An adaptive education content regulation method and system
By collecting learners' EEG signals in real time, calculating attention and affective valence indices, and dynamically adjusting teaching content, the problem of disconnect between teaching content and learners' states is solved, enabling personalized and dynamic teaching and improving teaching quality and efficiency.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- FOREIGN ECONOMIC & TRADE UNIV
- Filing Date
- 2026-03-11
- Publication Date
- 2026-06-05
Smart Images
Figure CN122152133A_ABST
Abstract
Description
Technical Field
[0001] This application relates to the interdisciplinary field of educational technology and biomedical engineering. Specifically, it relates to an adaptive educational content control method and system, particularly a closed-loop control method and system based on real-time acquisition and analysis of learners' electroencephalogram (EEG) signals to dynamically control the presentation of teaching content. Background Technology
[0002] In current educational practice, especially in teaching scenarios such as ideological and political theory courses that emphasize theoretical internalization and emotional resonance, the presentation of teaching content is often linear and standardized. Teachers deliver lectures according to a pre-set syllabus, making it difficult to quantitatively assess each learner's level of cognitive acceptance or emotional resonance with specific knowledge points in real time and objectively during the teaching process.
[0003] Traditional methods for evaluating teaching effectiveness, such as classroom questioning, post-class questionnaires, or observation of learners' facial expressions, have significant limitations. These methods are highly subjective, and the evaluation results are easily influenced by various factors. Furthermore, the data exhibits a significant lag, failing to capture changes in learners' cognitive states at the moment of information reception. In addition, some external behavioral manifestations can be misleading; learners may appear focused, but their minds may be wandering, resulting in a state of "ineffective learning," which is difficult to discern using traditional observation methods.
[0004] To address these issues, some computer vision-based online education attention monitoring systems have emerged, which infer learners' focus levels by capturing their eye movements, head posture, or facial micro-expressions through cameras. However, these methods still have limitations in specific educational scenarios. They struggle to distinguish between "effective fixation" and "ineffective fixation," failing to delve into the learner's deeper cognitive levels. Furthermore, when learners encounter abstract theories that are difficult to understand and experience cognitive blockage, vision-based systems cannot provide sufficient information to determine the cause of the blockage—whether it's due to inattention or excessively difficult content.
[0005] Furthermore, while some general AI recommendation algorithms can recommend content based on learners' historical behavior, they generally lack real-time feedback mechanisms based on physiological signals. This means that the system cannot immediately intervene and adjust teaching methods at the precise moment when learners experience cognitive confusion or overload, or instantly switch content to more easily understood concrete examples when learners are confused about an abstract theory.
[0006] Therefore, there is an urgent need for a technical solution that can perceive and respond to learners' internal cognitive and emotional states in real time and objectively, thereby achieving dynamic matching between teaching content and learners' acceptance ability, in order to solve the technical problem of the disconnect between teaching content and learners' actual state. Summary of the Invention
[0007] The main purpose of this application is to provide a self-adaptive educational content control method and system, which solves the technical problem in the prior art that the learner's cognitive and emotional state cannot be perceived in real time and objectively, resulting in a disconnect between teaching content and learner's acceptance ability.
[0008] To achieve the above objectives, this application provides an adaptive educational content control method, comprising:
[0009] The EEG signals of learners in the prefrontal and occipital regions are collected in real time using an EEG acquisition device.
[0010] While the teaching content is playing on the multimedia presentation terminal, a trigger marker is generated and sent at the beginning of each content segment.
[0011] The trigger marker is inserted into the EEG signal data stream formed by the EEG signal to achieve time alignment between the EEG signal and the teaching content;
[0012] The acquired EEG signals are preprocessed to remove artifacts;
[0013] Within a preset time window, based on the frequency domain characteristics of the preprocessed EEG signals, a focus index characterizing the learner's focus level is calculated. ) and the affective valence index, which represents the learner's emotional state ( );
[0014] According to the focus index ( ) and the emotional valence index ( The combination of these factors determines the learner's current cognitive-emotional state, which includes at least: a state of deep immersion with high concentration and positive emotions, a state of cognitive overload with high concentration and negative emotions, and a state of inattentiveness with low concentration.
