Composing, not-composing, and integrating: The neuroscience of multi-word expressions DOI
Paolo Canal, Francesco Vespignani, Nicola Molinaro

и другие.

Elsevier eBooks, Год журнала: 2024, Номер unknown, С. 410 - 420

Опубликована: Май 20, 2024

Язык: Английский

Expectation violations signal goals in novel human communication DOI Creative Commons

Tatia Buidze,

Tobias Sommer, Ke Zhao

и другие.

Nature Communications, Год журнала: 2025, Номер 16(1)

Опубликована: Фев. 26, 2025

Abstract Communication, often grounded in shared expectations, faces challenges when a Sender and Receiver lack common linguistic background. Our study explores how people instinctively turn to the fundamental principles of physical world overcome such barriers. Specifically, through an experimental game which Senders convey messages via trajectories, we investigate they develop novel strategies without relying on cues. We build computational model based principle expectancy violations set universal priors derived from movement kinetics. The replicates participant-designed with high accuracy shows its core variable—surprise—predicts Receiver’s physiological neuronal responses brain areas processing expectation violations. This work highlights adaptability human communication, showing surprise can be powerful tool forming new communicative language.

Язык: Английский

Процитировано

2

Language in Brains, Minds, and Machines DOI
Greta Tuckute, Nancy Kanwisher, Evelina Fedorenko

и другие.

Annual Review of Neuroscience, Год журнала: 2024, Номер 47(1), С. 277 - 301

Опубликована: Апрель 26, 2024

It has long been argued that only humans could produce and understand language. But now, for the first time, artificial language models (LMs) achieve this feat. Here we survey new purchase LMs are providing on question of how is implemented in brain. We discuss why, a priori, might be expected to share similarities with human system. then summarize evidence represent linguistic information similarly enough enable relatively accurate brain encoding decoding during processing. Finally, examine which LM properties—their architecture, task performance, or training—are critical capturing neural responses review studies using as silico model organisms testing hypotheses about These ongoing investigations bring us closer understanding representations processes underlie our ability comprehend sentences express thoughts

Язык: Английский

Процитировано

13

Predictive Use of Grammatical Gender During Noun Phrase Decoding: An Eye-Tracking Study With German Children With Developmental Language Disorder DOI

Jürgen Cholewa,

Annika Kirschenkern,

Frederike Steinke

и другие.

Journal of Speech Language and Hearing Research, Год журнала: 2025, Номер 68(3), С. 1056 - 1074

Опубликована: Фев. 6, 2025

Purpose: Predictive language comprehension has become a major topic in psycholinguistic research. The study described this article aims to investigate if German children with developmental disorder (DLD) use grammatical gender agreement predict the continuation of noun phrases same way as it been observed for typically developing (TD) children. also seeks differentiate between specific and general deficits predictive processing by exploring anticipatory semantic information. Additionally, research examines whether information varies speed stimulus presentation. Method: included 30 DLD (average age = 8.7 years) 26 TD 8.4 who participated visual-world eye-tracking study. Noun phrases, consisting an article, adjective, noun, were presented that matched only one two target pictures. contained cue, combination both, or none these cues. cues provided and/or adjective could be used identify picture before itself was presented. Results: Both groups, those DLD, utilized strategies response when decoding phrases. However, able consider at slower speech rate, even then, their certainty remained below typical level age. Conclusion: Based on findings, discusses potential relevance prediction framework explaining well clinical implications results.

Язык: Английский

Процитировано

1

Predicting the next sentence (not word) in large language models: What model-brain alignment tells us about discourse comprehension DOI Creative Commons
Shaoyun Yu,

Chanyuan Gu,

Kexin Huang

и другие.

Science Advances, Год журнала: 2024, Номер 10(21)

Опубликована: Май 23, 2024

Current large language models (LLMs) rely on word prediction as their backbone pretraining task. Although is an important mechanism underlying processing, human comprehension occurs at multiple levels, involving the integration of words and sentences to achieve a full understanding discourse. This study by using next sentence (NSP) task investigate mechanisms discourse-level comprehension. We show that NSP enhanced model’s alignment with brain data especially in right hemisphere demand network, highlighting contributions nonclassical regions high-level understanding. Our results also suggest can enable model better capture performance encode contextual information. demonstrates inclusion diverse learning objectives leads more human-like representations, investigating neurocognitive plausibility tasks LLMs shed light outstanding questions neuroscience.

