Can prediction and retrodiction explain whether frequent multi-word phrases are accessed ‘precompiled’ from memory or compositionally constructed on the fly?

Onnis, Luca; Huettig, Falk

doi:10.1016/j.brainres.2021.147674

Cited by 7 publications

(7 citation statements)

References 24 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…We propose that processing multiword units involves combining information about individual words and phrases. This is consistent with the large body of psycholinguistic evidence suggesting that language users predict upcoming words (see Baayen et al, 2013; Jacobs et al, 2016; Onnis & Huettig, 2021; N. J. Smith & Levy, 2013).…”

Section: Discussionsupporting

confidence: 87%

“…We found that the typology of language impacts language users' sensitivity to phrasal frequencies. Our findings align well with a single-system view of language processing (e.g., Baayen et al, 2013;Bybee, 1998;Christiansen & Chater, 2016;Onnis & Huettig, 2021) that combines information about individual words and larger units to explain language acquisition and processing. The findings are hard to align with proposals that multiword units are represented holistically (e.g., Wray, 2002).…”

Section: Discussionsupporting

confidence: 85%

See 1 more Smart Citation

Individual word and phrase frequency effects in collocational processing: Evidence from typologically different languages, English and Turkish.

Öksüz,

Brezina,

Monaghan

et al. 2024

Journal of Experimental Psychology: Learning, Memory, and Cogni

View full text Add to dashboard Cite

Collocations are understood to be integral building blocks of language processing, alongside individual words, but thus far evidence for the psychological reality of collocations has tended to be confined to English. In contrast to English, Turkish is an agglutinating language, utilizing productive morphology to convey complex meanings using a single word. Given this, we expected Turkish speakers to be less sensitive to phrasal frequencies than English speakers. In Study 1, we conducted a corpus analysis of translation-equivalent adjectivenoun collocations (e.g., front door) and found differences between the two languages in frequency counts.In Study 2, we conducted a reaction time experiment to determine the sensitivity of native speakers of English and Turkish to the frequency of adjectives, nouns, and whole collocations. Turkish speakers were less sensitive to whole-phrase frequencies, as predicted, indicating that collocations are processed less holistically in Turkish than English. Both groups demonstrated that processing collocations involves combining information about individual words and phrases. Taken together, we show that speakers are sensitive to frequency information at multiple grain sizes that are attuned to the typology of different languages.

show abstract

Section: Discussionsupporting

confidence: 87%

Section: Discussionsupporting

confidence: 85%

Individual word and phrase frequency effects in collocational processing: Evidence from typologically different languages, English and Turkish.

Öksüz,

Brezina,

Monaghan

et al. 2024

Journal of Experimental Psychology: Learning, Memory, and Cogni

View full text Add to dashboard Cite

show abstract

“…Thus, the model can be seen as a procedural approach to error-based associative learning, with prediction as its driving mechanism. This is in line with a long tradition of work that identifies a central role for word prediction, both based on preceding information in a sentence, or even based on the words that appear after a target word (which is often referred to as 'retrodiction' or integration, (Federmeier, 2007;Kutas et al, 2011;Onnis & Thiessen, 2013;Huettig, 2015;Onnis & Huettig, 2021;Alhama et al, 2021;Onnis et al, 2022).…”

Section: Discussionsupporting

confidence: 78%

How does linguistic context influence word learning?

2023

View full text Add to dashboard Cite

While there are well-known demonstrations that children can use distributional information to acquire multiple components of language, the underpinnings of these achievements are unclear. In the current paper, we investigate the potential pre-requisites for a distributional learning model that can explain how children learn their first words. We review existing literature and then present the results of a series of computational simulations with Vector Space Models, a type of distributional semantic model used in Computational Linguistics, which we evaluate against vocabulary acquisition data from children. We focus on nouns and verbs, and we find that: (i) a model with flexibility to adjust for the frequency of events provides a better fit to the human data, (ii) the influence of context words is very local, especially for nouns, and (iii) words that share more contexts with other words are harder to learn.

show abstract

“…Likewise, many adjective–noun combinations appear to be more backward‐looking (i.e., being more informative when integrating preceding context) than predictive (i.e., involving preactivation of upcoming words), as in strong tea but not *powerful tea , or powerful resources but not *strong resources . In the above cases, what distinguishes the acceptability (and processability, see Onnis & Huettig, 2021 ) of a multiword sequence may not be so much the forward conditional probability, which is sensibly low for noise given both do some or make some . This happens because numerous words in the language can follow verbs like do and make , and thus the forward probability values can be empirically estimated in a corpus of language to be low.…”

Section: Introductionmentioning

confidence: 99%

Is the Mind Inherently Predicting? Exploring Forward and Backward Looking in Language Processing

Onnis

Lim

Cheung³

et al. 2022

Cognitive Science

Self Cite

View full text Add to dashboard Cite

Prediction is one characteristic of the human mind. But what does it mean to say the mind is a “prediction machine” and inherently forward looking as is frequently claimed? In natural languages, many contexts are not easily predictable in a forward fashion. In English, for example, many frequent verbs do not carry unique meaning on their own but instead, rely on another word or words that follow them to become meaningful. Upon reading take a the processor often cannot easily predict walk as the next word. But the system can “look back” and integrate walk more easily when it follows take a (e.g., as opposed to *make|get|have a walk). In the present paper, we provide further evidence for the importance of both forward and backward‐looking in language processing. In two self‐paced reading tasks and an eye‐tracking reading task, we found evidence that adult English native speakers’ sensitivity to word forward and backward conditional probability significantly predicted reading times over and above psycholinguistic predictors of reading latencies. We conclude that both forward and backward‐looking (prediction and integration) appear to be important characteristics of language processing. Our results thus suggest that it makes just as much sense to call the mind an “integration machine” which is inherently backward ‘looking.’

show abstract

Can prediction and retrodiction explain whether frequent multi-word phrases are accessed ‘precompiled’ from memory or compositionally constructed on the fly?

Cited by 7 publications

References 24 publications

Individual word and phrase frequency effects in collocational processing: Evidence from typologically different languages, English and Turkish.

Individual word and phrase frequency effects in collocational processing: Evidence from typologically different languages, English and Turkish.

How does linguistic context influence word learning?

Is the Mind Inherently Predicting? Exploring Forward and Backward Looking in Language Processing

Contact Info

Product

Resources

About