A brain-computer interface based on a spiking neural network architecture - NeuCube and neuro-feedback

• Main Author: Bhattacharya, Wriju (Author)
• Other Authors: Kasabov, Nikola (Contributor), Wang, Grace (Contributor)
• Format: Others
• Published: Auckland University of Technology, 2015-11-12T21:26:19Z.
• Subjects: SNN; Neurofeedback; Neucube; Thesis
• Online Access: Get fulltext

 In an average human brain, there are about 100 billion neurons connected by synapses that transmits electro-chemical signals to the different parts of the body. Brain Computer Interfacing (BCI) can be used to study and record these signals and has gained a lot of interest over the last few decades. Many different methods are used for BCI, EEG being the most common of these methods. Although we now understand more than ever before about how a brain functions through collecting the spatial and temporal brain data and studying it, little work has been done towards actually using this data to enhance brain signals or treat brain related disorders such as Attention Deficit Hyperactivity Disorder (ADHD) in small children. This thesis identifies this need and proposes neuro-feedback as a process to provide positive feedback to a subject based on their brain activities in real time. Although some studies have been conducted in the past in the field of neuro-feedback, it has failed to gain credibility in the larger scientific community. This thesis outlines a gaming environment called NUN developed in which a subject is connected to an EEG machine and their brain data is collected while providing a visual stimulus via a computer game. The gaming environment is used to train a subject with a certain predefined change in patterns while expecting them to remember the same. The subject is then given feedback in terms of a score. The proposed idea is that the subject can keep playing the game until they have attained a perfect score and while doing so enhance their memory. The developed game was only tested on the developer, so future work is required to perform experiments on more than one subject to prove the concept. This paper also suggests that BCI and neuro-feedback experiments can be undertaken via the use of a very inexpensive, light and easy-to-use EEG machines and that BCI technology is not limited to larger more complex laboratory setups.