Reaction recognition method and device based on game scenarios

Abstract

This invention provides a reaction speed recognition method and device based on a game scenario. When a user enters a preset reaction speed game scenario wearing a preset wearable device, a reaction speed baseline task is initiated to acquire baseline EEG data during the user's operation of the reaction speed baseline task. Non-invasive physiological signal acquisition is achieved through the preset wearable device. After that, the formal game task is carried out. Based on the acquired real-time EEG data and baseline index data, an EEG reaction speed index score is calculated. Based on the acquired real-time behavioral data, a cognitive index score is calculated. The EEG reaction speed index score and the cognitive index score are fused to obtain a comprehensive reaction speed cognitive index score. Using the comprehensive reaction speed cognitive index score, the game event interval time is adjusted to match the task rhythm and the player's reaction speed level in real time, taking into account both player experience and the rigor of objective evaluation.

Description

Technical Field

[0001] This invention relates to the field of human-computer interaction diagnostic technology, and in particular to a method and apparatus for recognizing reaction force based on game scenarios. Background Technology

[0002] In numerous task scenarios, including education and training, driving simulation, rehabilitation intervention, sports competition, and immersive entertainment, reaction time is a key indicator for measuring an individual's perception, information processing, and speed of action execution. It not only directly affects learning efficiency and operational safety but is also closely related to reaction time maintenance, stress tolerance, and motor coordination. Instantaneous fluctuations in reaction time often determine whether an individual can maintain stable performance in dynamic tasks, such as a driver's emergency braking in complex road conditions, the action feedback of a rehabilitation patient during training, and the operational response of an e-sports player under high-pressure environments.

[0003] Traditional assessment methods typically rely on standardized tests or fixed tasks, which, while providing baseline metrics, lack the ability to capture real-time changes. With the development of interactive technologies, virtual gamification scenarios are gradually demonstrating inherent advantages: they can create highly immersive ecosystems while continuously collecting behavioral data without interrupting the experience. This transforms games from being limited to entertainment into a real-time quantitative tool for cognitive and behavioral abilities.

[0004] However, most existing systems use static scores or preset difficulty curves to measure player performance. This approach has significant limitations when dealing with fluctuations in reaction time: when players are in a bad state, the task pace is still too fast, easily leading to frequent errors and a negative experience; conversely, when players are in a good state, insufficient task intensity reduces focus and engagement. This mismatch between pace and state directly weakens the flow experience in training, teaching, and entertainment.

[0005] In summary, existing technologies suffer from a mismatch between the task pace and the player's current reaction time. Summary of the Invention

[0006] This invention provides a reaction speed recognition method and apparatus based on game scenarios to solve the defect of mismatch between task rhythm and player's current reaction speed state in the prior art, and to achieve reaction speed recognition based on game scenarios that matches task rhythm and player's current reaction speed state.

[0007] This invention provides a reaction speed recognition method based on game scenarios, comprising the following steps: In response to a user wearing a pre-set wearable device entering a pre-set reaction game scene, the reaction baseline task begins; During the user's operation of the reaction baseline task, the user's baseline EEG data is acquired, and baseline index data is calculated based on the baseline EEG data. In response to the completion of the reaction baseline task, different game events are displayed to the user based on the historical game event intervals; During the process of the user performing the game event according to the preset requirements, the user's real-time EEG data and real-time behavioral data are collected based on the preset wearable device, wherein the real-time behavioral data includes operation reaction time, number of effective reactions and number of correct operations; Based on the real-time EEG data and the baseline index data, the EEG reaction index score is calculated, and based on the real-time behavioral data, the cognitive index score is calculated. Based on the EEG reaction index score and the cognitive index score, a comprehensive reaction cognitive index score is calculated as the reaction recognition result; The game event interval is adjusted based on the comprehensive reaction ability cognitive index score.

[0008] According to the present invention, a reaction time recognition method based on a game scene is provided, wherein the reaction time baseline task includes a non-reaction part and a reaction part.

[0009] According to the present invention, a reaction recognition method based on a game scene is provided, wherein the preset wearable device includes a non-invasive EEG acquisition device, and the baseline EEG data and the real-time EEG data are acquired through the acquisition electrodes of the non-invasive EEG acquisition device.

[0010] According to the present invention, a reaction speed recognition method based on a game scenario calculates baseline index data based on the baseline EEG data, including: The baseline EEG data is filtered, and based on the filtering result, the corresponding baseline EEG data of the first preset frequency band and the second preset frequency band are extracted. Based on the corresponding baseline EEG data of the first preset frequency band and the corresponding baseline EEG data of the second preset frequency band, the frequency band energy ratio of the first preset frequency band and the second preset frequency band, as well as the inhibition ratio of the second preset frequency band, are calculated as baseline features. Calculate the average and maximum values ​​of the baseline features as calibration values.

[0011] According to the present invention, a reaction ability recognition method based on a game scenario calculates a brainwave reaction ability index score based on the real-time EEG data and the baseline index data, including: The real-time EEG data and the real-time behavioral data are fuzz-aligned and smoothed using a preset time window. The smoothed real-time EEG data is filtered, and based on the filtering result, the corresponding real-time EEG data of the first preset frequency band and the second preset frequency band are extracted. Based on the real-time EEG data corresponding to the first preset frequency band and the real-time EEG data corresponding to the second preset frequency band, the real-time frequency band energy ratio and the second preset frequency band suppression ratio are calculated as feature indicators. The feature index is centered based on the calibration value, and the result of the centered processing is mapped to a first preset interval to obtain the mapping result; The mapping result is smoothly mapped to a second preset interval to obtain the EEG reaction index score.

[0012] According to the present invention, a reaction speed recognition method based on a game scenario calculates a cognitive index score based on the real-time behavior data, including: Using the first preset formula, a cognitive index score is calculated based on the real-time behavioral data; The first preset formula includes:

[0013]

[0014]

[0015]

[0016]

[0017]

[0018] in, For accuracy scores, For the correct number of responses, The number of responses that should be received within the window. For the accuracy hyperparameter weights, The reaction rate fraction, Effective reaction time It is the optimal reaction time hyperparameter. It is the maximum reaction time hyperparameter. For the number of effective reactions, For velocity weighting hyperparameters, For consistency score, The standard deviation of the reaction time. The average reaction time. For consistency hyperparameter weights, For the total score, Here, k is the cognitive index score, k is the sigmoid slope parameter, and m is the sigmoid midpoint parameter.

