Evaluation system and method for brain intelligence cognition development ability of children

Abstract

The invention discloses a children brain intelligence cognition development ability evaluation method and system, and the method comprises the steps: receiving the identity information of a user, including the age, gender and education stage; an evaluation model covering six first-level dimensions and eight second-level dimensions is constructed, the first-level dimensions comprise memory, self-control ability, thinking ability, reaction ability, attention and space ability, and the second-level dimensions comprise short-term memory, work memory, suppression control, task switching, processing speed, logical thinking, concentration and visual perception space; and randomly extracting a target game task set from a task library, wherein the task set covers at least three first-level-dimension evaluation requirements. According to the evaluation method and system for the cognitive competence of the children, in the game evaluation process, the six cognitive fields are comprehensively covered by the system, and a scientific, comprehensive and interesting solution is provided for cognitive development evaluation of the children.

Description

Technical Field

[0001] The present invention belongs to the field of child cognitive development assessment, with the classification number of A61B, and particularly relates to a comprehensive evaluation method and system based on six core cognitive abilities (memory, attention, self-control, reaction, thinking, and spatial ability), aiming to provide an innovative, scientific, and interesting solution for the accurate assessment of children's cognitive abilities. Background Art

[0002] With the increasing attention of society to early childhood education and cognitive development, scientifically evaluating children's cognitive development level has become an important research and application direction. In modern society, people increasingly recognize the importance of early education for children's future learning and life. As the core component of early education, the scientific nature of cognitive development assessment methods and tools is directly related to the effectiveness of educational intervention measures. Therefore, researchers and educators are seeking more accurate and comprehensive assessment means to accurately grasp children's cognitive development status and provide them with personalized educational support.

[0003] Cognitive development involves children's perception, memory, thinking and other aspects of abilities. The development level of these abilities not only affects children's academic performance at school, but also their social skills, emotional regulation and environmental adaptation abilities. Therefore, scientifically evaluating children's cognitive development level not only helps to detect potential learning disabilities or developmental delays at an early stage, but also provides guidance for parents and educators to help them better understand children's needs and formulate appropriate educational plans.

[0004] Currently, there are various methods for evaluating children's cognitive development, including standardized tests, observation records, behavior scales, and neuropsychological assessments. These methods have their own advantages and limitations, and researchers are working on developing more comprehensive and sensitive assessment tools to more comprehensively reflect children's cognitive abilities. At the same time, with the development of technologies such as artificial intelligence and big data analysis, these technologies have also been introduced into the field of children's cognitive assessment, providing new possibilities for the accuracy and personalization of assessment. The invention patent with the publication number of CN111820922B discloses a method for evaluating children's computational thinking. This method integrates the key steps of tasks into an activity map, and children need to compile a program instruction set on a programming board to command a robot to successfully complete specific assessment tasks on the activity map and display the robot's movement trajectory in real time. In the programming practice, children can fully exercise and demonstrate their abstract thinking, spatial orientation cognition, logical thinking, task decomposition and computational abilities. However, this mode is still relatively single, only involving a part of the six core cognitive abilities and lacking comprehensiveness.

[0005] The existing technologies and products on the market currently mainly have the following deficiencies:

[0006] (1) Single - dimension evaluation, lacking comprehensiveness

[0007] Existing evaluation methods mostly focus on single - dimension tests, such as individual evaluations of concentration or memory, and cannot comprehensively reflect the overall level of children in multiple key areas of cognitive development.

[0008] (2) Insufficient interest in traditional testing methods

[0009] Traditional paper - and - pencil tests or boring operation tasks are difficult to fully arouse children's interest, resulting in the evaluation results may be affected by children's inattention or fatigue.

[0010] (3) Lack of intelligence and personalization

[0011] Existing evaluation technologies rarely combine intelligent algorithms and big data analysis for diagnosis and screening, making it difficult to accurately identify abnormal individuals and unable to provide personalized analysis and suggestions according to individual differences.

[0012] (4) Lack of scientific model support

[0013] Many evaluation products lack a theoretical framework based on brain and cognitive development science, and it is difficult to ensure the scientific nature, reliability, and validity of evaluation dimensions and task designs. Summary of the Invention

[0014] The present invention aims to provide a method and system for evaluating children's cognitive abilities based on six core cognitive abilities (memory, attention, self - control, reaction, thinking, and spatial ability), which comprehensively cover six cognitive fields of memory, attention, self - control, reaction, thinking, and spatial ability in the gamified evaluation process, and provide a scientific, comprehensive, and interesting solution for children's cognitive development evaluation.

