Related Experiment Video
Updated: May 18, 2026
11:39
Assessment of Audio-Tactile Sensory Substitution Training in Participants with Profound Deafness Using the Event-Related Potential Technique
Published on: September 7, 2022
A computational model for generation of the P300 evoked potential component
Bharat K Bonala1, Ben H Jansen
1Department of Electrical and Computer Engineering, University of Houston, N308-D2, Houston, TX 77204-4005, USA.
Journal of Integrative Neuroscience
|September 15, 2012
Summary
This study introduces a neural network model simulating the P300 brainwave, linked to working memory context updating. The model successfully replicates key P300 characteristics observed in human experiments.
Area of Science:
- Cognitive Neuroscience
- Computational Neuroscience
- Artificial Intelligence
Background:
- The P300 event-related potential is sensitive to informational content, not physical stimulus features.
- P300 is hypothesized to reflect context updating processes in human working memory.
Purpose of the Study:
- To develop a neural network model that simulates the learning and forgetting mechanisms underlying P300 generation.
- To validate the model by comparing its simulated P300 characteristics with experimental data.
Main Methods:
- A neural network model incorporating a modified Hebbian learning rule was developed.
- The model's weight dynamics were designed to mimic memory processes.
- Simulated P300 features were compared against empirical findings.
Main Results:
- The model demonstrated the ability to mimic the relationship between P300 amplitude and stimulus probability.
- The model also replicated the P300's response to task relevance.
- Simulated P300 characteristics closely matched experimental observations.
Conclusions:
- The proposed neural network model effectively simulates key aspects of P300 generation.
- This model provides insights into the neural mechanisms of working memory and context updating.
- The findings support the role of context updating in P300 generation.