[0019] According to the present invention, a reaction speed recognition method based on a game scenario calculates a cognitive index score based on the real-time behavior data, including: Using the second preset formula, a cognitive index score is calculated based on the real-time behavioral data; The second preset formula includes:

[0020]

[0021]

[0022]

[0023]

[0024] ;

[0025] in, To score the accuracy of mathematical tasks, For the weight hyperparameters of the mathematical task, The penalty coefficient hyperparameter, For the correct number of mathematical tasks, This represents the total number of mathematical tasks. Score the accuracy of key presses. This is a hyperparameter for the weight of key presses. The penalty coefficient hyperparameter, The correct number of key presses. This represents the total number of key presses. For effective reaction time, For the overall speed score, The optimal reaction time hyperparameter is... The maximum effective reaction time hyperparameter, For velocity weighting hyperparameters, For effective reaction quantity, This represents the total amount of reaction required. For consistency score, Standard deviation, The mean, For consistency weight hyperparameter, For the number of valid samples, For the overall score, Here, k is the cognitive index score, k is the sigmoid slope parameter, and m is the sigmoid midpoint parameter.

[0026] The present invention also provides a reaction recognition device based on a game scene, comprising the following modules: The baseline task unit is used to respond to the user wearing a preset wearable device and entering a preset reaction game scene to start the reaction baseline task; The baseline data unit is used to acquire the user's baseline EEG data during the user's operation of the reaction baseline task, and to calculate baseline index data based on the baseline EEG data. The game event unit is used to display different game events to the user in response to the completion of the reaction baseline task, based on the historical game event interval time. The data acquisition unit is used to collect the user's real-time EEG data and real-time behavioral data based on the preset wearable device during the process of the user operating the game event according to preset requirements. The real-time behavioral data includes operation reaction time, number of effective reactions, and number of correct operations. The score calculation unit is used to calculate the brainwave reaction index score based on the real-time EEG data and the baseline index data, and to calculate the cognitive index score based on the real-time behavioral data. The comprehensive score unit is used to calculate the comprehensive reaction cognitive index score based on the EEG reaction index score and the cognitive index score, as the reaction recognition result; The feedback adjustment unit is used to adjust the game event interval time based on the comprehensive reaction ability cognitive index score.

[0027] The present invention also provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement the reaction recognition method based on the game scene as described above.

[0028] The present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the reaction recognition method based on the game scene as described above.

[0029] The present invention also provides a computer program product, including a computer program that, when executed by a processor, implements the reaction recognition method based on any of the above-described game scenarios.

[0030] The present invention provides a reaction speed recognition method and device based on game scenarios. When a user enters a preset reaction speed game scenario wearing a preset wearable device, a reaction speed baseline task is initiated. Baseline EEG data is acquired during the user's operation of the baseline task, thereby calculating baseline index data to reduce individual user differences. The preset wearable device achieves non-invasive physiological signal acquisition, which is low-cost, easy to wear, and provides a good user experience. Then, the formal game task is performed. Based on the acquired real-time EEG data and baseline index data, an EEG reaction speed index score is calculated, and based on the acquired real-time behavioral data, a cognitive index score is calculated. The EEG reaction speed index score and the cognitive index score are fused to obtain a comprehensive reaction speed-cognitive index score, which accurately reflects the player's reaction speed fluctuations during real-time gameplay. Using the comprehensive reaction speed-cognitive index score, the game event interval time is adjusted to match the task rhythm with the player's reaction speed level in real time, maintaining the optimal load range during training or entertainment, avoiding "too fast leading to stress" or "too slow leading to fatigue," and balancing player experience with the rigor of objective evaluation. Attached Figure Description

[0031] To more clearly illustrate the technical solutions in this invention 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 some embodiments of this invention. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0032] Figure 1 This is a flowchart illustrating the reaction recognition method based on game scenarios provided by the present invention; Figure 2 This is a schematic diagram of the reaction recognition results of subject 1 in the Brain Rider game using the reaction recognition method based on game scenarios provided by this invention. Figure 3 This is a schematic diagram of the reaction recognition results of subject 2 in the Brain Rider game using the reaction recognition method based on game scenarios provided by this invention; Figure 4 This is a schematic diagram of the reaction recognition results of subject 1 in a parallel challenge game using the reaction recognition method based on game scenarios provided by this invention; Figure 5 This is a schematic diagram of the reaction recognition results of subject 2 in a parallel challenge game using the reaction recognition method based on game scenarios provided by this invention; Figure 6 This is a schematic diagram of the reaction force recognition device based on game scenes provided by the present invention; Figure 7 This is a schematic diagram of the structure of the electronic device provided by the present invention. Detailed Implementation

[0033] To make the objectives, technical solutions, and advantages of this invention clearer, the technical solutions of this invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of this invention. All other embodiments obtained by those skilled in the art based on the embodiments of this invention without creative effort are within the scope of protection of this invention.

[0034] In existing technologies, user reaction ability is typically assessed in the following three ways: 1) Pre-set psychological paradigm stimulation.

[0035] A common approach is to embed classic paradigms such as Simple Reaction Time (SRT), Choice Reaction Time (CRT), Stroop Task, Simon Task, and multi-object tracking as independent modules into the training or game process, or insert them between levels. Key metrics collected include average reaction time, reaction time distribution characteristics, error rate, operational latency, and stability over time. Sometimes, a sliding window method is used to analyze individual speed fluctuations and reaction stability at different stages. The advantages of this method are: high reproducibility (paradigm parameters are controllable, and results show good consistency under different experimental conditions); clear interpretation path (there is a relatively clear correspondence between reaction time and execution control and inhibition ability); and benchmark evaluation value (it can serve as a stage-wise measurement of a player's initial reaction speed and fault tolerance threshold). However, this method also has the following key drawbacks: Insufficient ecological validity: The experimental paradigm is disconnected from real-world interactive scenarios, lacking a natural feel and entertainment value; Poor real-time performance: It usually only provides interim results, making it difficult to track the player's second-level reaction fluctuations in a game; Disruption of the main flow: The paradigm task is inconsistent with the main gameplay logic, and its insertion disrupts immersion and continuity; Limited transferability: The score in the paradigm may not be consistent with the player's reaction performance in complex tasks; High engineering implementation cost: It requires high-precision time synchronization and reaction acquisition, otherwise data errors will be significant; Insufficient regulatory value: The output results are mostly statistical levels, which cannot be directly mapped to the rhythm or speed parameters in the game, and frequent insertions will reduce playability.

[0036] 2) Post-verification method based on questionnaires.

