Words, rules, and mechanisms of language acquisition

Endress, Ansgar D.; Bonatti, Luca L.

doi:10.1002/wcs.1376

Cited by 23 publications

(33 citation statements)

References 80 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…S4). These results are in line with the hypothesis that the output of the word boundary discovery may provide cues to edges of constituents which in turn can serve as scaffolding for the subsequent discovery of internal structure, i.e., which elements are contained and in which positions 38 .…”

Section: Discussionsupporting

confidence: 86%

Learning hierarchical sequence representations across human cortex and hippocampus

Henin

Turk‐Browne

Friedman

et al. 2019

Preprint

View full text Add to dashboard Cite

Sensory input arrives in continuous sequences that humans experience as units, e.g., words and events. The brain’s ability to discover extrinsic regularities is called statistical learning. Structure can be represented at multiple levels, including transitional probabilities, ordinal position, and identity of units. To investigate sequence encoding in cortex and hippocampus, we recorded from intracranial electrodes in human subjects as they were exposed to auditory and visual sequences containing temporal regularities. We find neural tracking of regularities within minutes, with characteristic profiles across brain areas. Early processing tracked lower-level features (e.g., syllables) and learned units (e.g., words); while later processing tracked only learned units. Learning rapidly shaped neural representations, with a gradient of complexity from early brain areas encoding transitional probability, to associative regions and hippocampus encoding ordinal position and identity of units. These findings indicate the existence of multiple, parallel computational systems for sequence learning across hierarchically organized cortico-hippocampal circuits.

show abstract

Section: Discussionsupporting

confidence: 86%

Learning hierarchical sequence representations across human cortex and hippocampus

Henin

Turk‐Browne

Friedman

et al. 2019

Preprint

View full text Add to dashboard Cite

show abstract

“…Below, we will call this mechanism the "positional" mechanism (though Henson (1998) called it the ordinal mechanism). Empirical and computational results suggest that humans have both mechanisms (and not only one of them as often claimed in the memory literature), but that they dissociate in multiple ways (e.g., Endress & Bonatti, 2007;Endress & Mehler, 2009a;Endress & Wood, 2011;Endress & Bonatti, 2016;Marchetto & Bonatti, 2013).…”

Section: Possible Mechanisms For Extracting Recurrent Units From Contmentioning

confidence: 94%

Transitional probabilities count more than frequency, but might not be used for memorization

Endress

Langus

2017

Cognitive Psychology

Self Cite

View full text Add to dashboard Cite

Learners often need to extract recurring items from continuous sequences, in both vision and audition. The best-known example is probably found in word-learning, where listeners have to determine where words start and end in fluent speech. This could be achieved through universal and experience-independent statistical mechanisms, for example by relying on Transitional Probabilities (TPs). Further, these mechanisms might allow learners to store items in memory. However, previous investigations have yielded conflicting evidence as to whether a sensitivity to TPs is diagnostic of the memorization of recurring items. Here, we address this issue in the visual modality. Participants were familiarized with a continuous sequence of visual items (i.e., arbitrary or everyday symbols), and then had to choose between (i) high-TP items that appeared in the sequence, (ii) high-TP items that did not appear in the sequence, and (iii) low-TP items that appeared in the sequence. Items matched in TPs but differing in (chunk) frequency were much harder to discriminate than items differing in TPs (with no significant sensitivity to chunk frequency), and learners preferred unattested high-TP items over attested low-TP items. Contrary to previous claims, these results cannot be explained on the basis of the similarity of the test items. Learners thus weigh within-item TPs higher than the frequency of the chunks, even when the TP differences are relatively subtle. We argue that these results are problematic for distributional clustering mechanisms that analyze continuous sequences, and provide supporting computational results. We suggest that the role of TPs might not be to memorize items per se, but rather to prepare learners to memorize recurring items once they are presented in subsequent learning situations with richer cues.

show abstract

“…To gain syntactic knowledge, they must catch the relationship between words, and create an abstract representation of grammar categories -like subject, object or verb -so as to infer the system of rules that combines these unities of speech. Therefore, it has been hypothesized that the processing of statistical dependencies hidden within the linguistic input is critical to speech segmentation and vocabulary acquisition (Saffran et al, 1996;Perruchet and Pacton, 2006), whereas the ability to detect and represent abstract rules would mediate the extraction of the grammatical structure (Marcus et al, 1999;Peña et al, 2002;Endress and Bonatti, 2016). Accordingly, the association between early SL abilities and language acquisition has been demonstrated by longitudinal studies.…”

Section: Introductionmentioning

confidence: 99%

Infants’ Learning of Rule-Based Visual Sequences Predicts Language Outcome at 2 Years

et al. 2020

View full text Add to dashboard Cite

The ability to learn and generalize abstract rules from sensory input-i.e., Rule Learning (RL)-is seen as pivotal to language development, and specifically to the acquisition of the grammatical structure of language. Although many studies have shown that RL in infancy is operating across different perceptual domains, including vision, no studies have directly investigated the link between infants' visual RL and later language acquisition. Here, we conducted a longitudinal study to investigate whether 7-montholds' ability to detect visual structural regularities predicts linguistic outcome at 2 years of age. At 7 months, infants were tested for their ability to extract and generalize ABB and ABA structures from sequences of visual shapes, and at 24 months their lexical and grammatical skills were assessed using the MacArthur-Bates CDI. Regression analyses showed that infants' visual RL abilities selectively predicted early grammatical abilities, but not lexical abilities. These results may provide the first evidence that RL mechanisms are involved in language acquisition, and suggest that RL abilities may act as an early neurocognitive marker for language impairments.

show abstract

Words, rules, and mechanisms of language acquisition

Cited by 23 publications

References 80 publications

Learning hierarchical sequence representations across human cortex and hippocampus

Learning hierarchical sequence representations across human cortex and hippocampus

Transitional probabilities count more than frequency, but might not be used for memorization

Infants’ Learning of Rule-Based Visual Sequences Predicts Language Outcome at 2 Years

Contact Info

Product

Resources

About