[0015] The object of the present invention is achieved by the following technical solutions:

[0016] A method for evaluating children's brain intelligence cognitive development ability, comprising:

[0017] Receiving user identity information, including age, gender, and education stage;

[0018] Constructing an evaluation model covering six first - level dimensions and eight second - level dimensions, the first - level dimensions including memory, self - control, thinking, reaction, attention, and spatial ability, and the second - level dimensions including short - term memory, working memory, inhibitory control, task switching, processing speed, logical thinking, concentration, and visual - perceptual space;

[0019] Randomly extract a set of target game tasks from the task library, where the task set covers the assessment requirements of at least three first-level dimensions; during the task execution process, randomly select game tasks from the second-level dimensions for assessment, and collect children's behavioral data during the tasks, including reaction time, accuracy, and strategy selection.

[0020] Construct a norm data table based on the behavioral data of children of different ages, where the norm data table contains the mean and standard deviation of each dimension.

[0021] Calculate the task score, first-level dimension score, and total score based on the behavioral data, and convert the scores into percentiles based on the norm data table, and divide them into six grades (A to F).

[0022] As a preferred method, memory includes two dimensions: short-term memory and working memory; self-control includes inhibitory control and task switching; thinking ability corresponds to logical thinking; reaction ability corresponds to processing speed; attention corresponds to concentration; spatial ability corresponds to visual perception space.

[0023] Based on a pre-trained cognitive ability assessment model, perform multi-dimensional feature fusion on the collected data to generate an ability vector including memory index, self-control index, thinking ability index, reaction ability index, attention index, and spatial ability index.

[0024] Compare and analyze the ability vector with the norm database, and output an assessment report with grade symbol markings; the norm database is a standardized database constructed based on the behavioral data of children of different ages, and is used to measure whether children's performance in each cognitive dimension meets the average level of their peer group; the norm database contains the following content: age segmentation, the norm database is segmented according to the age of children, and each age group corresponds to different norm data; the mean and standard deviation of each dimension: for each first-level dimension (such as memory, self-control, etc.) and / or second-level dimension (such as short-term memory, working memory, etc.), the database stores the mean (average value) and standard deviation of this dimension at a specific age; recommend targeted training tasks according to the grade and score.

[0025] The norm database is a standardized database constructed based on the behavioral data of children of different ages, and is used to measure whether children's performance in each cognitive dimension meets the average level of their peer group. Specifically, the norm database contains the following content: age segmentation: the norm database is segmented according to the age of children, and each age group corresponds to different norm data. The mean and standard deviation of each dimension: for each first-level dimension (such as memory, self-control, etc.) and / or second-level dimension (such as short-term memory, working memory, etc.), the database stores the mean (average value) and standard deviation of this dimension at a specific age.

[0026] As a preferred approach, the training process of the pre-trained cognitive ability assessment model includes: collecting an operation dataset of historical users, where the dataset is labeled with ability level tags; constructing a deep neural network model, with the input layer receiving the coordinate sequence of the operation trajectory, and the hidden layer containing LSTM units for extracting temporal features; using a contrastive loss function for model optimization, and the loss function calculates the difference between the predicted ability level and the expert annotation.

[0027] As a preferred approach, recommending targeted training tasks according to the level and score is completed by the personalized recommendation engine, which is configured to: analyze the abnormal dimensions in the ability level tag; match an intervention plan from the policy knowledge base, where the policy knowledge base stores a list of training activities associated with each level interval; generate an executable plan including gamified training task recommendations and family education guidance.

[0028] As a preferred approach, dynamically adjusting the difficulty of the training task includes: when it is detected that the user's operation accuracy rate is higher than the first threshold for three consecutive times, increasing the task difficulty level, and the increase includes increasing the moving speed of the stimulus or reducing the effective response time window; when it is detected that the user's error rate exceeds the second threshold, activating the auxiliary prompt mechanism, and the mechanism includes highlighting the target area or extending the task time limit.

[0029] As a preferred approach, constructing an age-segmented norm database, where the database stores the average reaction time and standard deviation of users of each age group in a preset task; using the Z algorithm to convert the user's original score, and the calculation formula is:

[0030]

[0031] where X is the user's score, μ is the norm mean of the corresponding age group, σ is the norm standard deviation; Z is the percentile, and an ability level tag is generated.

[0032] As a preferred approach, the construction method of the policy knowledge base involves constructing a comprehensive ability training framework, establishing a dynamic mapping relationship between ability types and intervention measures, including multi-level and hierarchical training programs; this framework covers multiple dimensions such as short-term memory, working memory, inhibitory control, task switching, processing speed, logical thinking, concentration, and visual perception space, and provides personalized training intensity, task complexity, and auxiliary mechanisms for different levels (level A to level F) to effectively improve cognitive function and optimize learning effects.