[0037] Specifically, after completing the task, individuals' feelings about their reaction speed, agility, or engagement are collected through questionnaires or subjective evaluations. Commonly used tools include NASA-TLX (including a time pressure dimension) and SWAT, with results often used as a control or auxiliary label for model training. The advantages of this method are: low cost (no complex equipment required, low implementation threshold); and broad coverage (suitable for large-scale population data collection). However, this method also has the following key drawbacks: non-real-time (only provides retrospective information, unable to reflect instantaneous fluctuations in reaction time, disconnected from online closed-loop control); significant subjective bias (easily influenced by emotions, task impressions, and answering habits, limiting the reliability of results); poor consistency (different groups perceive "fast" and "slow" differently, lacking cross-cultural and cross-group comparability); difficulty in directly driving tasks (questionnaire scores cannot be directly mapped to continuously adjustable parameters, such as operational rhythm and task refresh rate); and data quality issues (common omissions and arbitrary answers increase the difficulty of subsequent data cleaning).

[0038] 3) Assessment methods based on a single physiological signal.

[0039] Specifically, this method relies on single physiological or behavioral signals related to the reaction, such as: EEG (motor readiness potential, power changes in specific frequency bands), eye movements (fixation delay, scan rate), HRV and skin conductance (autonomic nervous system response under stress), and motor parameters (click delay, hand acceleration, trajectory deviation). These signals, combined with algorithmic models, can be used to infer reaction speed and stability. Some methods attempt to collect data in real time and fuse it with behavioral indicators. The advantages of this method are: strong objectivity (directly related to neural or motor pathways, difficult to fake); and online acquisition (some signals can be collected in real time during the task). However, this method also has the following key drawbacks: complex equipment (wearing multiple sensors affects natural operation and immersion); insufficient stability (lighting, posture, and motion artifacts can easily introduce noise); and limited interpretability (the correspondence between signals and reaction processes often depends on laboratory conditions, and the interpretability decreases after being transferred to game scenarios).

[0040] In summary, existing reaction assessment methods each have their value, but all have significant shortcomings: Standardized paradigms: high internal validity, but poor ecological validity and insufficient real-time performance, making it difficult to seamlessly integrate into dynamic games or training scenarios; Questionnaire surveys: simple to implement and suitable for large-scale implementation, but can only provide retrospective, low-resolution information, lacking immediacy and objectivity; Single physiological signals: objective, but lack robustness and interpretability, and are greatly affected by external environment and individual differences.

[0041] Therefore, the industry still lacks a unified engineering methodology that can characterize responsiveness in a real-time, stable, and individualized manner within a natural interactive environment, and directly drive closed-loop regulation. This gap results in: the pace of games or training failing to match an individual's current responsiveness; teaching and rehabilitation scenarios struggling to adapt to instantaneous fluctuations; and product operations lacking explainable and executable adjustment mechanisms, making it difficult to maintain a dynamic balance between challenge and achievement in immersive experiences.

[0042] Based on this, the present invention provides a method and apparatus for recognizing reaction force in game scenarios.

[0043] The following is combined with Figures 1-5 The present invention describes a reaction recognition method based on game scenarios. Figure 1 This is a flowchart illustrating the reaction recognition method based on game scenarios provided by the present invention, as shown below. Figure 1 As shown, the method includes: Step 110: In response to the user wearing the preset wearable device and entering the preset reaction game scene, start the reaction baseline task.

[0044] Due to individual differences among users, a standardized reaction baseline task is set up to collect reference calibration data of users' EEG data in order to avoid insufficient response caused by individual differences.

[0045] Wearable devices are used to collect users' electroencephalogram (EEG) data, aiming to reduce costs and improve user experience. Pre-defined wearable devices refer to those that facilitate EEG data collection, such as EEG headbands or portable EEG devices, designed to collect users' physiological signals with low equipment cost and minimal wearing burden.

[0046] Compared to traditional laboratory methods, this invention reduces equipment costs and wearing burden, while ensuring the ecological effectiveness of data collection and algorithm inference, and maintaining the user's immersive experience.

[0047] Step 120: During the user's operation of the reaction baseline task, acquire the user's baseline EEG data and calculate baseline index data based on the baseline EEG data.

[0048] The user performs a baseline reaction task. During the operation, the user's EEG data is acquired based on a preset wearable device and recorded as baseline EEG data.

[0049] Indicator features are extracted from baseline EEG data and denoted as baseline indicator data.

[0050] Step 130: In response to the completion of the reaction baseline task, display different game events to the user based on the historical game event intervals.

[0051] After the baseline task is completed, the formal reaction test game begins. During the reaction test game, game events are displayed at intervals based on the game event intervals in the historical time window.

[0052] The game content includes displaying game events to the user, who then performs actions based on the prompts for different game events.

[0053] The game difficulty is mapped based on the time interval between game events.

[0054] In one specific embodiment, the previous time window is selected as the historical time window.

[0055] Step 140: During the process of the user operating the game event according to the preset requirements, the user's real-time EEG data and real-time behavioral data are collected based on the preset wearable device, wherein the real-time behavioral data includes operation reaction time, number of effective reactions and number of correct operations.

[0056] Users participate in a reaction test game, responding to displayed game events and taking actions based on prompts.

[0057] During operation, real-time EEG data of the user is collected based on a preset wearable device, and real-time behavioral data of the user is collected according to the game situation.

[0058] Real-time behavioral data includes operation response time, number of effective responses, and number of correct operations.

[0059] Furthermore, the acquired real-time EEG data and real-time behavioral data can be saved to a database, and then read from the database in subsequent score calculation steps.

[0060] This invention integrates behavioral data and physiological signals to establish a high-frequency, low-invasive online reaction time estimation model, avoiding the problems of sensitivity to noise and poor stability associated with single signals. Through feature extraction and fusion modeling, it significantly improves reaction speed, accuracy, and the precision of identifying fluctuation trends, achieving millisecond- to second-level responses in complex and dynamic interactive environments.

[0061] Understandably, existing single-method approaches often fail to fully characterize the multidimensional features of responsiveness: psychological paradigms ensure internal validity of experiments but lack ecological transferability; questionnaires are simple and easy to use but lack real-time and detailed capabilities; physiological or motor signals can provide objective data but face limitations in robustness and applicability.

[0062] Therefore, this invention employs a multimodal fusion method to characterize the multidimensional features of responsiveness, including behavioral and physiological indicators.

[0063] Among them, behavioral indicators such as reaction time, operational accuracy, and task completion speed directly reflect outward performance; physiological indicators such as EEG characteristics, heart rate fluctuations, eye movement patterns, and electromyographic signals reveal an individual's internal response readiness and executive load.

