Jove
Visualize
Contact Us
JoVE
x logofacebook logolinkedin logoyoutube logo
ABOUT JoVE
OverviewLeadershipBlogJoVE Help Center
AUTHORS
Publishing ProcessEditorial BoardScope & PoliciesPeer ReviewFAQSubmit
LIBRARIANS
TestimonialsSubscriptionsAccessResourcesLibrary Advisory BoardFAQ
RESEARCH
JoVE JournalMethods CollectionsJoVE Encyclopedia of ExperimentsArchive
EDUCATION
JoVE CoreJoVE BusinessJoVE Science EducationJoVE Lab ManualFaculty Resource CenterFaculty Site
Terms & Conditions of Use
Privacy Policy
Policies

Related Experiment Videos

EMPATH: a neural network that categorizes facial expressions.

Matthew N Dailey1, Garrison W Cottrell, Curtis Padgett

  • 1Computer Science and Engineering, University of California, San Diego 92093, USA.

Journal of Cognitive Neuroscience
|December 24, 2002
PubMed
Summary

A neural network model explains facial expression recognition, showing it can be both categorical and graded. This biologically plausible model predicts psychological data without tuning, unifying competing theories of emotion perception.

Related Concept Videos

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by
Same author

Canine radiography and echocardiography sensor fusion for deep learning-based detection and classification of myxomatous mitral valve disease.

Journal of thoracic disease·2026
Same author

Monkey upload: Improving robustness using multi-stage neural alignment.

Journal of vision·2026
Same author

SMART deep learning tools to accelerate the characterization of natural product structures from their NMR data.

Methods in enzymology·2026
Same author

AI-powered evaluation of dementia severity based on clinical data and visual scoring systems (MTA, ERICA, GCA) from MRI.

Scientific reports·2026
Same author

"Dense Amygdala": Extensive Complex-valued Functional MRI of the Ventral and Medial Temporal Lobe during Passive Movie Watching in Three Individuals.

Scientific data·2026
Same author

SPECTRE: A Multimodal Spectral Transformer for Small Molecule Annotation.

Journal of chemical information and modeling·2026

Area of Science:

  • Cognitive Neuroscience
  • Computational Neuroscience

Background:

  • Facial expression recognition theories are divided between categorical perception (discrete categories) and graded perception (continuous space).
  • Existing psychological data supports both competing theories, creating a paradox in understanding facial expression perception.

Purpose of the Study:

  • To develop and test a biologically plausible neural network model for facial expression recognition.
  • To determine if a single model can account for data supporting both categorical and graded perception theories.

Main Methods:

  • A simple neural network was trained to classify facial expressions into six basic emotions.
  • The model's predictions were compared quantitatively and qualitatively with existing psychological data.

Related Experiment Videos

Main Results:

  • The model successfully predicted psychological data related to categorization, similarity, reaction times, and discrimination.
  • The model's performance supported both categorical and graded perception phenomena without parameter tuning.

Conclusions:

  • Facial expression perception can be explained as a natural consequence of neural network implementation in the brain.
  • A unified computational framework can reconcile competing theories of facial expression recognition.