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

Statistical models for neural encoding, decoding, and optimal stimulus design.

Liam Paninski1, Jonathan Pillow, Jeremy Lewi

  • 1Department of Statistics and Center for Theoretical Neuroscience, Columbia University, New York, NY, USA. liam@stat.columbia.edu

Progress in Brain Research
|October 11, 2007
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

Sparse component analysis: A method that uncovers separable computations within neural population activity.

Neuron·2026
Same author

Beast3D: Animal behavioral analysis and neural encoding from multi-view video via Gaussian splatting.

ArXiv·2026
Same author

Lightning Pose 3D: an uncertainty-aware framework for data-efficient multi-view animal pose estimation.

bioRxiv : the preprint server for biology·2026
Same author

Computational optimization of two-photon holographic stimulation sites<i>in vivo</i>.

Journal of neural engineering·2026
Same author

Spectral learning of Bernoulli linear dynamical systems models.

Transactions on machine learning research·2026
Same author

Animal behavioral analysis and neural encoding with transformer-based self-supervised pretraining.

ArXiv·2026
Same journal

Preface.

Progress in brain research·2025
Same journal

Mindfulness and meditation: Promoting emotional and cognitive health.

Progress in brain research·2025
Same journal

Cognitive stimulation enhancing memory and mental function.

Progress in brain research·2025
Same journal

The science behind non-pharmacological interventions.

Progress in brain research·2025
Same journal

Technology-assisted interventions for neuropsychiatric disorders.

Progress in brain research·2025
Same journal

Ethical consideration in non-pharmacological treatments for neuropsychiatric disorders.

Progress in brain research·2025
See all related articles

Statistical models offer a unified approach to analyzing neural data, addressing both encoding and decoding problems. These methods efficiently predict neural activity and decode stimuli from spike trains.

Area of Science:

  • Computational Neuroscience
  • Statistical Modeling
  • Neural Data Analysis

Background:

  • Neural data analysis faces challenges in understanding how neural spike trains encode stimuli and how to decode stimuli from these trains.
  • Existing methods often struggle to capture complex dependencies like stimulus history and interneuronal interactions.

Purpose of the Study:

  • To present statistical model-based techniques providing a unified solution for neural encoding and decoding problems.
  • To introduce methods that account for stimulus dependencies, spike history, and population neural interactions.

Main Methods:

  • Utilizing statistical models, including those related to integrate-and-fire neurons, to analyze neural spike trains.
  • Employing flexible, likelihood-based methods for fitting encoding models and performing optimal decoding.

Related Experiment Videos

  • Leveraging a key concavity property of the model likelihood for computational tractability.

    • Main Results:

    • Demonstrated a unified framework for tackling both neural encoding and decoding problems.
    • Showcased the ability of models to capture complex neural dynamics, including stimulus and history effects.
    • Established computational tractability for complex analysis tasks due to model properties.

    Conclusions:

    • Statistical models provide a powerful and unified approach to neural data analysis.
    • These methods enable efficient prediction of neural responses and accurate stimulus decoding.
    • The framework supports adaptive stimulus optimization for efficient model parameter inference.