Artificial Grammar Learning Task¶

HED Task ID: hedtsk_artificial_grammar_learning

Also known as: AGL, Artificial Grammar

Exposure to letter strings generated by a finite-state grammar followed by a grammaticality judgment on novel strings; indexes implicit rule learning.

Description¶

Participants are first exposed to letter strings (e.g., “MXRVXT,” “VXTTRM”) generated by a finite-state grammar, often under the cover story that they are memorizing the strings. In a subsequent test phase, they classify novel letter strings as grammatical or non-grammatical, performing above chance despite being unable to articulate the underlying rules. Reber (1967) introduced this paradigm as the first laboratory demonstration of implicit learning — the acquisition of complex, rule-governed knowledge without conscious awareness. The paradigm has been central to debates about the nature of implicit vs. explicit knowledge, the role of consciousness in learning, and the neural systems supporting rule abstraction.

Inclusion test¶

Procedure	Participants study letter strings generated by a finite-state grammar during a training phase, then classify novel strings as grammatical or not.
Manipulation	Grammar complexity; training duration; surface features (same vs. changed letter set at test); chunk strength.
Measurement	Endorsement rate (proportion judged grammatical); d-prime separating grammatical from ungrammatical; RT.

Variations¶

Variation	Description	Justification
Standard AGL (Reber)	Exposure to grammatical strings followed by grammaticality judgment of novel strings.	Canonical implicit grammar learning: exposure to strings, then grammaticality judgment
Transfer Version	Training on strings from one letter set, testing on strings from a different letter set that preserves the abstract grammar; tests rule abstraction vs. surface-feature learning.	Novel surface elements at test; isolates abstract rule knowledge from surface memorization
Chunk Strength Control	Equating the frequency of letter bigrams and trigrams between grammatical and non-grammatical test items to rule out fragment-based classification.	Stimuli equated for associative chunk strength; controls for alternative explanation
Production Task	Participants generate strings they believe are grammatical rather than classifying; tests the nature of acquired knowledge.	Participant generates grammatical strings rather than judging them; different output requirement
Sequential AGL	Presenting strings one letter at a time to study online prediction and temporal learning.	Motor/spatial sequential learning of grammar; different modality and response type
Hierarchical AGL	Grammars with center-embedded or hierarchical structure; tests whether implicit learning extends to recursive rules.	Nested recursive grammar; structurally distinct grammar type requiring different parsing
Cross-Modal AGL	Auditory or tactile sequences governed by the same grammar; tests modality-independence of implicit learning.	Stimulus modality change (visual→auditory or vice versa) at test; cross-modal transfer paradigm

Cognitive processes¶

This task engages the following cognitive processes:

Key references¶

{‘authors’: ‘Reber, A. S.’, ‘year’: 1967, ‘title’: ‘Implicit learning of artificial grammars’, ‘venue’: ‘Journal of Verbal Learning and Verbal Behavior’, ‘venue_type’: ‘journal’, ‘journal’: ‘Journal of Verbal Learning and Verbal Behavior’, ‘volume’: ‘6’, ‘issue’: ‘6’, ‘pages’: ‘855-863’, ‘doi’: ‘10.1016/s0022-5371(67)80149-x’, ‘openalex_id’: None, ‘pmid’: None, ‘citation_string’: ‘Reber, A. S. (1967). Implicit learning of artificial grammars. Journal of Verbal Learning and Verbal Behavior, 6(6), 855–863.’, ‘url’: ‘https://doi.org/10.1016/s0022-5371(67)80149-x’, ‘source’: ‘crossref’, ‘confidence’: ‘high’, ‘verified_on’: ‘2026-04-20’}
{‘authors’: ‘Reber, A. S.’, ‘year’: 1989, ‘title’: ‘Implicit learning and tacit knowledge.’, ‘venue’: ‘Journal of Experimental Psychology: General’, ‘venue_type’: ‘journal’, ‘journal’: ‘Journal of Experimental Psychology: General’, ‘volume’: ‘118’, ‘issue’: ‘3’, ‘pages’: ‘219-235’, ‘doi’: ‘10.1037/0096-3445.118.3.219’, ‘openalex_id’: None, ‘pmid’: None, ‘citation_string’: ‘Reber, A. S. (1989). Implicit learning and tacit knowledge. Journal of Experimental Psychology: General, 118(3), 219–235.’, ‘url’: ‘https://doi.org/10.1037/0096-3445.118.3.219’, ‘source’: ‘crossref’, ‘confidence’: ‘high’, ‘verified_on’: ‘2026-04-20’}
{‘authors’: ‘Knowlton, B. J., & Squire, L. R.’, ‘year’: 1996, ‘title’: ‘Artificial grammar learning depends on implicit acquisition of both abstract and exemplar-specific information.’, ‘venue’: ‘Journal of Experimental Psychology: Learning, Memory, and Cognition’, ‘venue_type’: ‘journal’, ‘journal’: ‘Journal of Experimental Psychology: Learning, Memory, and Cognition’, ‘volume’: ‘22’, ‘issue’: ‘1’, ‘pages’: ‘169-181’, ‘doi’: ‘10.1037/0278-7393.22.1.169’, ‘openalex_id’: None, ‘pmid’: None, ‘citation_string’: ‘Knowlton, B. J., & Squire, L. R. (1996). Artificial grammar learning depends on implicit acquisition of both abstract and exemplar-specific information. Journal of Experimental Psychology: Learning, Memory, and Cognition, 22(1), 169–181.’, ‘url’: ‘https://doi.org/10.1037/0278-7393.22.1.169’, ‘source’: ‘crossref’, ‘confidence’: ‘high’, ‘verified_on’: ‘2026-04-20’}

