Search bioRxiv⌕ Search
Home/Archive/bioRxiv

bioRxiv · 10.1101/2024.08.05.606517

Shared Neural Dynamics of Facial Expression Processing

Ely, M. M.·Ambrus, G. G.
View original source ↗Export metadata ↓

Abstract

The ability to recognize and interpret facial expressions is fundamental to human social cognition, enabling navigation of complex interpersonal interactions and understanding of others emotional states. The extent to which neural patterns associated with facial expression processing are shared between observers remains unexplored, and no study has yet examined the neural dynamics specific to different emotional expressions. Additionally, the neural processing dynamics of facial attributes such as sex and identity in relation to facial expressions have not been thoroughly investigated. In this study, we investigated the shared neural dynamics of emotional face processing using an explicit facial emotion recognition task, where participants made two-alternative forced choice (2AFC) decisions on the displayed emotion. Our data-driven approach employed cross-participant multivariate classification and representational dissimilarity analysis on EEG data. The results demonstrate that EEG signals can effectively decode the sex, emotional expression, and identity of face stimuli across different stimuli and participants, indicating shared neural codes for facial expression processing. Multivariate classification analyses revealed that sex is decoded first, followed by identity, and then emotion. Emotional expressions (angry, happy, sad) were decoded earlier when contrasted with neutral expressions. While identity and sex information were modulated by image-level stimulus features, the effects of emotion were independent of visual image properties. Importantly, our findings suggest enhanced processing of face identity and sex for emotional expressions, particularly for angry faces and, to a lesser extent, happy faces.

Source connections

Explore related subjects

scientific communication and educationPreprint

Keep this discovery

Explore connections, maps & timelines

BibTeXRIS

Ely, M. M., Ambrus, G. G.. 2024-08-07. Shared Neural Dynamics of Facial Expression Processing. https://doi.org/10.1101/2024.08.05.606517

Cite the original work for its findings. Save a collection to share your selection of sources.

KEEP EXPLORING

Related preprints

bioRxiv2026

Evaluating Large Language Models as Tools to Navigate Researchers in Rapidly Evolving Research Landscapes: A Case Study in Cancer Drug Response Prediction

Large Language Models (LLMs) have emerged as promising tools for assisting researchers in automating and accelerating the synthesis of literature reviews. However, their reliability is a significant concern due to issues like factual inaccuracies and hallucinations. The key question is whether LLMs can reliably provide comprehensive, up-to-date overviews and analyses. This study evaluates the performance of three leading LLMs (OpenAI's ChatGPT, Google's Gemini, and DeepSeek) on the complex task of generating a comprehensive survey paper on deep learning for cancer Drug Response Prediction (DRP). By testing both standard and Deep Research (DR) / Deep Think (DT) modes of LLMs with prompts of varying detail, this paper assesses key academic dimensions, including reference management, content quality, and analytical depth. Key findings reveal that while DR modes of LLMs significantly improve reliability by eliminating hallucinations, performance variations exist across models and prompts. A trade-off between reference quantity and integration quality was observed, and even the best-performing models lacked the analytical depth of human experts, often requiring extensive human supervision. The study concludes that LLMs currently serve as powerful assistive tools but still cannot replace the critical validation and synthesis provided by human researchers. Choosing the best LLM to use depends on the task in hand, while several strategies can be implemented to improve the produced output.

scientific communication and education↗
bioRxiv2025

A thirty-year trend of increasing clinical orientation at the National Institutes of Health

It is widely recognized that funding for biomedical research supports the development of major medical advances. However, little systematic effort has been made to determine whether a link exists between the types of funding opportunities that are available to scientists and progress towards new treatments in the clinic. To better understand this relationship, we analyzed the funding opportunities offered by the National Institutes of Health (NIH) over a span of thirty-two years, together with the resulting portfolio of applications and awards. We found NIH funding opportunity announcements became more numerous and increasingly clinically oriented over that span, following a trend that parallels the increasing clinical and translational orientation of both NIH grant applications and NIH-funded publications. Surprisingly, this increase appears to be independent of the representation of clinician-scientists in the NIH workforce.

scientific communication and education↗
bioRxiv2025

Science-wide mapping and ranking of institutions based on affiliated authors

Standardized information on the size of institutions and on their concentration of high-impact scientists is missing. Here, we generate institution-level data on 6,979 institutions with >=50 authors who have had >=15 Scopus-indexed items. We present data on the number and proportion of top-cited authors (according to a composite citation indicator) for each institution. The primary analysis considers 1,214,702 authors (145,704 top-cited in career-long impact) who started publishing in 1980 or later, have published >=40 Scopus-indexed items and >=5 single-, first- or last-authored items. There is modest overlap among top institutions based on volume and based on number of top-cited authors. However, there is minimal overlap when institutions with the highest proportions of top-cited authors are considered; these include mostly research institutes and tech organizations and a minority of universities. Institutions with few eligible authors have large uncertainty. We propose a percentile ranking of the 2,380 largest institutions that adjusts the number of top-cited authors for the proportion of those who have high (>95th percentile) self-citation rates and high rates of publications in Scopus-discontinued titles and further penalizes for authorships of retracted papers not due to publisher/journal errors. Institutions with the highest self-citation, discontinuation, and retraction penalties cluster in specific countries. Among countries with >=15 institutions among the 2,380, Saudi Arabia, China, Malaysia, Iran, India and Indonesia have the lowest median percentile rankings of their institutions (3.9th-21st percentile) due to high penalties. The publicly available datasets that we provide allow institution-level research assessments balancing high impact against retractions and gaming practices.

scientific communication and education↗