Development of a neurofeedback system focusing on attentional state and validation based on source location and effective connectivity
Name: WAGNER DIAS CASAGRANDE
Publication date: 28/10/2024
Advisor:
Name |
Role |
|---|---|
| ANSELMO FRIZERA NETO | Advisor |
Examining board:
| Name |
Role |
|---|---|
| RICARDO CARMINATI DE MELLO | Examinador Interno |
Pages
Summary: This work addresses the development of a neurofeedback system to support the treatment of attention disorders through the recovery of compromised cognitive control, as well as the experimental validation of the system developed to improve the understanding of which brain regions are directly related to the state of attention and how connectivity occurs between them, which can serve as a guide for neurofeedback sessions applied to the treatment of attention disorders. The protocol defined in the neurofeedback sessions is related to theta and beta frequencies, as there is scientific evidence relating them to attention deficits. Electroencephalography data were collected during three neurofeedback sessions, where subjects played a serious game to stimulate attention, followed by data pre-processing, filtering, and classification. Initial validation was performed with 6 participants, followed by 40 additional participants for more robust results. To validate the neurofeedback system, a methodology was used to investigate source location and effective connectivity, comparing cortical activity between two groups (low and high difficulty — groups) differentiated by the difficulty level of the serious game. The inverse solution based on Cortical Current Density was applied to identify brain regions related to attentional state. From there, the effective connectivity between these regions was estimated using the Directed Transfer Function. With methods applied here, it was possible to identify that brain regions related to the state of attention, including both sides of the medial and dorsolateral prefrontal and right temporal region, showed a localized higher beta activity and interconnectivity in subjects performing greater demanding task (high-difficulty group). This pattern was not observed in subjects performing lower demanding tasks (low-difficulty group), which showed more dispersed flows. In addition, Statistical validation was conducted using ANOVA and analysis of the theta/beta ratio distribution between the two groups. Among the results found, a one-way repeated measures ANOVA demonstrated that when comparing the theta/beta ratios between the three study sessions (S1 vs. S2 vs. S3), for the high difficulty group, the regions that presented significant differences were: right dorsolateral prefrontal (F(2.57) = 3.158, F = 3.225) and left temporal (F(2.57) = 3.158, F = 3.731), demonstrating that the game’s difficulty level influences the theta/beta ratio of the mentioned regions. Future work proposes the development and application of a virtual reality environment to further investigate brain responses to attentional stimuli, in addition to promoting greater participant engagement. It is also proposed to apply the developed system to participants with attention deficits, aiming to improve treatment methods.