[0033] The construction method of the policy knowledge base includes: establishing a mapping relationship between ability types and intervention measures, and the mapping relationship table covers all six first-level dimensions and the level intervals of the corresponding second-level dimensions, specifically including:

[0034] Short-term memory: Level A, training intensity: maintain 3 times a week, 15 minutes each time. Task difficulty: increase the complexity of information, such as increasing from 4 digits to 5 digits, and combining color and position. Auxiliary mechanism: no additional assistance, appropriately increase the time limit of the task, for example, require to complete the task within 30 seconds. Level B, training intensity: maintain 5 times a week, 15 minutes each time. Task difficulty: gradually increase the amount of information, such as increasing from 4 digits to 5 digits. Auxiliary mechanism: the correct answer will be displayed when there is an error to help the learner understand the reason for the error. Level C, training intensity: increase the training frequency to 7 times a week, and the single duration remains 15 minutes. Task difficulty: maintain the baseline difficulty, increase the hint, such as increasing the display duration of the memory object to 5 seconds. Auxiliary mechanism: automatically replay the memory content when there is an error to help children review the forgotten information. Level D, training intensity: increase the training frequency to 7 times a week, and the single duration increases to 20 minutes. Task difficulty: maintain the baseline difficulty, but increase the hint, such as increasing the display duration of the memory object to 5 seconds. Auxiliary mechanism: automatically replay the memory content when there is an error to help children review the forgotten information. Level E, training intensity: reduce the training frequency to 5 times a week, and the single duration remains 15 minutes. Task difficulty: reduce the baseline difficulty and the amount of information, for example, reducing from 5 digits to 3 digits. Auxiliary mechanism: increase immediate feedback and hints, and extend the memory time, for example, increasing to 40 seconds. Level F, training intensity: reduce the training frequency to 3 times a week, and the single duration increases to 20 minutes. Task difficulty: greatly reduce the difficulty, only require to remember 2 digits, and extend the memory time, for example, increasing to 60 seconds. Auxiliary mechanism: provide multi-sensory support, such as sound and visual assistance, to ensure understanding of the task requirements.

[0035] Working memory: Level A, training intensity: maintain 3 times a week, 15 minutes each time. Task difficulty: increase complexity, such as adding multi-step working memory tasks, for example, sorting numbers according to rules. Auxiliary mechanism: no additional assistance, appropriately increase the time limit of the task, for example, require the task to be completed within 30 seconds. Level B, training intensity: maintain 5 times a week, 15 minutes each time. Task difficulty: gradually increase the amount of information, such as increasing from 3 numbers to 4 numbers. Auxiliary mechanism: the correct answer will be displayed when there is an error to help the learner understand the reason for the error. Level C, training intensity: increase the training frequency to 7 times a week, with a single duration of 15 minutes. Task difficulty: maintain the baseline difficulty, increase prompts, such as showing task switching signals. Auxiliary mechanism: automatically replay the task instructions when there is an error to help children understand the task requirements. Level D, training intensity: increase the training frequency to 7 times a week, and increase the single duration to 20 minutes. Task difficulty: maintain the baseline difficulty, but increase prompts, such as showing task switching signals. Auxiliary mechanism: automatically replay the task instructions when there is an error to help children understand the task requirements. Level E, training intensity: reduce the training frequency to 5 times a week, with a single duration of 15 minutes. Task difficulty: reduce the baseline difficulty and the amount of information, for example, reducing from 4 numbers to 2 numbers. Auxiliary mechanism: increase immediate feedback and prompts, and extend the task time, for example, increasing to 40 seconds. Level F, training intensity: reduce the training frequency to 3 times a week, and increase the single duration to 20 minutes. Task difficulty: significantly reduce the difficulty, only require to complete a simple working memory task, for example, memorize 1 number. Auxiliary mechanism: provide multi-sensory support, such as sound and visual aids, to ensure understanding of the task requirements.

[0036] Inhibitory control: Level A, Training intensity: Maintain 3 times a week, 15 minutes each time. Task difficulty: Increase complexity, such as increasing the number of interference sources. Auxiliary mechanism: No additional assistance, appropriately increase the time limit of the task, for example, require the task to be completed within 30 seconds. Level B, Training intensity: Maintain 5 times a week, 15 minutes each time. Task difficulty: Gradually increase the amount of information, such as increasing from 3 interference sources to 4 interference sources. Auxiliary mechanism: The correct answer will be displayed when there is an error to help the learner understand the reason for the error. Level C, Training intensity: Increase the training frequency to 7 times a week, with a single session duration of 15 minutes. Task difficulty: Maintain the baseline difficulty, increase prompts, such as highlighting the correct option. Auxiliary mechanism: Automatically extend the response time window when there is an error to help children complete the task. Level D, Training intensity: Increase the training frequency to 7 times a week, with a single session duration increased to 20 minutes. Task difficulty: Maintain the baseline difficulty, but increase prompts, such as highlighting the correct option. Auxiliary mechanism: Automatically extend the response time window when there is an error to help children complete the task. Level E, Training intensity: Reduce the training frequency to 5 times a week, with a single session duration of 15 minutes. Task difficulty: Reduce the baseline difficulty, reduce the amount of information, for example, from 4 interference sources to 2 interference sources. Auxiliary mechanism: Increase immediate feedback and prompts, extend the task time, for example, increase to 40 seconds. Level F, Training intensity: Reduce the training frequency to 3 times a week, with a single session duration increased to 20 minutes. Task difficulty: Greatly reduce the difficulty, only require to complete a simple inhibitory control task, such as selecting the correct option. Auxiliary mechanism: Provide multi-sensory support, such as sound and visual assistance, to ensure understanding of the task requirements.