Recent references¶

{‘authors’: ‘Pothos, E. M.’, ‘year’: 2007, ‘title’: ‘Theories of artificial grammar learning.’, ‘venue’: ‘Psychological Bulletin’, ‘venue_type’: ‘journal’, ‘journal’: ‘Psychological Bulletin’, ‘volume’: ‘133’, ‘issue’: ‘2’, ‘pages’: ‘227-244’, ‘doi’: ‘10.1037/0033-2909.133.2.227’, ‘openalex_id’: None, ‘pmid’: None, ‘citation_string’: ‘Pothos, E. M. (2007). Theories of artificial grammar learning. Psychological Bulletin, 133(2), 227–244.’, ‘url’: ‘https://doi.org/10.1037/0033-2909.133.2.227’, ‘source’: ‘crossref’, ‘confidence’: ‘high’, ‘verified_on’: ‘2026-04-20’}
{‘authors’: ‘Rohrmeier, M. A., & Cross, I.’, ‘year’: 2014, ‘title’: ‘Modelling unsupervised online-learning of artificial grammars: Linking implicit and statistical learning’, ‘venue’: ‘Consciousness and Cognition’, ‘venue_type’: ‘journal’, ‘journal’: ‘Consciousness and Cognition’, ‘volume’: ‘27’, ‘issue’: None, ‘pages’: ‘155-167’, ‘doi’: ‘10.1016/j.concog.2014.03.011’, ‘openalex_id’: None, ‘pmid’: None, ‘citation_string’: ‘Rohrmeier, M. A., & Cross, I. (2014). Modelling unsupervised online-learning of artificial grammars: Linking implicit and statistical learning. Consciousness and Cognition, 27, 155–167.’, ‘url’: ‘https://doi.org/10.1016/j.concog.2014.03.011’, ‘source’: ‘crossref’, ‘confidence’: ‘high’, ‘verified_on’: ‘2026-04-20’}
{‘authors’: ‘Batterink, L. J., Reber, P. J., Neville, H. J., & Paller, K. A.’, ‘year’: 2015, ‘title’: ‘Implicit and explicit contributions to statistical learning’, ‘venue’: ‘Journal of Memory and Language’, ‘venue_type’: ‘journal’, ‘journal’: ‘Journal of Memory and Language’, ‘volume’: ‘83’, ‘issue’: None, ‘pages’: ‘62-78’, ‘doi’: ‘10.1016/j.jml.2015.04.004’, ‘openalex_id’: None, ‘pmid’: None, ‘citation_string’: ‘Batterink, L. J., Reber, P. J., Neville, H. J., & Paller, K. A. (2015). Implicit and explicit contributions to statistical learning. Journal of Memory and Language, 83, 62–78.’, ‘url’: ‘https://doi.org/10.1016/j.jml.2015.04.004’, ‘source’: ‘crossref’, ‘confidence’: ‘high’, ‘verified_on’: ‘2026-04-20’}