Язык: Английский

Процитировано

8

The role of general cognitive skills in integrating visual and linguistic information during sentence comprehension: individual differences across the lifespan DOI Creative Commons
Florian Hintz, Cesko C. Voeten, Dorottya Dobó

и другие.

Scientific Reports, Год журнала: 2024, Номер 14(1)

Опубликована: Авг. 1, 2024

Abstract Individuals exhibit massive variability in general cognitive skills that affect language processing. This is partly developmental. Here, we recruited a large sample of participants (N = 487), ranging from 9 to 90 years age, and examined the involvement nonverbal processing speed (assessed using visual auditory reaction time tasks) working memory forward backward Digit Span world task. Participants saw two objects on screen heard sentence referred one them. In half sentences, target object could be predicted based verb-selectional restrictions. We observed evidence for anticipatory predictable compared non-predictable trials. Visual had main effects comprehension facilitated predictive processing, as evidenced by an interaction. only weak comprehension. Age nonlinear effect (younger adults responded faster than children older adults), but it did not differentially modulate non-predictive nor memory. Our results contribute delineating are involved language-vision interactions.

Язык: Английский

Процитировано

5

Predictive Processing Among Individuals with Autism Spectrum Disorder During Online Language Comprehension: A Preliminary Systematic Review DOI

Junli Qi,

Jing Peng, Xin Kang

и другие.

Journal of Autism and Developmental Disorders, Год журнала: 2025, Номер unknown

Опубликована: Янв. 6, 2025

Язык: Английский

Процитировано

0

Informativity enhances memory robustness against interference in sentence comprehension DOI Creative Commons
Weijie Xu, Richard Futrell

Journal of Memory and Language, Год журнала: 2025, Номер 142, С. 104603 - 104603

Опубликована: Янв. 18, 2025

Язык: Английский

Процитировано

0

Emotional valence, cloze probability, and entropy: Completion norms for 403 French sentences DOI
Jérémy Brunel, Émilie Dujardin, Sandrine Delord

и другие.

Behavior Research Methods, Год журнала: 2025, Номер 57(2)

Опубликована: Янв. 31, 2025

Язык: Английский

Процитировано

0

Predictive coding and dimension-selective attention enhance the lateralization of spoken language processing DOI Creative Commons
Basil C. Preisig, Martin Meyer

Neuroscience & Biobehavioral Reviews, Год журнала: 2025, Номер unknown, С. 106111 - 106111

Опубликована: Март 1, 2025

Hemispheric lateralization in speech and language processing exemplifies functional brain specialization. Seminal work patients with left hemisphere damage highlighted the left-hemispheric dominance functions. However, is not confined to hemisphere. Hence, some researchers associate auditory asymmetries: slow temporal fine spectral acoustic information preferentially processed right regions, while faster primarily handled by regions. Other scholars posit that relates more linguistic processing, particularly for speech-like stimuli. We argue these seemingly distinct accounts are interdependent. Linguistic analysis of relies on top-down processes, such as predictive coding dimension-selective attention, which enhance lateralized engaging left-lateralized sensorimotor networks. Our review highlights weaker simple sounds, stronger strongest meaningful speech. Evidence shows selective attention lateralization. illustrate processes rely networks provide insights into role processing.

Язык: Английский

Процитировано

0

Wie Kinder mit Sprachentwicklungsstörungen das grammatische Genus im prädiktiven Sprachverstehen nutzen DOI
Thomas Günther,

Jürgen Cholewa

Sprache · Stimme · Gehör, Год журнала: 2025, Номер unknown

Опубликована: Апрель 14, 2025

Abstract The article addresses predictive language comprehension in children with development disorders (DLD). It shows that these have difficulties utilizing grammatical gender cues. While healthy do this effectively, SES are limited their abilities. emphasizes the need to develop diagnostic procedures and therapeutic interventions promote use of knowledge.

Язык: Английский

Процитировано

0