[0015] Based on the determined cognitive-emotional state, the content generation engine is driven to adaptively adjust the presentation method of the teaching content in real time; wherein, when the determined cognitive-emotional state is the cognitive overload state, the dimensionality reduction interpretation method is triggered to switch the abstract theoretical content to concrete case content.
[0016] In a preferred embodiment, the calculation of the focus index ( ) and the emotional valence index ( The specific steps include: performing a Fourier transform on the EEG signal within the time window to obtain the power spectral density of each frequency band; and calculating the focus index based on the power spectral density of the β band, the α band, and the θ band. Based on the alpha band power spectral density of the EEG signals acquired from the corresponding electrodes on the left and right sides of the prefrontal cortex, the emotional valence index used to characterize prefrontal asymmetry is calculated. ).
[0017] In a preferred embodiment, the focus index ( ) through formula Perform calculations, where The average power spectral density of the β band, The average power spectral density of the α band. The average power spectral density of the θ band; the emotional valence index ( ) through formula Perform calculations, where The α-band average power spectral density of the right prefrontal cortex electrode is given. This represents the average power spectral density of the α-band of the left prefrontal cortex electrode.
[0018] In a preferred embodiment, the step of adaptively adjusting the presentation method according to the determined cognitive-emotional state further includes: when the cognitive-emotional state is determined to be the deep immersion state, driving the content generation engine to increase the abstraction level or information density of subsequent teaching content; when the cognitive-emotional state is determined to be the distracted state, driving the content generation engine to insert interactive questions or key concept reinforcement review segments related to the current knowledge point into the subsequent teaching content.
[0019] In a preferred embodiment, the step of determining the learner's current cognitive-emotional state is achieved by using the focus index ( ) and the emotional valence index ( This is achieved by comparing the focus level with a preset threshold, which includes a first focus threshold. ), second focus threshold ( First affective valence threshold ( ) and second affective valence threshold ( The condition for determining the cognitive overload state is as follows: and The criteria for determining the state of deep immersion are: and The criteria for determining the state of inattention are: .
[0020] In a preferred embodiment, the preprocessing step includes: performing bandpass filtering on the EEG signal from 0.5 Hz to 50 Hz; and using an independent component analysis algorithm to identify and remove artifacts related to eye movements, blinking, and muscle activity.
[0021] In a preferred embodiment, the method further includes: constructing a learner group cognitive-emotional state database to store the cognitive-emotional states of multiple learners for the same teaching content segment at corresponding times; when the frequency of occurrence of the cognitive overload state associated with a certain content segment recorded in the database exceeds a preset resistance judgment threshold, the content segment is marked as a high cognitive resistance point, and a teaching content optimization suggestion report is generated.
[0022] To achieve the above objectives, this application also provides an adaptive educational content control system, comprising:
[0023] EEG acquisition equipment is used to collect learners' EEG signals in the prefrontal and occipital regions in real time;
[0024] Multimedia presentation terminal, used to play teaching content;
[0025] The processing unit is electrically connected to both the EEG acquisition device and the multimedia presentation terminal.
[0026] The processing unit is configured as follows:
[0027] While the multimedia presentation terminal is playing the teaching content, a trigger marker is generated and sent at the beginning of each content segment.
[0028] The trigger marker is inserted into the EEG signal data stream formed by the EEG signals acquired by the EEG acquisition device;
[0029] The EEG signals are preprocessed;
[0030] Within a preset time window, based on the frequency domain characteristics of the preprocessed EEG signals, an attention index is calculated. ) and affective valence index ( );
[0031] According to the focus index ( ) and the emotional valence index ( The combination of these factors determines the learner’s current cognitive-emotional state, which includes at least a state of deep immersion, a state of cognitive overload, and a state of inattention.
