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

Selective networks capable of representative transformations, limited generalizations, and associative memory

G M Edelman, G N Reeke

    Proceedings of the National Academy of Sciences of the United States of America
    |March 1, 1982
    PubMed
    Summary

    This study introduces a novel automaton with two neuron-like networks for limited 2D pattern recognition. It develops associative memory through experience-based connection adjustments without explicit instructions.

    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

    Complexity and coherency: integrating information in the brain.

    Trends in cognitive sciences·2011
    Same author

    Modulation of expression and cell surface distribution of N-CAM during myogenesis in vitro.

    Neurochemistry international·2010
    Same author

    Differential translation and fragile X syndrome.

    Genes, brain, and behavior·2005
    Same author

    Theoretical neuroanatomy and the connectivity of the cerebral cortex.

    Behavioural brain research·2002
    Same author

    Efficient marking of neural stem cell-derived neurons with a modified murine embryonic stem cell virus, MESV2.

    Gene therapy·2002
    Same author

    Binding of neural cell adhesion molecules (N-CAMs) to the cellular prion protein.

    Journal of molecular biology·2001

    Area of Science:

    • Computational Neuroscience
    • Artificial Intelligence
    • Automata Theory

    Background:

    • Traditional automata often rely on explicit programming or fixed rules.
    • Developing systems with adaptive learning capabilities remains a key challenge in AI.

    Purpose of the Study:

    • To create a novel automaton capable of limited two-dimensional pattern recognition.
    • To investigate the generation of associative memory through interconnected neural networks.

    Main Methods:

    • Utilized two parallel sets of selective networks with neuron-like elements.
    • Implemented preassigned, experience-alterable connection strengths.
    • Integrated local feature detection with global feature correlation via reentrant interactions.

    Related Experiment Videos

    Main Results:

    • The automaton demonstrated limited recognition of two-dimensional patterns.
    • A new function, associative memory, was generated through network interactions.
    • The system learned without forced learning, explicit semantic rules, or a priori instructions.

    Conclusions:

    • The proposed automaton architecture effectively integrates local and global processing for pattern recognition.
    • The model highlights a biologically plausible mechanism for associative memory formation.
    • This approach offers a new paradigm for developing adaptive, rule-free computational systems.