[0064] The fused reaction force estimate is supported by both game-related performance and objective physiological factors, thereby improving the accuracy and stability of the assessment and providing a feasible path for real-time monitoring and closed-loop intervention in natural and dynamic game scenarios.

[0065] Step 150: Calculate the EEG reaction index score based on the real-time EEG data and the baseline index data, and calculate the cognitive index score based on the real-time behavioral data.

[0066] Baseline index data is used to centrally process real-time EEG data, and then the EEG responsiveness index score is calculated.

[0067] Cognitive index scores are calculated based on real-time behavioral data.

[0068] Step 160: Calculate the comprehensive reaction cognitive index score based on the EEG reaction index score and the cognitive index score, and use it as the reaction recognition result.

[0069] A comprehensive reaction time recognition model was constructed by integrating behavioral indicators and non-invasive physiological signals. The comprehensive reaction time cognitive index score is defined as follows: , Where k is the cognitive intensity hyperparameter, The score of EEG responsiveness index, The score represents the cognitive indicator.

[0070] This invention not only maintains stability and robustness across different population groups and scenarios, but also outputs reaction level in continuous numerical form, avoiding the fuzzy interpretation problem caused by traditional "discrete gradation" methods. This invention combines interpretability and transferability, and can automatically adapt to different task types and population characteristics, providing a universal solution for multiple application scenarios such as education and training, rehabilitation intervention, and entertainment experience.

[0071] Furthermore, the reaction recognition results are fed back to the top left corner of the game scene for visualization.

[0072] Furthermore, the reactive power level can be output in continuous numerical form, avoiding the interpretation ambiguity problem caused by the traditional "discrete hierarchical" method.

[0073] This invention can identify a player's reaction time in real time in a dynamic interactive environment. The algorithm not only outputs continuous numerical values ​​but also has clear interpretability, revealing the sources of reaction lag, error rate, and stability fluctuations, thus providing a reliable basis for closed-loop regulation.

[0074] Step 170: Adjust the game event interval time based on the comprehensive reaction ability cognitive index score.

[0075] Based on the comprehensive reaction ability cognitive index score, the game event interval time of the next time window is adjusted. This enables the establishment of an individual cognitive profile through physiological signal detection, thereby achieving personalized control, dynamically adjusting the difficulty of game tasks, and achieving real-time adaptation. This allows players to maintain a dynamic balance between challenge and ability, enhancing learning efficiency and immersive experience.

[0076] According to the example embodiment, the game event interval is: , [] represents the floor function.

[0077] This invention can instantly adjust the difficulty and pace of tasks based on the identified reaction level, achieving continuous, smooth, and personalized dynamic control. This closed-loop mechanism effectively avoids the problems of "frustration due to excessive difficulty" or "boredom due to excessive ease," ensuring that players / learners maintain a flow experience throughout training and gameplay. In applications such as driving / flight simulation, cognitive and physical training, rehabilitation intervention, education, and immersive entertainment, this mechanism significantly improves participation, satisfaction, and long-term value, while enhancing safety and effectiveness.

[0078] This invention enables the observation and modeling of dynamic changes in reactivity in an ecologically efficient virtual environment, providing an interpretable, real-time, and personalized control mechanism. It is widely applicable to fields such as education and training, rehabilitation intervention, flight simulation, competitive games, and immersive entertainment.

[0079] Understandably, this invention aims to address the following issues: training tasks are difficult to match with players' real-time states, leading to decreased learning and rehabilitation efficiency; the game and teaching processes struggle to maintain optimal challenge levels, limiting user experience; and the product operation side lacks a quantifiable and traceable closed-loop mechanism, resulting in insufficient user stickiness. The goal is to achieve a technical system with the following characteristics: rapid response: real-time estimation of reaction performance during tasks without interrupting natural interaction; individual adaptation: considering individual baseline differences and instantaneous fluctuations to achieve differentiated adjustments; task compatibility: applicable to various application scenarios such as driving, rehabilitation, competition, and education; interpretability and stability: ensuring transparent and observable control logic for easy teaching and operational applications.

[0080] This invention provides a reaction speed recognition method based on game scenarios. When a user enters a preset reaction speed game scenario wearing a pre-set wearable device, a reaction speed baseline task is initiated. Baseline EEG data is acquired during the user's operation of the baseline task, thereby calculating baseline index data to reduce individual user differences. The pre-set wearable device achieves non-invasive physiological signal acquisition, which is low-cost, easy to wear, and provides a good user experience. Then, the formal game task is performed. Based on the acquired real-time EEG data and baseline index data, an EEG reaction speed score is calculated, and based on the acquired real-time behavioral data, a cognitive index score is calculated. The EEG reaction speed score and the cognitive index score are fused to obtain a comprehensive reaction speed-cognitive index score, which accurately reflects the player's reaction speed fluctuations during real-time gameplay. Using this comprehensive reaction speed-cognitive index score, the game event interval time is adjusted to match the task rhythm with the player's reaction speed level in real time, maintaining the optimal load range during training or entertainment, avoiding "too fast leading to stress" or "too slow leading to fatigue," thus balancing player experience with the rigor of objective evaluation.

[0081] The reactive baseline task is further described below. In some embodiments, the reactive baseline task includes a non-reactive portion and a reactive portion.

[0082] According to the example embodiment, the non-reactive baseline portion employs a 30-second eye-closed task, while the reactive baseline portion employs a rapid response paradigm to induce a response.

[0083] The following provides a further description of the preset wearable device. In some embodiments, the preset wearable device includes a non-invasive EEG acquisition device, through which the baseline EEG data and the real-time EEG data are acquired via the acquisition electrodes of the non-invasive EEG acquisition device.

[0084] In this embodiment of the invention, a non-invasive electroencephalogram (EEG) acquisition device is used as a pre-set wearable device for acquiring physiological signals.

[0085] According to an example embodiment, the non-invasive EEG acquisition device includes a non-invasive four-lead EEG headband with acquisition electrodes F1, FP1, FP2 and F2, a reference electrode and a ground electrode located in the mastoid process, and a sampling rate of 250 Hz.

[0086] The calculation of baseline index data is further explained below. In some embodiments, the calculation of baseline index data based on the baseline EEG data includes: Step 121: Filter the baseline EEG data, and based on the filtering result, extract the corresponding baseline EEG data of the first preset frequency band and the second preset frequency band.

[0087] Baseline EEG data were filtered to remove noise.

[0088] According to the example embodiment, the filtering includes a 1-30Hz bandpass filter and a 50Hz notch filter.

[0089] Extract the frequency band energy of the first preset frequency band and the second preset frequency band, and use the frequency band energy ratio as the baseline feature.

