
Exploring Emotional Modulation of the P300 in EEG Data
By Fabiana Ojeda 歐瑩忻
Published on June 14, 2025
June 14, 2025

Project definition
Background
The P300 is an event-related potential (ERP) component often associated with attention and stimulus evaluation. It is typically elicited by infrequent, task-relevant stimuli in oddball paradigms. This project used the MNE sample dataset to examine P300 responses to standard and deviant tones in a cross-modal oddball task.
Tools
- Python
- MNE-Python
- Jupyter Notebook
- Git & GitHub
- EEG plotting and ERP comparison tools
Data
The project uses the MNE sample dataset, which includes auditory and visual stimuli recorded during a multimodal oddball paradigm. For this analysis, auditory standard (left) and deviant (right) tones were selected for ERP comparison.
Deliverables
- ERP plots comparing standard and deviant tones
- A difference wave plot showing the increased P300 amplitude for deviant tones
- A Jupyter Notebook with complete code and results
- README and documentation hosted on GitHub
Results
Event Summary
Event | Description | Count |
---|---|---|
1 | Auditory left | 72 |
2 | Auditory right (standard) | 73 |
3 | Visual right | 73 |
4 | Visual left | 71 |
5 | Auditory deviant (rare) | 15 |
32 | Button press | 16 |
For this project, I focused on event 2 (auditory standard) and event 5 (auditory deviant) because they align with the auditory oddball paradigm known to elicit the P300.
ERP Comparison
The deviant tone produced a larger positive deflection around 300 ms, consistent with the P300 component. This supports attentional engagement triggered by unexpected stimuli.
Difference Wave
Subtracting standard from deviant responses revealed a clear P300 difference peaking at ~300 ms.
Conclusion and Acknowledgements
This project demonstrated the classic P300 effect using auditory tones in a cross-modal oddball paradigm. It served as a practical introduction to MNE-Python, EEG data analysis, and ERP visualization. Future directions could explore emotional modulation of the P300 in similar paradigms.
Big thanks to the Brainhack School organizers, TAs, and professors!