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

Are sparse-coding simple cell receptive field models physiologically plausible?

Paul A Watters1

  • 1Department of Computing, Macquarie University, NSW 2109, Australia. pwatters@ics.mq.edu.au

Journal of Integrative Neuroscience
|November 28, 2006
PubMed
Summary

The Olshausen and Field model for visual processing in V1 did not accurately predict spatial-frequency tuning in primate simple cells. Model results more closely matched cat neurons, not primate area 17 as anticipated.

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

Editorial: Neural Dynamics - Models and Complexity.

Frontiers in neuroscience·2022
Same author

Effects of Automated Messages on Internet Users Attempting to Access "Barely Legal" Pornography.

Sexual abuse : a journal of research and treatment·2021
Same author

A Survey of Context-Aware Access Control Mechanisms for Cloud and Fog Networks: Taxonomy and Open Research Issues.

Sensors (Basel, Switzerland)·2020
Same author

Accessible virtual reality therapy using portable media devices.

Studies in health technology and informatics·2010
Same author

Selecting parameters for phase space reconstruction of the electrocorticogram (ECoG).

Journal of integrative neuroscience·2005
Same author

A method for estimating long-range power law correlations from the electroencephalogram.

Biological psychology·2004

Area of Science:

  • Computational neuroscience
  • Visual cortex modeling
  • Natural scene processing

Background:

  • The Olshausen and Field model (1996) proposed a simple cell receptive field model for V1 using unsupervised learning.
  • This model optimized basis functions for an overcomplete representation of visual space, qualitatively matching V1 simple cell responses.
  • Previous simulations used whitened natural scenes, mimicking pre-cortical filtering.

Purpose of the Study:

  • To quantitatively assess the spectral domain tuning of the Olshausen and Field model's basis functions.
  • To evaluate the model's goodness-of-fit to known spatial-frequency bandwidths of simple cells in primate V1.
  • To investigate the impact of various model parameters on the spectral tuning characteristics.

Main Methods:

Related Experiment Videos

  • Quantitative estimation of basis function tuning in the spectral domain using a Gaussian model.
  • Simulation experiments varying basis function size, representation completeness (complete vs. overcomplete), sparseness factor, and learning rate.
  • Comparison of model-derived spatial-frequency tuning with empirical data from primate V1 and cat visual areas.

    • Main Results:

    • The model's spatial-frequency tuning consistently failed to match primate area 17 simple cells across all tested conditions.
    • Model outputs showed a closer resemblance to unclassified neurons in cat area 19.
    • Exceptions were noted, but the overall prediction for primate area 17 was not supported.

    Conclusions:

    • The Olshausen and Field model, despite its initial qualitative success, does not fully capture the quantitative spatial-frequency tuning of primate V1 simple cells.
    • The model's parameters and architecture may require significant revision to align with primate visual processing.
    • Further research is needed to reconcile computational models with the specific neurophysiological properties of primate visual cortex.