[0032] Based on the determined cognitive and emotional state, the content generation engine is driven to adaptively adjust the presentation logic of the teaching content in the next moment, and when the cognitive overload state is determined, the dimensionality reduction explanation logic is triggered.
[0033] In a preferred embodiment, the processing unit includes a microprocessor and a non-transitory computer-readable storage medium coupled to the microprocessor, the non-transitory computer-readable storage medium storing computer program instructions that, when executed by the microprocessor, implement the method described in any of the preceding claims.
[0034] In a preferred embodiment, the EEG acquisition device is a head-mounted EEG cap, on which the electrodes for acquiring the EEG signals are arranged according to the international 10-20 standard electrode placement method, at electrode points including Fp1, Fp2, F3, F4, O1, and O2.
[0035] This application decodes learners' cognitive-emotional states by collecting their EEG signals in real time and combining these signals with two dimensions: focus and affective valence. This allows for a more precise differentiation between easily confused learning states such as "deep immersion" and "cognitive overload." Based on this, the system can trigger targeted content regulation logic. For example, when learners experience cognitive overload, the system automatically switches to more easily understood concrete examples, achieving precise intervention for learning difficulties. Simultaneously, by triggering markers, the system achieves precise synchronization between EEG signals and teaching knowledge points, providing objective and quantitative data support for identifying "high cognitive resistance points" in the course and optimizing instructional design. In summary, this application achieves personalized and dynamic teaching centered on learners' real-time brain states, ensuring that teaching content is always within the learner's "zone of proximal development," significantly improving the internalization efficiency of theoretical knowledge and the quality of teaching. Attached Figure Description
[0036] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.
[0037] Figure 1 This is a structural block diagram of an adaptive educational content control system provided in an embodiment of this application.
[0038] Figure 2 This is a flowchart of an adaptive educational content control method provided in an embodiment of this application.
[0039] Figure 3This is a two-dimensional spatial diagram provided in the embodiments of this application for dividing cognitive emotional states.
[0040] Figure 4 This is a schematic diagram of the closed-loop optimization mechanism for teaching content provided in the embodiments of this application. Detailed Implementation
[0041] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.
[0042] This embodiment provides an adaptive educational content adjustment method, which can be applied to a neurofeedback teaching system based on a B / S architecture. See also... Figure 1 The system may include an EEG acquisition device 101, a multimedia presentation terminal 102, and a processing unit 103 in terms of hardware. The processing unit 103 may further include an edge computing gateway and a central processing server. The various components of the system can communicate with each other via wired or wireless networks, following the corresponding network communication protocols. The system also includes a power supply module to power each hardware device.
[0043] like Figure 2 As shown, the specific process S200 of the adaptive educational content control method described in this embodiment is as follows:
[0044] Step S201: Collect the learner's brain signals in real time using an EEG acquisition device.
[0045] In this embodiment, the EEG acquisition device 101 is a non-invasive head-mounted EEG cap with a semi-circular structure and embedded dry comb-shaped electrodes for easy wear by learners. The electrode layout follows the international 10-20 standard electrode placement method, focusing on covering the prefrontal and occipital regions. Specifically, to acquire signals related to emotional valence, electrodes are placed at least at the Fp1, Fp2, F3, and F4 points in the prefrontal lobe; to acquire signals related to visual processing, electrodes are placed at least at the O1 and O2 points in the occipital lobe. The EEG acquisition device 101 has a built-in analog-to-digital converter module for converting the acquired analog EEG signals into digital signals and transmitting the digitized EEG signal data stream to the processing unit 103 in real time via wireless transmission modules such as Bluetooth 5.0.
[0046] Step S202: Generate and synchronously trigger markers while playing the teaching content.
