Shared representations of human actions across vision and language DOI Creative Commons
Diana C. Dima,

Sugitha Janarthanan,

Jody C. Culham

и другие.

Neuropsychologia, Год журнала: 2024, Номер 202, С. 108962 - 108962

Опубликована: Июль 22, 2024

Humans can recognize and communicate about many actions performed by others. How are organized in the mind, is this organization shared across vision language? We collected similarity judgments of human depicted through naturalistic videos sentences, tested four models action categorization, defining at different levels abstraction ranging from specific (action verb) to broad target: whether an directed towards object, another person, or self). The reflected a representations determined mainly target actions, even after accounting for other semantic features. Furthermore, language model embeddings predicted behavioral captured information alongside unique information. Together, our results show that concepts similarly mind language, reflects socially relevant goals.

Язык: Английский

A shared model-based linguistic space for transmitting our thoughts from brain to brain in natural conversations DOI Creative Commons
Zaid Zada, Ariel Goldstein, Sebastian Michelmann

и другие.

Neuron, Год журнала: 2024, Номер 112(18), С. 3211 - 3222.e5

Опубликована: Авг. 2, 2024

Язык: Английский

Процитировано

16
A High-Efficiency Modelling Method for Analog Integrated Circuits DOI Creative Commons
Dongdong Chen, Yunqi Yang, Xianglong Wang

и другие.

Chip, Год журнала: 2025, Номер unknown, С. 100135 - 100135

Опубликована: Март 1, 2025

Язык: Английский

Процитировано

1
Contextual feature extraction hierarchies converge in large language models and the brain DOI
Gavin Mischler,

Yinghao Aaron Li,

Stephan Bickel

и другие.

Nature Machine Intelligence, Год журнала: 2024, Номер unknown

Опубликована: Ноя. 26, 2024

Язык: Английский

Процитировано

6
Animal models of the human brain: Successes, limitations, and alternatives DOI
Nancy Kanwisher

Current Opinion in Neurobiology, Год журнала: 2025, Номер 90, С. 102969 - 102969

Опубликована: Фев. 1, 2025

Язык: Английский

Процитировано

0
Linguistic coupling between neural systems for speech production and comprehension during real-time dyadic conversations DOI Creative Commons
Zaid Zada, Samuel A. Nastase, Sebastian Speer

и другие.

bioRxiv (Cold Spring Harbor Laboratory), Год журнала: 2025, Номер unknown

Опубликована: Фев. 16, 2025

The core use of human language is communicating complex ideas from one mind to another in everyday conversations. In conversations, comprehension and production processes are intertwined, as speakers soon become listeners, listeners speakers. Nonetheless, the neural systems underlying these faculties typically studied isolation using paradigms that cannot fully engage our capacity for interactive communication. Here, we used an fMRI hyperscanning paradigm measure activity simultaneously pairs subjects engaged real-time, We contextual word embeddings a large model quantify linguistic coupling between within across individual brains. found highly overlapping network regions involved both spanning much cortical network. Our findings reveal shared representations extend beyond into areas associated with social cognition. Together, results suggest specialized speech perception align on common set features encoded broad

Язык: Английский

Процитировано

0
A vectorial code for semantics in human hippocampus DOI Open Access
Melissa Franch,

Elizabeth A. Mickiewicz,

James L. Belanger

и другие.

bioRxiv (Cold Spring Harbor Laboratory), Год журнала: 2025, Номер unknown

Опубликована: Фев. 23, 2025

ABSTRACT As we listen to speech, our brains actively compute the meaning of individual words. Inspired by success large language models (LLMs), hypothesized that brain employs vectorial coding principles, such is reflected in distributed activity single neurons. We recorded responses hundreds neurons human hippocampus, which has a well-established role semantic coding, while participants listened narrative speech. find encoding contextual word simultaneous whose selectivities span multiple unrelated categories. Like embedding vectors models, distance between neural population correlates with distance; however, this effect was only observed (like BERT) and reversed non-contextual Word2Vec), suggesting depends critically on contextualization. Moreover, for subset highly semantically similar words, even embedders showed an inverse correlation distances; attribute pattern noise-mitigating benefits contrastive coding. Finally, further support critical context, range covaries lexical polysemy. Ultimately, these results hypothesis hippocampus follows principles.

Язык: Английский

Процитировано

0
Natural language processing models reveal neural dynamics of human conversation DOI Creative Commons
Jing Cai, Alex E. Hadjinicolaou, Angelique C. Paulk

и другие.

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

Опубликована: Апрель 9, 2025

Through conversation, humans engage in a complex process of alternating speech production and comprehension to communicate. The neural mechanisms that underlie these complementary processes through which information is precisely conveyed by language, however, remain poorly understood. Here, we used pre-trained deep learning natural language processing models combination with intracranial neuronal recordings discover signals reliably reflected production, comprehension, their transitions during conversation between individuals. Our findings indicate the activities were broadly distributed throughout frontotemporal areas across multiple frequency bands. We also find specific words sentences being they dependent on word's context order. Finally, demonstrate patterns partially overlapped listener-speaker associated specific, time-aligned changes activity. Collectively, our reveal dynamical organization subserve harness use understanding underlying human language.

Язык: Английский

Процитировано

0
Approximating the semantic space: word embedding techniques in psychiatric speech analysis DOI Creative Commons

Claudio Palominos,

Rui He,

Karla Fröhlich

и другие.

Schizophrenia, Год журнала: 2024, Номер 10(1)

Опубликована: Дек. 2, 2024

Abstract Large language models provide high-dimensional representations (embeddings) of word meaning, which allow quantifying changes in the geometry semantic space mental disorders. A pattern a more condensed (‘shrinking’) marked by an increase mean similarity between words has been recently documented psychosis across several languages. We aimed to explore this further picture descriptions provided transdiagnostic German sample patients with schizophrenia spectrum disorders (SSD) ( n = 42), major depression (MDD, 43), and healthy controls 44). Compared controls, both clinical groups showed restricted dynamic navigational patterns as captured time series distances crossed, while also showing differential total trajectories navigated. These findings demonstrate alterations centred on dynamics flow meaning SSD MDD, preserving previous indications towards shrinking cases.

Язык: Английский

Процитировано

2
Shared representations of human actions across vision and language DOI Creative Commons
Diana C. Dima,

Sugitha Janarthanan,

Jody C. Culham

и другие.

Neuropsychologia, Год журнала: 2024, Номер 202, С. 108962 - 108962

Опубликована: Июль 22, 2024

Humans can recognize and communicate about many actions performed by others. How are organized in the mind, is this organization shared across vision language? We collected similarity judgments of human depicted through naturalistic videos sentences, tested four models action categorization, defining at different levels abstraction ranging from specific (action verb) to broad target: whether an directed towards object, another person, or self). The reflected a representations determined mainly target actions, even after accounting for other semantic features. Furthermore, language model embeddings predicted behavioral captured information alongside unique information. Together, our results show that concepts similarly mind language, reflects socially relevant goals.

Язык: Английский

Процитировано

0