Related Experiment Video
Updated: Jun 28, 2026
07:34
A Simple Stimulatory Device for Evoking Point-like Tactile Stimuli: A Searchlight for LFP to Spike Transitions
Published on: March 25, 2014
Decoding a bistable percept with integrated time-frequency representation of single-trial local field potential
Zhisong Wang1, Nikos K Logothetis, Hualou Liang
1School of Health Information Sciences, University of Texas Health Science Center at Houston, 7000 Fannin, Suite 600, Houston, TX 77030, USA.
Journal of Neural Engineering
|October 31, 2008
Summary
Researchers decoded bistable perception using neural activity from the visual cortex. They found specific gamma-band features in local field potentials (LFPs) are key for distinguishing perception states, aiding brain-computer interface development.
Area of Science:
- Neuroscience
- Computational Neuroscience
- Visual Perception
Background:
- Bistable perception involves alternating between two states with constant visual input, offering a way to study perception's neural basis.
- Analyzing local field potentials (LFPs) from the middle temporal (MT) cortex in macaque monkeys can reveal neural correlates of perception.
Purpose of the Study:
- To decode bistable structure-from-motion (SFM) perception using dynamic LFP activity.
- To identify robust features from LFP time-frequency representations for accurate perception decoding.
Main Methods:
- Utilized relaxation (RELAX) and sequential forward selection (SFS) algorithms on integrated LFP spectrograms.
- Employed Support Vector Machines (SVM) and Linear Discriminant Analysis (LDA) for single-trial decoding.
- Focused on features from the gamma frequency band (30-100 Hz) within specific temporal windows.
Main Results:
- Integrated-spectrogram based feature selection outperformed instantaneous-spectrogram methods.
- Decoding of bistable perception achieved excellent performance using selected LFP features.
- Gamma-band LFP features within specific time windows proved most discriminative.
Conclusions:
- The integrated-spectrogram approach effectively decodes bistable perception.
- Gamma-band LFP activity in the MT cortex contains critical information for perception.
- This feature selection method shows promise for applications like brain-computer interfaces (BCI).