[0047] The multimedia presentation terminal 102 is responsible for playing teaching content to the learner. To ensure that the EEG signals are precisely aligned with the teaching content in time during subsequent analysis, the multimedia presentation terminal 102 generates a timestamp at the start of playing each individual teaching content segment and encodes it as a trigger flag. In this embodiment, the trigger flag can be an 8-bit data packet. The trigger flag is sent to the processing unit 103.
[0048] Step S203: Insert the trigger marker into the EEG signal data stream.
[0049] After receiving a trigger marker from the multimedia presentation terminal 102, the processing unit 103 immediately inserts it into the EEG signal data stream that is being received in real time by the EEG acquisition device 101. In this way, each trigger marker acts as a time anchor point, precisely associating a specific EEG signal segment with the teaching content segment that elicited the response.
[0050] Step S204: Preprocess the EEG signals.
[0051] The raw EEG signal is very weak and easily interfered with by various internal and external noises, which are called artifacts. To extract a pure EEG signal that reflects brain activity, the processing unit 103 performs a series of preprocessing steps. The received raw EEG signal data stream is digitally filtered; specifically, a bandpass filter is applied, preferably with a passband range of 0.5Hz to 50Hz. The 0.5Hz high-pass cutoff frequency is used to filter out baseline drift caused by head micro-movements or poor electrode-scalp contact, while the 50Hz low-pass cutoff frequency is used to suppress power frequency interference introduced by the AC power grid. For physiological artifacts commonly generated during learning, such as eye movements, blinking, and muscle activity, this embodiment uses an independent component analysis algorithm for removal. This algorithm decomposes the multi-channel EEG signal into a set of statistically independent source signal components. By calculating the correlation between each independent component and a pre-recorded or standard electrooculography (EOG) reference signal, artifact components highly correlated with eye movements, blinking, and other activities can be identified. In this embodiment, when the correlation coefficient between a certain independent component and the electrooculogram (EOG) reference signal is greater than a preset artifact detection threshold, that component is identified as an artifact and removed from the signal. The remaining pure EEG signal components are then used to reconstruct the artifact-free EEG signal for subsequent feature calculations.
[0052] Step S205: Calculate the focus index and the affective valence index.
[0053] Processing unit 103 segments the preprocessed EEG signal within a preset time window. Within each time window, two core quantitative indicators are calculated through frequency domain analysis.
[0054] Specifically, the processing unit 103 performs a fast Fourier transform on the EEG signal within the time window to obtain the energy distribution of the signal at different frequencies, i.e., the power spectral density.
[0055] One is the focus index ( This index quantifies a learner's level of focus. It is calculated based on the power spectral density of a specific frequency band. In this embodiment, the focus index is implemented using the following formula:
[0056]
[0057] in, The average power spectral density of the β band is positively correlated with the brain's excitation, alertness, and active thinking states. This represents the average power spectral density of the alpha band, whose energy is associated with a relaxed and calm state of the brain. Let θ be the average power spectral density of the frequency band, whose energy is associated with drowsiness or a state of mental distraction. Therefore, The higher the value, the higher the learner's level of concentration.
[0058] The second is the affective valence index ( The affective valence index is used to quantify whether a learner's emotional state is positive or negative. It is calculated using the prefrontal asymmetry theory, which states that relatively weaker alpha wave activity in the left prefrontal cortex is associated with positive emotions, while relatively weaker alpha wave activity in the right prefrontal cortex is associated with negative emotions. In this embodiment, the affective valence index is implemented using the following formula:
[0059]
[0060] in, The average power spectral density of the alpha band of the EEG signal acquired by the right electrode in the prefrontal cortex. Let be the average power spectral density of the α-band of the left prefrontal cortex electrode. According to this formula, when... A value greater than 0 indicates that the learner is in a positive or identifying emotional state; when... When the value is less than 0, it indicates that the learner is in a negative, resistant, or confused emotional state.
[0061] Step S206: Determine the learner's cognitive and emotional state.
