Research On Learning Evaluation System Based On Hybrid Brain-Computer Interface

For a long time,the standard for measuring a student's learning effect is often only the test scores in exam-oriented education environment,but attention is also an important factor which affects the learning effect of the students.Many students are distressed by the lack of concentration.Currently,there is a lack of tools in the field of education that can feed back students' attention.Applying attention detection technology to education can detect the learner's attention in real time,which provides a second reliable dimension for measuring the learning effect besides test scores.Therefore,this thesis designs a learning evaluation system based on hybrid brain-computer interface.The system incorporates attention into the evaluation of learning effects,which mainly includes two parts: EEG signal acquisition equipment and English word learning system.The main work of this thesis is as follows:1)A single-channel EEG signal acquisition device is designed to collect EEG and EOG signals.Firstly,the signal is extracted by silver chloride or hydrogel electrodes,followed by instrument amplification,low-pass filtering and post-amplification,and then converted into digital signals through the ADC.Finally,the signals are preprocessed by the MCU and sent to the PC via Bluetooth low energy.Common signal acquisition equipment is too large to carry,and its high-level power consumption results in short work time.In the thesis,the high-integration op amp,the MCU with built-in 2.4G wireless transceiver and the multi-layer layout of the circuit board make the size of the acquisition device as small as 1.8cm×3.0cm.Low-power Bluetooth technology is used as the communication protocol which reduces the power consumption of the device.And the device can work for 15.24 hours continuously.Thus,this device has the advantages of low cost,low power consumption and portability.2)An English word learning system is designed.The system uses the EEG-based attention detection result as the switching signal and the EOG as the decision signal.EEG-based attention detection detects the users' “attention” and “inattention” states.During the training stage,the training samples are evaluated by the TOVA test,which improves the reliability of the training samples.Attention detection uses power spectral density(PSD)as features and support vector machine(SVM)as the classifier.Ten subjects have participated in the attention test,and the average accuracy rate is 86.81%.This system use EOG as the decision signal to choose the correct answer in the English learning system.In order to shorten decision time,this thesis designs two stimulation paradigms,random and fixed.Paradigm is changed according to the number of flashing buttons displayed on the learning interface.As the experiment results shows that the average decision time of 10 subjects is 8.75 s and the decision accuracy rate is 97.98%.