Ensemble Empirical Mode Decomposition Analysis of EEG Data Collected during a Contour Integration Task
Karema Al-Subari1, Saad Al-Baddai1, Ana Maria Tomé2
1Department of Biology, Institute of Biophysics, University of Regensburg, Regensburg, Germany; Department of Linguistics, Literature and Culture, Institute of Information Science, University of Regensburg, Regensburg, Germany.
Plos One
|April 25, 2015
Summary
This study used ensemble empirical mode decomposition (EEMD) to analyze electroencephalography (EEG) data during a visual contour integration task. Findings reveal distinct event-related modes (ERMs) differentiating contour from non-contour stimuli, supporting network activity models.
Area of Science:
- Neuroscience
- Cognitive Science
- Signal Processing
Background:
- Visual processing involves integrating simple elements into coherent shapes.
- Electroencephalography (EEG) and functional Magnetic Resonance Imaging (fMRI) are key tools for studying brain activity.
- Contour integration is a fundamental aspect of visual perception.
Purpose of the Study:
- To analyze EEG data from a combined EEG/fMRI study on visual contour integration.
- To identify characteristic features of event-related modes (ERMs) using ensemble empirical mode decomposition (EEMD).
- To investigate differences in brain responses to contour versus non-contour visual stimuli.
Main Methods:
- Data-driven analysis of EEG data.
- Ensemble Empirical Mode Decomposition (EEMD) for signal analysis.
- Combined EEG/fMRI recording during a contour integration task.
Main Results:
- Identified significant differences in ERMs for contour vs. non-contour stimuli.
- Observed early (P100, N200) and late responses in occipital and frontal areas, respectively.
- Found bimodal early/late response signatures in central brain areas.
- Localized statistically significant differences using head topographies.
Conclusions:
- Contour integration elicits distinct neural responses detectable via EEG.
- Specific ERMs (P100, N200) show differential patterns related to stimulus type and brain region.
- Findings support models of contour integration relying on distributed brain network activity.