[0090] The first and second preset frequency bands can be set according to actual conditions. For example, in a specific embodiment, the θ band (4-8 Hz) is selected as the first preset frequency band, and the α band (8-13 Hz) is selected as the second preset frequency band.

[0091] Step 122: Based on the corresponding baseline EEG data of the first preset frequency band and the corresponding baseline EEG data of the second preset frequency band, calculate the frequency band energy ratio of the first preset frequency band and the second preset frequency band, as well as the suppression ratio of the second preset frequency band, as baseline features.

[0092] Calculate the band energy ratio of the first preset frequency band and the second preset frequency band, and use the band energy ratio and the suppression ratio of the second preset frequency band as baseline features.

[0093] Step 123: Calculate the average and extreme values ​​of the baseline features as calibration values.

[0094] Then, the average and extreme values ​​of the baseline characteristics of the non-reactive and reactive parts were recorded as calibration values ​​in the game.

[0095] The following further explains the calculation of the EEG reaction ability index score. In one embodiment, the EEG reaction ability index score is calculated based on the real-time EEG data and the baseline index data, including: Step 151: Perform fuzzy alignment on the real-time EEG data and the real-time behavioral data, and smooth them using a preset time window.

[0096] Real-time EEG data and real-time behavioral data are fuzzily aligned according to timestamps, with a preset time window to offset instantaneous fluctuations and smooth the data.

[0097] According to an example embodiment, the preset length time window includes a time window that extracts the previous 35 seconds based on the current time.

[0098] Step 152: Filter the smoothed real-time EEG data, and based on the filtering result, extract the corresponding real-time EEG data of the first preset frequency band and the second preset frequency band.

[0099] The filtering of real-time EEG data and the extraction of the first and second preset frequency bands can be referenced from the relevant processing methods of baseline EEG data, which will not be elaborated here.

[0100] Step 153: Calculate the real-time frequency band energy ratio and the second preset frequency band suppression ratio based on the corresponding real-time EEG data of the first preset frequency band and the corresponding real-time EEG data of the second preset frequency band, and use them as feature indicators.

[0101] For the smoothed real-time EEG data, the band energy ratio is calculated in the same way as the baseline EEG data, and this is used as the real-time band energy ratio. Simultaneously, the second preset frequency band suppression ratio is calculated, and the real-time band energy ratio and the second preset frequency band suppression ratio are used as characteristic indicators.

[0102] Step 154: Center the feature index according to the calibration value, and map the result of the centering process to the first preset interval to obtain the mapping result.

[0103] The calculated calibration values ​​are used to standardize the feature indicators, eliminating individual differences. The result of the centering process is then mapped to a first preset interval to obtain the mapping result.

[0104] According to the example embodiment, the average indexes of FP1 and FP2 electrodes are centered using the statistical values ​​from the training phase, and then linearly mapped to the first preset interval [-2,2] to obtain the mapping result.

[0105] Step 155: Smoothly map the mapping result to the second preset interval to obtain the EEG reaction index score.

[0106] The mapping result is mapped to the second preset interval to obtain the EEG responsiveness index score.

[0107] According to the example embodiment, the formula is as follows: The mapping result x is smoothed and mapped to the second preset interval [0,100], which is used as the brain electrical responsiveness index score.

[0108] This invention, through feature extraction and baseline correction of the energy ratio of EEG bands, significantly optimizes real-time computation and algorithm inference speed while ensuring accuracy and robustness, ensuring low-latency operation in consumer devices and immersive applications, and is suitable for highly interactive and time-sensitive scenarios.

[0109] The calculation of the cognitive index score is further explained below. In some embodiments, the cognitive index score is calculated based on the real-time behavioral data, including: Using the first preset formula, a cognitive index score is calculated based on the real-time behavioral data; The first preset formula includes:

[0110]

[0111]

[0112]

[0113]

[0114]

[0115] in, For accuracy scores, For the correct number of responses, The number of responses that should be received within the window. For the accuracy hyperparameter weights, The reaction rate fraction, Effective reaction time It is the optimal reaction time hyperparameter. It is the maximum reaction time hyperparameter. For the number of effective reactions, For velocity weighting hyperparameters, For consistency score, The standard deviation of the reaction time. The average reaction time. For consistency hyperparameter weights, For the total score, Here, k is the cognitive index score, k is the sigmoid slope parameter, and m is the sigmoid midpoint parameter.

[0116] This embodiment uses the Brain Rider game as an example to illustrate a formal reaction ability test. The game lasts for 5 minutes. Players need to operate according to the traffic lights in the upper right corner. Red lights and lightning bolt lights require stopping, yellow lights require no operation, and green lights require moving forward. The upper part displays the real-time reaction ability value, which controls the current speed of the electric vehicle, i.e., the event interval time. The game collects EEG data from two channels, FP1 and FP2, operation timestamps, traffic light display timestamps, and information such as whether the answer was correct.

[0117] The cognitive metrics of mental riders are calculated based on three core dimensions: in-window accuracy, reaction speed, and consistency.

[0118] Specifically, accuracy is defined by measuring the proportion of correct responses within a window. Accuracy scores are calculated using the following formula. : , in For the correct number of responses, The number of responses that should be received within the window. Weights are used for accuracy hyperparameters.

[0119] For the reaction speed dimension, the evaluation considers the speed of reaction time, taking into account the optimal reaction time; the faster the reaction, the higher the score. Specifically, for effective reaction time... Calculate the reaction rate fraction using the following formula. : , , in, It is the optimal reaction time hyperparameter. It is the maximum reaction time hyperparameter. For the number of effective reactions, This is the velocity weight hyperparameter.

[0120] The consistency dimension measures the stability of reaction time, avoiding inconsistencies. The consistency score is calculated using the following formula. : , in, The standard deviation of the reaction time. The average reaction time. Weights are the consistency hyperparameters. Then, the accuracy, reaction speed, and consistency within the window are weighted and summed to calculate the total score. : , Next, the total score is smoothed using the sigmoid function, and the cognitive index score is calculated: , in, Here, k is the cognitive index score, k is the sigmoid slope parameter, and m is the sigmoid midpoint parameter.