[0062] See Figure 3 Processing unit 103 constructs a focus index (... ) is a one-dimensional, affective valence index ( Let be another two-dimensional state space. This is achieved through computation. and The learner's current cognitive-emotional state is determined by comparing their current state to a set of preset thresholds. These preset thresholds include the first attention threshold. Second attention threshold First emotional valence threshold Second emotional valence threshold .
[0063] In this embodiment, at least the following three core cognitive-emotional states are defined:
[0064] 1. Cognitive Overload State 301: The criteria for this state are that the attention index is above a high threshold, but the affective valence index is below a negative threshold. and This situation reflects that although learners are trying to concentrate and think, they feel confused, frustrated, or resistant to the teaching content, which is a key challenge in the teaching process.
[0065] 2. Deep Immersion State 302: The criteria for its determination are that both the focus index and the affective valence index are high, i.e. and This indicates that the learner is highly focused and has a positive attitude towards the content, which is an ideal learning state.
[0066] 3. Inattention State 303: The criterion for this is that the focus index is below a low threshold, i.e. This indicates that the learner's mind is wandering and they are not investing cognitive resources in the current learning task.
[0067] Step S207: Adaptively adjust the presentation method of teaching content.
[0068] Processing unit 103 runs a content generation engine connected to a hierarchical material database. The materials in this database can be categorized into different layers, such as core theory, historical case studies, and emotional rendering. Based on the cognitive-emotional state determined in step S206, the content generation engine adaptively adjusts the teaching content to be presented to learners in real time.
[0069] The specific regulatory logic is as follows:
[0070] When a cognitive overload state 301 is detected, the system triggers the "dimensionality reduction explanation method." The content generation engine immediately pauses the current abstract theoretical description and retrieves more concrete content corresponding to the current knowledge point from the material database for presentation, such as a specific historical case video or a vivid animated demonstration. At the same time, to reduce the learner's cognitive load, the parameters of the presented content can be adjusted, reducing the playback speed of the video or audio by 15%.
[0071] When the system is identified as being in a deep immersion state (302), it indicates that the difficulty and depth of the current content are appropriate, or even slightly below the learner's comprehension level. At this point, the system can drive the content generation engine to appropriately increase the abstraction level or information density of subsequent teaching content to maintain the learner's sense of challenge and keep them in the "zone of proximal development."
[0072] When an inattentive state (303) is detected, the system triggers the "sensory arousal method." The content generation engine can insert attention-grabbing elements into subsequent teaching content, such as an interactive question related to the current knowledge point, or a review segment reinforcing key concepts. Furthermore, it can stimulate the senses by adjusting the physical parameters of multimedia presentations, such as increasing screen contrast by 10% or boosting the gain of high-frequency components in audio by 3dB, to awaken the learner's attention.
[0073] Through the aforementioned dual-dimensional state decoding and multi-state regulation logic, this method can accurately identify learners' learning states and provide targeted teaching interventions, thereby achieving personalized adaptive adjustment of teaching content.
[0074] In addition, see Figure 4 This method also includes a closed-loop feedback mechanism for iterative optimization of teaching content. Processing unit 103 constructs and maintains a learner group cognitive-emotional state database 401. This database stores the cognitive-emotional state sequences generated by multiple learners during corresponding time periods identified by trigger markers when learning the same teaching content segment. When the system performs offline analysis of the data in the database, if it finds that a specific content segment generally triggers cognitive overload among the learner group, the system will automatically mark the content segment as a "high cognitive resistance point" 403. Subsequently, the system will generate a teaching content optimization suggestion report 404, which will be fed back to the teaching research system or curriculum designer, providing objective evidence based on massive amounts of physiological data for subsequent curriculum iteration and optimization. This mechanism realizes closed-loop optimization from teaching implementation to instructional design.
[0075] This embodiment provides an adaptive educational content control system, the specific hardware structure and connection relationships of which are as follows: Figure 1 As shown. The system includes an EEG acquisition device 101, a multimedia presentation terminal 102, and a processing unit 103.
