Analysis of Posture Influence on Memory Using fNIRS-BCI

This study explores the impact of three varied body postures-sitting, standing, and upper-body movement on human memory using functional near-infrared spectroscopy (fNIRS) based brain- computer interface (BCI). The study involves two distinct cognitive activities i.e., the mental arithmetic and picture recall activity. The fNIRS brain signals from 10 healthy subjects (5 male and 5 female) were recorded, and pre-processed. After extracting features, supervised learning was used for classification to analyze how altering body postures affects the memory The Gaussian Naïve Bayes (NB), Decision Tree (DT), and Random Forest (RF) classifiers were applied. For the Mental Arithmetic Activity, sitting posture emerged as the optimal posture, demonstrating the maximum achieved accuracy of 84.2%. Conversely, for the Picture Recall Activity, the upper-body movement emerged as the optimal posture, achieving the maximum accuracy of 87.4%. Through comprehensive analysis, this research contributes insights into the intricate relationship between body postures and cognitive functions.