[0121] The calculation of the cognitive index score is further explained below. In some embodiments, the cognitive index score is calculated based on the real-time behavioral data, including: Using the second preset formula, a cognitive index score is calculated based on the real-time behavioral data; The second preset formula includes:

[0122]

[0123]

[0124]

[0125]

[0126] ;

[0127] in, To score the accuracy of mathematical tasks, For the weight hyperparameters of the mathematical task, The penalty coefficient hyperparameter, For the correct number of mathematical tasks, This represents the total number of mathematical tasks. Score the accuracy of key presses. This is a hyperparameter for the weight of key presses. The penalty coefficient hyperparameter, The correct number of key presses. This represents the total number of key presses. For effective reaction time, For the overall speed score, The optimal reaction time hyperparameter is... The maximum effective reaction time hyperparameter, For velocity weighting hyperparameters, For effective reaction quantity, This represents the total amount of reaction required. For consistency score, Standard deviation, The mean, For consistency weight hyperparameter, For the number of valid samples, For the overall score, Here, k is the cognitive index score, k is the sigmoid slope parameter, and m is the sigmoid midpoint parameter.

[0128] This embodiment uses the Brain Rider game as an example to illustrate a formal reaction ability test. The game consists of three rounds, each with 10 questions. The game simultaneously displays calculation questions on the left and shape questions on the right. For calculation questions, players must judge whether the answer is correct or not. For shape questions, if the shape is not a red triangle, a button must be pressed; otherwise, no action is required. The top displays the real-time reaction ability value, which controls the maximum reaction time for the current question. During the game, the FP1 and FP2 EEG data, operation timestamps, question display timestamps, and information such as whether the answer is correct are recorded in real time.

[0129] The cognitive metrics for parallel challenges are calculated based on three core dimensions: mathematical tasks, key press tasks, reaction speed, and consistency.

[0130] Cognitive computational responsiveness is assessed through computational tasks, and the accuracy score for mathematical tasks is calculated using the following formula. : , in, To score the accuracy of mathematical tasks, For the weight hyperparameters of the mathematical task, The penalty coefficient hyperparameter, For the correct number of mathematical tasks, This represents the total number of mathematical tasks.

[0131] Response accuracy is assessed through key press tasks, and the key press task accuracy score is calculated using the following formula. : , in, Score the accuracy of key presses. This is a hyperparameter for the weight of key presses. The penalty coefficient hyperparameter, The correct number of key presses. This represents the total number of button presses.

[0132] The ability to make rapid decisions was assessed using the reaction speed dimension, for each effective reaction time. The overall speed fraction is calculated using the following formula. : , , in, The optimal reaction time hyperparameter is... The maximum effective reaction time hyperparameter, For velocity weighting hyperparameters, For effective reaction quantity, This represents the total amount of reaction required.

[0133] Response stability is assessed using the consistency dimension, and the consistency score is calculated using the following formula. : , in, Standard deviation, The mean, For consistency weight hyperparameter, This represents the number of valid samples.

[0134] The weighted sum of the math task, key press task, speed score, and consistency score is used to calculate the overall score. : .

[0135] Smoothing the overall score using the sigmoid function yields the cognitive index score B: , Where k is the sigmoid slope parameter and m is the sigmoid midpoint parameter.

[0136] In one specific embodiment, the experiment was set up with two participants engaging in a mental rider and parallel challenge game, and the corresponding experimental results are as follows: Figures 2-5 As shown. Figure 2 This is a schematic diagram illustrating the reaction recognition results of Subject 1 in the Brain Rider game. Figure 3 This is a schematic diagram illustrating the reaction recognition results of Subject 2 in the Brain Rider game. Figure 4 This is a schematic diagram illustrating the reaction speed recognition results of Subject 1 in a parallel challenge game. Figure 5 This diagram illustrates the reaction time recognition results for Subject 2 in a parallel challenge game. Different shaded areas represent different experimental conditions. As can be seen from the diagram, although the algorithm exhibits some system latency introduced by smoothing, its overall accuracy is high, with accurate scores under both reaction and non-reaction conditions. It also demonstrates high real-time performance, providing real-time feedback on the player's reaction time. Furthermore, due to the incorporation of physiological signals such as EEG, the algorithm is highly robust to different players, showing excellent performance across all players. The average single-run time is 0.6 seconds, and the memory usage is 27Mb, allowing for parallel operation by multiple players, thus possessing certain commercial and application value.

[0137] Experiments have shown that this invention can accurately reflect the fluctuations in a player's reaction time during real-time interaction; it enables instant adjustment, effectively avoiding the lag and rigidity problems of traditional methods; and it enhances the flow experience, improving immersion and interaction quality while ensuring the effectiveness of scientific evaluation.

[0138] Furthermore, this invention supports simultaneous participation by multiple players, enabling personalized closed-loop control of each player's real-time reaction speed, and achieving rhythm coordination and task allocation optimization at the team level. This invention can be widely applied in scenarios such as cognitive decline intervention for the elderly, collaborative education, and group intervention for rehabilitation training, providing broad application prospects and practical value for the healthcare, education, and entertainment industries.

[0139] The reaction recognition device based on game scenes provided by the present invention is described below. The reaction recognition device based on game scenes described below and the reaction recognition method based on game scenes described above can be referred to in correspondence. Figure 6 This is a schematic diagram of the reaction recognition device based on game scenes provided by the present invention, as shown below. Figure 6 As shown, the device includes the following modules: The baseline task unit 610 is used to start the reaction baseline task in response to a user wearing a preset wearable device entering a preset reaction game scene; The baseline data unit 620 is used to acquire the user's baseline EEG data during the user's operation of the reaction baseline task, and to calculate baseline index data based on the baseline EEG data. The game event unit 630 is used to display different game events to the user in response to the completion of the reaction baseline task, based on the historical game event interval time. The data acquisition unit 640 is used to collect the user's real-time EEG data and real-time behavioral data based on the preset wearable device during the process of the user operating the game event according to the preset requirements. The real-time behavioral data includes operation reaction time, number of effective reactions, and number of correct operations. The score calculation unit 650 is used to calculate the brain electrical reaction index score based on the real-time EEG data and the baseline index data, and to calculate the cognitive index score based on the real-time behavioral data. The comprehensive score unit 660 is used to calculate a comprehensive reaction cognitive index score based on the EEG reaction index score and the cognitive index score, as the reaction recognition result; The feedback adjustment unit 670 is used to adjust the game event interval time based on the comprehensive reaction ability cognitive index score.

[0140] According to the present invention, a reaction force recognition device based on a game scene is provided, wherein the reaction force baseline task includes a non-reaction part and a reaction part.

[0141] According to the present invention, a reaction recognition device based on a game scene is provided, wherein the preset wearable device includes a non-invasive EEG acquisition device, and the baseline EEG data and the real-time EEG data are acquired through the acquisition electrodes of the non-invasive EEG acquisition device.

