L2 grammar as to any cognitive skill.

DeKeyser’s Skill Acquisition Theory:

DeKeyser (2003, 2015) argues, drawing on Anderson’s ACT* model, that:

Knowledge is initially declarative (explicit rule).
Through proceduralization, the rule is applied increasingly quickly and automatically.
Through automatization (massive practice), the rule application becomes implicit knowledge — truly automatized, no longer requiring controlled processing.

This strong interface position has been empirically supported in laboratory studies where explicit grammar instruction plus practice produces automatized performance. Critics (Krashen, N. Ellis) argue that this is automatized explicit knowledge rather than truly implicit knowledge — the neural and processing substrate may remain different.

Reber’s artificial grammar learning:

Reber (1967, 1989) demonstrated that subjects could learn the rules of an artificial grammar (letter strings generated by a finite-state automaton) through exposure alone, without being told there were rules — and could classify new strings by whether they followed the grammar without being able to state the rules. This was the first demonstration of implicit learning in adults, and it established that statistical regularities can be extracted unconsciously from the input. SLA researchers extended this to natural language patterns.

N. Ellis’s statistical learning framework:

N. C. Ellis (2002, 2005) argues that L2 acquisition is primarily a statistical learning process — learners track frequency and co-occurrence patterns in the input, gradually building implicit probabilistic representations of the target language. Under this framework:

High-frequency, regular patterns are most acquirable implicitly.
Low-frequency, irregular patterns may require explicit attention to become noticed and encoded.
Explicit instruction sensitizes the implicit system to patterns it might otherwise miss.

Evidence from neuroscience:

Declarative/Procedural model (Ullman 2001): Explicit/declarative lexical knowledge relies on the hippocampus and temporal cortex (declarative memory system); implicit/procedural grammar relies on the basal ganglia and frontal cortex (procedural memory system). This neurological dissociation is consistent with an implicit/explicit distinction.
ERP evidence: Event-related potential (ERP) studies show different processing signatures for violations of implicit vs. explicit knowledge — supporting real processing-level distinctions.

Japanese and implicit/explicit learning:

Agglutinative morphology: Japanese verb and adjective morphology is rich and regular — it may be more susceptible to statistical implicit learning than English’s irregular morphology, because the regularity in the input provides reliable statistical patterns.
Particles as implicit targets: Research on Japanese particle acquisition (wa/ga/wo) shows that high-frequency particles (wa, ga) tend to be acquired implicitly through input, while lower-frequency, pragmatically specialized patterns may require explicit instruction.
Formulaic sequences and implicit learning: Many Japanese pragmatic routines (polite forms, keigo structures) are acquired formulaically — implicitly, as holistic chunks — before their internal morphology is explicitly analyzed.
JLPT effects: JLPT-focused study produces explicit knowledge of grammar forms that may remain declarative-only rather than converting to implicit use knowledge — a pedagogical concern.

History

1967: Reber—artificial grammar learning paradigm; establishes implicit learning in adults.
1982: Krashen—acquisition/learning distinction; non-interface position.
1989: Reber—explicit/implicit memory dissociation; theoretical review.
1994: Ellis—weak interface position; explicit attention → noticing → implicit acquisition.
1995: DeKeyser—strong interface; automatization of explicit to implicit.
2001: Ullman—declarative/procedural model of language; neurological dissociation.
2002–2005: N. C. Ellis—statistical learning framework; frequency and chunking.
2003: DeKeyser—Skill Acquisition Theory applied to SLA.
2015: DeKeyser—updated skill acquisition and interface review.

Common Misconceptions

“Adults can only use explicit learning.” Adults demonstrate substantial implicit learning in laboratory studies (Reber 1967; Williams 2005). The claim is that adult implicit learning capacity may be reduced relative to children’s (or differentially weighted), not that it is absent.

