A Computational Evaluation of Two Models of Retrieval Processes in Sentence Processing in Aphasia

Lissón, Paula; Pregla, Dorothea; Nicenboim, Bruno; Paape, Dario; Nederend, Mick L. van het; Burchert, Frank; Stadie, Nicole; Caplan, David; Vasishth, Shravan

doi:10.1111/cogs.12956

Cited by 19 publications

(6 citation statements)

References 65 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…One fruitful approach that we used in the present work is to establish a linking hypothesis between online measures such as reading times and offline measures such as acceptability judgments or answers to comprehension questions, and to analyze both within a single latent-process model (e.g., Ferreira & Yang, 2019;Nicenboim & Vasishth, 2018;. Another promising avenue is to model reading times and reaction times with mixture distributions (e.g., Lissón et al, 2021;Vasishth, Chopin, Ryder, & Nicenboim, 2017;Vasishth, Jäger, & Nicenboim, 2017), that is, to assume that the data points within conditions may be generated by different processes with different mean latencies and potentially different amounts of variability. By using these approaches, new insights can be gathered from existing data sets.…”

Section: Discussionmentioning

confidence: 99%

Estimating the true cost of garden-pathing: A computational model of latent cognitive processes

Paape¹,

Vasishth²

2022

Preprint

Self Cite

View full text Add to dashboard Cite

What is the processing cost of being garden-pathed by a temporary syntactic ambiguity? We argue that comparing average reading times in garden-path versus non-garden-path sentences is not enough to answer this question. Trial-level contaminants such as inattention, the fact that garden-pathing may occur non-deterministically in the ambiguous condition, and “triage” (rejecting the sentence without reanalysis; Fodor & Inoue, 2000) lead to systematic underestimates of the true cost of garden-pathing. Furthermore, the pure garden-path effect due to encountering an unexpected word needs to be separated from the additional cost of syntactic reanalysis. To get more realistic estimates for the individual processing costs of garden-path surprisal and syntactic reanalysis, we implement a novel computational model that includes trial-level contaminants as probabilistically occurring latent cognitive processes. The model shows good predictive fit to existing reading-time and judgment data. Furthermore, the latent-process approach captures differences between NP/Z garden-path sentences and RRC garden-path sentences: The NP/Z garden path occurs nearly deterministically but can be mostly eliminated by adding a comma. By contrast, the RRC garden path occurs with a lower probability, but disambiguation via semantic plausibility is not always effective.

show abstract

Section: Discussionmentioning

confidence: 99%

Estimating the true cost of garden-pathing: A computational model of latent cognitive processes

Paape¹,

Vasishth²

2022

Preprint

Self Cite

View full text Add to dashboard Cite

show abstract

“…For previous applications of race models to sentence processing, see e.g Nicenboim and Vasishth (2018),Paape and Zimmermann (2020),Lissón et al (2021),Logačev and Vasishth (2016)…”

mentioning

confidence: 99%

How do linguistic illusions arise? Rational inference and good-enough processing as competing latent processes within individuals

Paape¹

2023

Preprint

View full text Add to dashboard Cite

Non-literal interpretations of implausible sentences such as "The mother gave the candle the daughter" have been taken as evidence for a rational error-correction mechanism that reconstructs the intended utterance from the ill-formed input ("... gave the daughter the candle"). However, the good-enough processing framework offers an alternative explanation: readers sometimes miss problematic aspects of sentences because they are only processing them superficially, which leads to acceptability illusions. As a synthesis of these accounts, I propose that rational inferences about errors on the one hand and good-enough processing on the other are competing latent processes that simultaneously occur within the same comprehender. In support of this view, I present data from a two-dimensional grammaticality/ interpretability judgment task with different types of subtly ill-formed sentences. Both rational inference and good-enough processing predict positive interpretability judgments for such sentences, but only good-enough processing also predicts positive grammaticality judgments. By fitting a lognormal race model jointly to judgments and response latencies, I show that rational inference and good-enough processing actively trade off with each other during reading. Furthermore, individual differences measures reveal that participant traits such as linguistic pedantry, interpretational charity, and analytic/intuitive cognitive styles contribute to variability in the processing patterns.

show abstract

“…Informally, high retrieval activation threshold represents higher rigor: a structure is built only when the model is confident that it should be built. However, the best‐fitting value of noise seems implausible as the estimated value (0.5) was higher than the estimate obtained for participants with aphasia (0.45, Lissón et al., 2021; Mätzig, Vasishth, Engelmann, Caplan, & Burchert, 2018).…”

Section: Computational Simulation With the Lewis And Vasishth Modelmentioning

confidence: 79%

Semantic Attraction in Sentence Comprehension

Laurinavichyute

Malsburg

2022

Cognitive Science

View full text Add to dashboard Cite

Agreement attraction is a cross-linguistic phenomenon where a verb occasionally agrees not with its subject, as required by grammar, but instead with an unrelated noun ('The key to the cabinets were . . . '). Despite the clear violation of grammatical rules, comprehenders often rate these sentences as acceptable. Contenders for explaining agreement attraction fall into two broad classes: Morphosyntactic accounts specifically designed to explain agreement attraction, and more general sentence processing models, such as the Lewis and Vasishth model ( 2005), which explain attraction as a consequence of how linguistic structure is stored and accessed in content-addressable memory. In the present research, we disambiguate between these two classes by testing a surprising prediction made by the Lewis and Vasishth model but not by the morphosyntactic accounts, namely that attraction should not be limited to morphosyntax, but that semantic features of unrelated nouns equally induce attraction. A recent study by Cunnings and Sturt (2018) provided initial evidence that this may be the case. Here, we report three single-trial experiments in English (N 1 = 1, 072, N 2 = 1, 426, and N 3 = 2, 454) that compared semantic and agreement attraction and tested whether and how the two interact. All three experiments showed strong semantically induced attraction effects closely mirroring agreement attraction effects. We complement these results with computational simulations confirming that the Lewis and Vasishth model can faithfully reproduce the observed results. In sum, our findings suggest that attraction is a more general phenomenon than is commonly believed, and therefore favor more general sentence processing models such as the Lewis and Vasishth model.

show abstract

A Computational Evaluation of Two Models of Retrieval Processes in Sentence Processing in Aphasia

Cited by 19 publications

References 65 publications

Estimating the true cost of garden-pathing: A computational model of latent cognitive processes

Estimating the true cost of garden-pathing: A computational model of latent cognitive processes

How do linguistic illusions arise? Rational inference and good-enough processing as competing latent processes within individuals

Semantic Attraction in Sentence Comprehension

Contact Info

Product

Resources

About