[0142] According to the present invention, a reaction recognition device based on a game scene calculates baseline index data based on the baseline EEG data, including: The baseline EEG data is filtered, and based on the filtering result, the corresponding baseline EEG data of the first preset frequency band and the second preset frequency band are extracted. Based on the corresponding baseline EEG data of the first preset frequency band and the corresponding baseline EEG data of the second preset frequency band, the frequency band energy ratio of the first preset frequency band and the second preset frequency band, as well as the inhibition ratio of the second preset frequency band, are calculated as baseline features. Calculate the average and maximum values ​​of the baseline features as calibration values.

[0143] According to the present invention, a reaction ability recognition device based on a game scene calculates a brainwave reaction ability index score based on the real-time brainwave data and the baseline index data, including: The real-time EEG data and the real-time behavioral data are fuzz-aligned and smoothed using a preset time window. The smoothed real-time EEG data is filtered, and based on the filtering result, the corresponding real-time EEG data of the first preset frequency band and the second preset frequency band are extracted. Based on the real-time EEG data corresponding to the first preset frequency band and the real-time EEG data corresponding to the second preset frequency band, the real-time frequency band energy ratio and the second preset frequency band suppression ratio are calculated as feature indicators. The feature index is centered based on the calibration value, and the result of the centered processing is mapped to a first preset interval to obtain the mapping result; The mapping result is smoothly mapped to a second preset interval to obtain the EEG reaction index score.

[0144] According to the present invention, a reaction recognition device based on a game scene calculates a cognitive index score based on the real-time behavioral data, including: Using the first preset formula, a cognitive index score is calculated based on the real-time behavioral data; The first preset formula includes:

[0145]

[0146]

[0147]

[0148]

[0149]

[0150] in, For accuracy scores, For the correct number of responses, The number of responses that should be received within the window. For the accuracy hyperparameter weights, The reaction rate fraction, Effective reaction time It is the optimal reaction time hyperparameter. It is the maximum reaction time hyperparameter. For the number of effective reactions, For velocity weighting hyperparameters, For consistency score, The standard deviation of the reaction time. The average reaction time. For consistency hyperparameter weights, For the total score, Here, k is the cognitive index score, k is the sigmoid slope parameter, and m is the sigmoid midpoint parameter.

[0151] According to the present invention, a reaction recognition device based on a game scene calculates a cognitive index score based on the real-time behavioral data, including: Using the second preset formula, a cognitive index score is calculated based on the real-time behavioral data; The second preset formula includes:

[0152]

[0153]

[0154]

[0155]

[0156] ;

[0157] in, To score the accuracy of mathematical tasks, For the weight hyperparameters of the mathematical task, The penalty coefficient hyperparameter, For the correct number of mathematical tasks, This represents the total number of mathematical tasks. Score the accuracy of key presses. This is a hyperparameter for the weight of key presses. The penalty coefficient hyperparameter, The correct number of key presses. This represents the total number of key presses. For effective reaction time, For the overall speed score, The optimal reaction time hyperparameter is... The maximum effective reaction time hyperparameter, For velocity weighting hyperparameters, For effective reaction quantity, This represents the total amount of reaction required. For consistency score, Standard deviation, The mean, For consistency weight hyperparameter, For the number of valid samples, For the overall score, Here, k is the cognitive index score, k is the sigmoid slope parameter, and m is the sigmoid midpoint parameter.

[0158] Figure 7 An example is a schematic diagram of the physical structure of an electronic device, such as... Figure 7 As shown, the electronic device may include: a processor 710, a communication interface 720, a memory 730, and a communication bus 740, wherein the processor 710, the communication interface 720, and the memory 730 communicate with each other through the communication bus 740. The processor 710 can call logic instructions in the memory 730 to execute a reaction speed recognition method based on a game scenario. This method includes: in response to a user wearing a preset wearable device entering a preset reaction speed game scenario, starting a reaction speed baseline task; during the user's operation of the reaction speed baseline task, acquiring the user's baseline EEG data and calculating baseline index data based on the baseline EEG data; in response to the completion of the reaction speed baseline task, displaying different game events to the user based on historical game event intervals; during the user's operation of the game events according to preset requirements, collecting the user's real-time EEG data and real-time behavioral data based on the preset wearable device, wherein the real-time behavioral data includes operation reaction time, number of effective reactions, and number of correct operations; calculating an EEG reaction speed index score based on the real-time EEG data and the baseline index data, and calculating a cognitive index score based on the real-time behavioral data; calculating a comprehensive reaction speed cognitive index score based on the EEG reaction speed index score and the cognitive index score, as the reaction speed recognition result; and adjusting the game event interval time based on the comprehensive reaction speed cognitive index score.

[0159] Furthermore, the logical instructions in the aforementioned memory 730 can be implemented as software functional units and, when sold or used as independent products, can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, essentially, or the part that contributes to the prior art, or a part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0160] On the other hand, the present invention also provides a computer program product, the computer program product comprising a computer program, which can be stored on a non-transitory computer-readable storage medium. When the computer program is executed by a processor, the computer is able to execute the reaction recognition method based on the game scene provided by the above methods. The method includes: in response to a user wearing a preset wearable device entering a preset reaction game scene, starting a reaction baseline task; during the user's operation of the reaction baseline task, acquiring the user's baseline EEG data and calculating baseline index data based on the baseline EEG data; in response to the completion of the reaction baseline task, based on the historical game event interval time, sending data to the computer program... The user is presented with different game events; during the user's operation of the game events according to preset requirements, real-time EEG data and real-time behavioral data of the user are collected based on the preset wearable device, wherein the real-time behavioral data includes operation reaction time, number of effective reactions, and number of correct operations; based on the real-time EEG data and the baseline index data, an EEG reaction ability index score is calculated, and based on the real-time behavioral data, a cognitive index score is calculated; based on the EEG reaction ability index score and the cognitive index score, a comprehensive reaction ability cognitive index score is calculated as the reaction ability recognition result; based on the comprehensive reaction ability cognitive index score, the interval time of the game events is adjusted.

[0161] In another aspect, the present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon. When executed by a processor, the computer program implements the reaction time recognition method based on the game scenario provided by the methods described above. The method includes: in response to a user wearing a preset wearable device entering a preset reaction time game scenario, initiating a reaction time baseline task; during the user's operation of the reaction time baseline task, acquiring the user's baseline EEG data and calculating baseline index data based on the baseline EEG data; in response to the completion of the reaction time baseline task, displaying different game events to the user based on historical game event intervals; and in the... During the user's operation of the game events according to preset requirements, real-time EEG data and real-time behavioral data of the user are collected based on the preset wearable device. The real-time behavioral data includes operation reaction time, number of effective reactions, and number of correct operations. Based on the real-time EEG data and the baseline index data, an EEG reaction ability index score is calculated, and based on the real-time behavioral data, a cognitive index score is calculated. Based on the EEG reaction ability index score and the cognitive index score, a comprehensive reaction ability cognitive index score is calculated as the reaction ability recognition result. Based on the comprehensive reaction ability cognitive index score, the game event interval time is adjusted.