“Explicit grammar study is useless because it doesn’t become implicit.” Even if explicit knowledge doesn’t fully convert to implicit knowledge, it may:

Guide noticing of input patterns.
Reduce cognitive load in monitored conditions.
Scaffold production in planned discourse.

“Implicit = natural; explicit = unnatural.” Both are real cognitive processes. L1 acquisition in children relies heavily on implicit learning, but adult implicit learning is a real capacity. The difference is developmental, not categorical.

Criticisms

The implicit/explicit distinction, while intuitive, is methodologically difficult to operationalize cleanly — measures of “implicit” knowledge (timed grammaticality judgments) may still involve some explicit processing.
The strong interface claim (explicit → implicit via practice) is debated — critics argue laboratory evidence involves tasks with limited ecological validity.
Krashen’s non-interface position has been influential but is not well-supported empirically; most current theorists accept some form of interface.

Social Media Sentiment

The implicit/explicit question maps onto popular debates: “Is grammar study worth it?” vs. “Just use input.” Comprehensible input advocates generally hold a non-interface or weak-interface position: explicit grammar study has minimal value because it doesn’t become acquisition. Academic grammar learners report that explicit rule knowledge helps them understand why they’re making errors and notice corrections in input. Current empirical consensus supports a weak interface — explicit instruction plus input plus practice produces better outcomes than input alone.

Last updated: 2026-04

Practical Application

Use explicit grammar to guide noticing: Study a Japanese grammar structure explicitly (e.g., てもらう vs. てもらえる); then, when you encounter it in input, you’re more likely to notice the contrast and build implicit representations.
Practice automatization: After explicit learning, practice in meaning-focused activities (conversation, timed output, dictation) without consulting rules — this supports proceduralization.
Distinguish rule knowledge from use knowledge: Test whether you have declarative knowledge (can you state the rule for Japanese potential forms?) vs. procedural knowledge (do you automatically produce the correct form in speech?). Direct practice at the productive level.
Spaced retrieval for explicit knowledge: For explicit vocabulary and grammar knowledge, spaced repetition (Anki) strengthens declarative knowledge and may support the transition to implicit knowledge through repeated activation.

Related Terms

Research

Reber, A. S. (1967). Implicit learning of artificial grammars. Journal of Verbal Learning and Verbal Behavior, 6(6), 855–863. [Summary: Foundational implicit learning study; demonstrates adults can extract grammar rules from letter sequences without conscious awareness; artificial grammar paradigm; establishes implicit learning as empirically demonstrable cognitive process.]

DeKeyser, R. (2003). Implicit and explicit learning. In C. J. Doughty & M. H. Long (Eds.), The Handbook of Second Language Acquisition (pp. 313–348). Blackwell. [Summary: Comprehensive review of implicit/explicit learning in SLA; defines key terms; defends strong interface and skill acquisition positions; summarizes laboratory and classroom evidence; essential chapter.]

Ellis, N. C. (2002). Frequency effects in language processing: A review with implications for theories of implicit and explicit language acquisition. Studies in Second Language Acquisition, 24(2), 143–188. [Summary: Statistical learning framework; frequency and co-occurrence in implicit acquisition; argues L2 acquisition is primarily implicit pattern detection; reviews psycholinguistic and corpus evidence; influential theoretical statement.]

Ullman, M. T. (2001). The declarative/procedural model of lexicon and grammar. Journal of Psycholinguistic Research, 30(1), 37–69. [Summary: Neurological model; lexicon relies on declarative memory (hippocampus), grammar on procedural memory (basal ganglia, frontal cortex); dissociation provides biological grounding for implicit/explicit distinction; influential neurolinguistic framework.]

Krashen, S. D. (1982). Principles and Practice in Second Language Acquisition. Pergamon Press. [Summary: Non-interface position; learned (explicit) knowledge functions only as Monitor, cannot convert to acquired (implicit) knowledge; acquisition requires comprehensible input; foundational but empirically challenged statement of non-interface.]

Mikey Does