[0076] The EEG acquisition device 101 is used to acquire the learner's EEG signals in real time. As described in the previous embodiment, the device is preferably a head-mounted EEG cap containing dry electrodes. The electrodes are placed at specific locations on the learner's scalp according to the international 10-20 standard electrode placement method, including but not limited to the Fp1, Fp2, F3, and F4 points in the prefrontal cortex, and the O1 and O2 points in the occipital lobe. The device processes the raw signal through built-in signal amplification, filtering, and analog-to-digital conversion circuits, and transmits the digitized EEG signal data stream through a wireless communication module.
[0077] Multimedia presentation terminal 102 is used to play teaching content to learners. This terminal can be a personal computer, tablet, smartphone, or dedicated VR / AR device. Client software runs on this terminal, which is responsible not only for decoding and presenting the content but also for generating and sending trigger markers to processing unit 103 at key time points.
[0078] The processing unit 103 is the control core of the entire system. It establishes electrical or wireless communication connections with the EEG acquisition device 101 and the multimedia presentation terminal 102, respectively. In one specific implementation, the processing unit 103 can be a single high-performance server or a distributed system consisting of an edge computing gateway and a cloud-based central processing server. The processing unit 103 includes at least one microprocessor and a non-transitory computer-readable storage medium coupled to the microprocessor. The storage medium stores computer program instructions.
[0079] When the microprocessor executes these computer program instructions, it is configured to perform the adaptive educational content control method described in the foregoing embodiments. Specifically, the processing unit 103 is configured to:
[0080] 1. Simultaneously receive EEG signal data streams from EEG acquisition device 101 and trigger flags from multimedia presentation terminal 102.
[0081] 2. Perform time synchronization operations to precisely insert trigger markers into the EEG signal data stream.
[0082] 3. Perform signal preprocessing algorithms, including bandpass filtering and ICA-based artifact removal.
[0083] 4. Within the preset time window, perform frequency domain analysis to calculate the focus index ( ) and affective valence index ( ).
[0084] 5. Based on these two indices and the preset threshold logic, determine the learner's current cognitive and emotional state.
[0085] 6. Based on the determined state, send control commands to the content generation engine deployed on itself or the multimedia presentation terminal 102 to drive the adaptive adjustment of the teaching content in the next moment, and trigger the dimensionality reduction explanation logic when the state of cognitive overload is determined.
[0086] In summary, this system, through the collaborative work of its various hardware components, achieves real-time monitoring and decoding of learners' cognitive and emotional states, as well as closed-loop control of teaching content based on this, thereby effectively improving the personalization level of teaching and learning efficiency.
[0087] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some or all of the technical features therein. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of this application.
Claims
1. An adaptive educational content control method, characterized in that, include: The EEG data of learners in the prefrontal and occipital regions is collected in real time using an EEG acquisition device; while the teaching content is played on a multimedia presentation terminal, a trigger marker is generated and sent at the beginning of each content segment; the trigger marker is inserted into the EEG signal data stream formed by the EEG signals to achieve time alignment between the EEG signals and the teaching content. The acquired EEG signals are preprocessed to remove artifacts; Within a preset time window, based on the frequency domain characteristics of the preprocessed EEG signals, a focus index characterizing the learner's focus level is calculated. ) and the affective valence index, which represents the learner's emotional state ( According to the aforementioned focus index ( ) and the emotional valence index ( The system uses a combination of factors to determine the learner's current cognitive-emotional state, which includes at least: a state of deep immersion with high concentration and positive emotions, a state of cognitive overload with high concentration and negative emotions, and a state of low concentration and inattentiveness. Based on the determined cognitive-emotional state, the content generation engine is driven to adaptively adjust the presentation method of the teaching content in real time. When the cognitive-emotional state is determined to be the cognitive overload state, a dimensionality reduction interpretation method is triggered to switch the abstract theoretical content to concrete case content.