[0162] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs. Those skilled in the art can understand and implement this without any creative effort.

[0163] Through the above description of the embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus necessary general-purpose hardware platforms, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solutions, in essence or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods described in the various embodiments or some parts of the embodiments.

[0164] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention 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 of the technical features; and these 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 the present invention.

Claims

1. A reaction time recognition method based on game scenarios, characterized in that, include: In response to a user wearing a pre-set wearable device entering a pre-set reaction game scene, the reaction baseline task begins; During the user's operation of the reaction baseline task, the user's baseline EEG data is acquired, and baseline index data is calculated based on the baseline EEG data. In response to the completion of the reaction baseline task, different game events are displayed to the user based on the historical game event intervals; During the process of the user performing the game event according to the preset requirements, the user's real-time EEG data and real-time behavioral data are collected based on the preset wearable device, wherein the real-time behavioral data includes operation reaction time, number of effective reactions and number of correct operations; Based on the real-time EEG data and the baseline index data, the EEG reaction index score is calculated, and based on the real-time behavioral data, the cognitive index score is calculated. Based on the EEG reaction index score and the cognitive index score, a comprehensive reaction cognitive index score is calculated as the reaction recognition result; The game event interval is adjusted based on the comprehensive reaction ability cognitive index score.

2. The reaction recognition method based on game scenes according to claim 1, characterized in that, The reactive baseline task includes a non-reactive portion and a reactive portion.

3. The reaction recognition method based on game scenes according to claim 1, characterized in that, The preset wearable device includes a non-invasive EEG acquisition device, through which the baseline EEG data and the real-time EEG data are acquired via the acquisition electrodes of the non-invasive EEG acquisition device.

4. The reaction recognition method based on game scenes according to claim 1, characterized in that, Baseline index data are calculated based on the baseline EEG data, including: The baseline EEG data is filtered, and based on the filtering result, the corresponding baseline EEG data of the first preset frequency band and the second preset frequency band are extracted. Based on the corresponding baseline EEG data of the first preset frequency band and the corresponding baseline EEG data of the second preset frequency band, the frequency band energy ratio of the first preset frequency band and the second preset frequency band, as well as the inhibition ratio of the second preset frequency band, are calculated as baseline features. Calculate the average and maximum values ​​of the baseline features as calibration values.

5. The reaction recognition method based on game scenes according to claim 4, characterized in that, Based on the real-time EEG data and the baseline index data, the EEG responsiveness index score is calculated, including: The real-time EEG data and the real-time behavioral data are fuzz-aligned and smoothed using a preset time window. The smoothed real-time EEG data is filtered, and based on the filtering result, the corresponding real-time EEG data of the first preset frequency band and the second preset frequency band are extracted. Based on the real-time EEG data corresponding to the first preset frequency band and the real-time EEG data corresponding to the second preset frequency band, the real-time frequency band energy ratio and the second preset frequency band suppression ratio are calculated as feature indicators. The feature index is centered based on the calibration value, and the result of the centered processing is mapped to a first preset interval to obtain the mapping result; The mapping result is smoothly mapped to a second preset interval to obtain the EEG reaction index score.

6. The reaction recognition method based on game scenes according to claim 1, characterized in that, Based on the real-time behavioral data, a cognitive index score is calculated, including: Using the first preset formula, a cognitive index score is calculated based on the real-time behavioral data; The first preset formula includes: in, For accuracy scores, For the correct number of responses, The number of responses that should be made within the window. For the accuracy hyperparameter weights, The reaction rate fraction, Effective reaction time It is the optimal reaction time hyperparameter. It is the maximum reaction time hyperparameter. For the number of effective reactions, For velocity weighting hyperparameters, For consistency score, The standard deviation of the reaction time. The average reaction time. For consistency hyperparameter weights, For the total score, Here, k represents the cognitive index score, k is the sigmoid slope parameter, and m is the sigmoid midpoint parameter.

7. The reaction recognition method based on game scenes according to claim 1, characterized in that, Based on the real-time behavioral data, a cognitive index score is calculated, including: Using the second preset formula, a cognitive index score is calculated based on the real-time behavioral data; The second preset formula includes: ; in, To score the accuracy of mathematical tasks, For the weight hyperparameters of the mathematical task, The penalty coefficient hyperparameter, For the correct number of mathematical tasks, This represents the total number of mathematical tasks. Score the accuracy of key presses. This is a hyperparameter for the weight of key presses. The penalty coefficient hyperparameter, The correct number of key presses. This represents the total number of key presses. For effective reaction time, The overall speed score, The optimal reaction time hyperparameter is... The maximum effective reaction time hyperparameter, For velocity weighting hyperparameters, For the effective reaction quantity, This represents the total amount of reaction required. For consistency score, Standard deviation, The mean, For consistency weight hyperparameter, For the number of valid samples, For the overall score, Here, k represents the cognitive index score, k is the sigmoid slope parameter, and m is the sigmoid midpoint parameter.

8. A reaction recognition device based on a game scene, characterized in that, include: The baseline task unit is used to respond to the user wearing a preset wearable device entering a preset reaction game scene and start the reaction baseline task; The baseline data unit is used to acquire the user's baseline EEG data during the user's operation of the reaction baseline task, and to calculate baseline index data based on the baseline EEG data. The game event unit is used to display different game events to the user in response to the completion of the reaction baseline task, based on the historical game event interval time. The data acquisition unit is used to collect the user's real-time EEG data and real-time behavioral data based on the preset wearable device during the process of the user operating the game event according to preset requirements. The real-time behavioral data includes operation reaction time, number of effective reactions, and number of correct operations. The score calculation unit is used to calculate the brainwave reaction index score based on the real-time EEG data and the baseline index data, and to calculate the cognitive index score based on the real-time behavioral data. The comprehensive score unit is used to calculate the comprehensive reaction cognitive index score based on the EEG reaction index score and the cognitive index score, as the reaction recognition result; The feedback adjustment unit is used to adjust the game event interval time based on the comprehensive reaction ability cognitive index score.

9. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the program, it implements the reaction recognition method based on the game scene as described in any one of claims 1 to 7.

10. A non-transitory computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the reaction recognition method based on the game scene as described in any one of claims 1 to 7.

Images