DocumentCode
718212
Title
Long-term paired sensory stimulation training for improved motor imagery BCI performance via pavlovian conditioning theory
Author
Lin Yao ; Tao Xie ; Xiaokang Shu ; Xinjun Sheng ; Dingguo Zhang ; Xiangyang Zhu
Author_Institution
State Key Lab. of Mech. Syst. & Vibration, Shanghai Jiao Tong Univ., Shanghai, China
fYear
2015
fDate
22-24 April 2015
Firstpage
90
Lastpage
93
Abstract
In this work, paired sensory stimulation training via pavlovian conditioning (PSSPC) was proposed to improve motor imagery BCI performance, especially targeted in those poor performing BCI users. Motor imagery task was paired with the sensory stimulus to establish the conditioned responses through the long-term classic conditioning training. Three poor performing subjects were recruited to participate the PSSPC experiment lasting for about one month in eight sessions. R2 contrast image have shown that the discriminative brain pattern was emerged out in the sensorimotor area of the brain after several sessions training. In addition, up to 80% BCI performance was achieve to some subjects, and it has also shown that learning was evolved in the PSSPC training, complementary to the feedback based training (also termed operant conditioning). The PSSPC methodology has the potential in improving those poor performing BCI subjects, and laid the potential to guide the cortical plastic changes for those with motor impairments.
Keywords
bioelectric potentials; brain-computer interfaces; electroencephalography; neurophysiology; brain computer interface; discriminative brain pattern; improved motor imagery BCI performance; long-term paired sensory stimulation training; motor imagery task; motor impairments; pavlovian conditioning theory; Accuracy; Brain-computer interfaces; Electroencephalography; Neural engineering; Plastics; Training; Vibrations;
fLanguage
English
Publisher
ieee
Conference_Titel
Neural Engineering (NER), 2015 7th International IEEE/EMBS Conference on
Conference_Location
Montpellier
Type
conf
DOI
10.1109/NER.2015.7146567
Filename
7146567
Link To Document