2. The method according to claim 1, characterized in that, The calculation of the focus index ( ) and the emotional valence index ( The specific steps include: performing a Fourier transform on the EEG signal within the time window to obtain the power spectral density of each frequency band; and calculating the focus index based on the power spectral density of the β band, the α band, and the θ band. Based on the alpha band power spectral density of the EEG signals acquired from the corresponding electrodes on the left and right sides of the prefrontal cortex, the emotional valence index used to characterize prefrontal asymmetry is calculated. ).
3. The method according to claim 2, characterized in that, The focus index ( ) through formula Perform calculations, where The average power spectral density of the β band, The average power spectral density of the α band. The average power spectral density of the θ band; The emotional valence index ( ) through formula Perform calculations, where The α-band average power spectral density of the right prefrontal cortex electrode is given. This represents the average power spectral density of the α-band of the left prefrontal cortex electrode.
4. The method according to claim 1, characterized in that, The step of adaptively adjusting the presentation method based on the determined cognitive-emotional state further includes: when the determined cognitive-emotional state is the deep immersion state, driving the content generation engine to increase the abstraction level or information density of subsequent teaching content; when the determined cognitive-emotional state is the distracted state, driving the content generation engine to insert interactive questions or key concept reinforcement review segments related to the current knowledge point into subsequent teaching content.
5. The method according to claim 4, characterized in that, The step of determining the learner's current cognitive and emotional state involves using the focus index ( ) and the emotional valence index ( This is achieved by comparing the focus level with a preset threshold, which includes a first focus threshold. ), second focus threshold ( First affective valence threshold ( ) and second affective valence threshold ( The condition for determining the cognitive overload state is as follows: and The criteria for determining the state of deep immersion are: and The criteria for determining the state of inattention are: .
6. The method according to claim 1, characterized in that, The preprocessing steps include: performing bandpass filtering on the EEG signal from 0.5 Hz to 50 Hz; and using an independent component analysis algorithm to identify and remove artifacts related to eye movements, blinking, and muscle activity.
7. The method according to claim 1, characterized in that, The method further includes: constructing a learner group cognitive-emotional state database to store the cognitive-emotional states of multiple learners for the same teaching content segment at corresponding times; when the frequency of occurrence of the cognitive overload state associated with a certain content segment recorded in the database exceeds a preset resistance judgment threshold, the content segment is marked as a high cognitive resistance point, and a teaching content optimization suggestion report is generated.
8. An adaptive educational content control system, characterized in that, include: EEG acquisition equipment is used to collect learners' EEG signals in the prefrontal and occipital regions in real time; Multimedia presentation terminal, used to play teaching content; The processing unit is electrically connected to the EEG acquisition device and the multimedia presentation terminal, respectively; the processing unit is configured to generate and send a trigger flag at the start time of each content segment while the teaching content is being played on the multimedia presentation terminal. The trigger marker is inserted into the EEG signal data stream formed by the EEG signals acquired by the EEG acquisition device; The EEG signals are preprocessed; Within a preset time window, based on the frequency domain characteristics of the preprocessed EEG signals, an attention index is calculated. ) and affective valence index ( According to the aforementioned focus index ( ) and the emotional valence index ( The combination of these factors determines the learner's current cognitive-emotional state, which includes at least a state of deep immersion, a state of cognitive overload, and a state of distracted attention. Based on the determined cognitive-emotional state, the content generation engine is driven to adaptively adjust the presentation logic of the teaching content in the next moment, and when the cognitive overload state is determined, the dimensionality reduction explanation logic is triggered.
9. The system according to claim 8, characterized in that, The processing unit includes a microprocessor and a non-transitory computer-readable storage medium coupled to the microprocessor, wherein the non-transitory computer-readable storage medium stores computer program instructions that, when executed by the microprocessor, implement the method of claim 1.
10. The system according to claim 8, characterized in that, The EEG acquisition device is a head-mounted EEG cap, on which the electrodes for acquiring the EEG signals are arranged according to the international 10-20 standard electrode placement method, including electrode points Fp1, Fp2, F3, F4, O1, and O2.