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

Nonlinear ideal observation and recurrent preprocessing in perceptual learning.

L Zhaoping1, Michael H Herzog, Peter Dayan

  • 1Department of Psychology, University College, London WC1E 6BT, UK.

Network (Bristol, England)
|June 7, 2003
PubMed
Summary
This summary is machine-generated.

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

Silver bullets and sensory horizons.

The Behavioral and brain sciences·2026
Same author

Examining the relationship between ssVEP and psychophysical measures of contrast sensitivity, grating acuity, and orientation discrimination.

iScience·2026
Same author

Starting a revolution with a refuted model?

The Behavioral and brain sciences·2025
Same author

Object recognition from sparse simulated phosphenes and curved segments.

Vision research·2025
Same author

Dynamics of vision: Grouping takes longer than crowding.

Journal of vision·2025
Same author

Novelty as a drive of human exploration in complex stochastic environments.

Proceedings of the National Academy of Sciences of the United States of America·2025
Same journal

Enhancing IoT security: A Creative Swagger Optimization algorithm for DDoS defence.

Network (Bristol, England)·2026
Same journal

Parametric optimization for electrical discharge diamond grinding (EDDG) system using dual approach.

Network (Bristol, England)·2025
Same journal

A novel lung cancer diagnosis model using hybrid convolution (2D/3D)-based adaptive DenseUnet with attention mechanism.

Network (Bristol, England)·2025
Same journal

Hybrid optimization enabled Eff-FDMNet for Parkinson's disease detection and classification in federated learning.

Network (Bristol, England)·2025
Same journal

AI-driven plant disease detection with tailored convolutional neural network.

Network (Bristol, England)·2025
Same journal

Layer modified residual Unet++ for speech enhancement using Aquila Black widow optimizer algorithm.

Network (Bristol, England)·2025
See all related articles

Visual positional variations alter ideal observer strategies in hyperacuity tasks. A recurrent neural network model demonstrates improved discrimination by adapting to these variations, enhancing spatial resolution performance.

Area of Science:

  • Neuroscience
  • Computational Vision
  • Psychophysics

Background:

  • Visual tasks, especially those requiring high spatial resolution (hyperacuity), are affected by unavoidable variations in stimulus presentation on the retina due to factors like micro-saccades and fixation errors.
  • These positional uncertainties are particularly critical for hyperacuity tasks, which demand precise spatial discrimination.

Purpose of the Study:

  • To investigate how small positional variations impact the ideal observer's strategy in hyperacuity-like visual discrimination tasks.
  • To propose and evaluate a computational model, inspired by recurrent processing in early vision, that can account for improved discrimination performance despite positional uncertainties.

Main Methods:

  • Developed a theoretical model of an ideal observer for a hyperacuity task, analyzing how positional variations change the optimal discrimination strategy.

Related Experiment Videos

  • Introduced a recurrent preprocessor model for noisy neural activities, followed by a linear discriminator, to simulate visual processing under positional uncertainty.
  • Compared the model's performance against theoretical predictions and discussed its implications for perceptual learning.

    • Main Results:

    • Showed that positional variations cause the ideal observer's optimal strategy to shift from linear to quadratic dependence on noisy neural activities.
    • Demonstrated that the recurrent preprocessor model significantly improves discrimination performance, even with positional variations exceeding the task's threshold acuity.
    • The model's structure and performance are consistent with psychophysical observations of hyperacuity improvement during perceptual learning.

    Conclusions:

    • Positional variations necessitate adaptive strategies in visual processing for hyperacuity tasks.
    • Recurrent neural network models offer a viable mechanism for enhancing visual discrimination under retinal positional uncertainty.
    • The findings provide insights into the neural basis of perceptual learning and its role in optimizing visual performance.