[0037] Task switching: Level A, Training intensity: Maintain 3 times a week, 15 minutes each time. Task difficulty: Increase complexity, such as adding multitasking tasks. Auxiliary mechanism: No additional assistance, appropriately increase the time limit of the task, for example, require the task to be completed within 30 seconds. Level B, Training intensity: Maintain 5 times a week, 15 minutes each time. Task difficulty: Gradually increase the amount of information, such as increasing from 2 tasks to 3 tasks. Auxiliary mechanism: The correct answer will be displayed when there is an error to help the learner understand the reason for the error. Level C, Training intensity: Increase the training frequency to 7 times a week, with a single duration of 15 minutes. Task difficulty: Maintain the baseline difficulty, increase prompts, such as displaying task switching signals. Auxiliary mechanism: Automatically replay the task instructions when there is an error to help children understand the task requirements. Level D, Training intensity: Increase the training frequency to 7 times a week, and increase the single duration to 20 minutes. Task difficulty: Maintain the baseline difficulty, but increase prompts, such as displaying task switching signals. Auxiliary mechanism: Automatically replay the task instructions when there is an error to help children understand the task requirements. Level E, Training intensity: Reduce the training frequency to 5 times a week, with a single duration of 15 minutes. Task difficulty: Reduce the baseline difficulty and the amount of information, for example, reduce from 3 tasks to 2 tasks. Auxiliary mechanism: Increase immediate feedback and prompts, and extend the task time, for example, increase to 40 seconds. Level F, Training intensity: Reduce the training frequency to 3 times a week, and increase the single duration to 20 minutes. Task difficulty: Greatly reduce the difficulty, only require to complete a simple task switching task, such as switching a single task. Auxiliary mechanism: Provide multi-sensory support, such as sound and visual assistance, to ensure understanding of the task requirements.

[0038] Processing speed: Level A, training intensity: maintain 3 times a week, 15 minutes each time. Task difficulty: increase complexity, such as reducing the stimulus presentation time. Auxiliary mechanism: no additional assistance, appropriately increase the time limit of the task, for example, require the task to be completed within 30 seconds. Level B, training intensity: maintain 5 times a week, 15 minutes each time. Task difficulty: gradually increase the amount of information, such as increasing from 3 tasks to 4 tasks. Auxiliary mechanism: the correct answer will be displayed when there is an error to help the learner understand the reason for the error. Level C, training intensity: increase the training frequency to 7 times a week, and the single session duration remains 15 minutes. Task difficulty: maintain the baseline difficulty, increase prompts, such as showing task-related graphics. Auxiliary mechanism: automatically extend the task presentation time when there is an error to help children complete the task. Level D, training intensity: increase the training frequency to 7 times a week, and the single session duration increases to 20 minutes. Task difficulty: maintain the baseline difficulty, but keep the task presentation time unchanged. Auxiliary mechanism: automatically extend the task presentation time when there is an error to help children complete the task. Level E, training intensity: reduce the training frequency to 5 times a week, and the single session duration remains 15 minutes. Task difficulty: reduce the baseline difficulty and the amount of information, for example, reducing from 4 tasks to 2 tasks. Auxiliary mechanism: increase immediate feedback and prompts, and appropriately extend the task presentation time. Level F, training intensity: reduce the training frequency to 3 times a week, and the single session duration increases to 20 minutes. Task difficulty: greatly reduce the difficulty, only require to complete a simple processing speed task, such as identifying a single stimulus. Auxiliary mechanism: provide multi-sensory support, such as sound and visual assistance, to ensure understanding of the task requirements.

[0039] Logical thinking: Level A, training intensity: maintain 3 times a week, 15 minutes each time. Task difficulty: increase complexity, such as adding multi-step logical reasoning tasks. Auxiliary mechanism: no additional assistance, appropriately increase the time limit for tasks, for example, require completion within 30 seconds. Level B, training intensity: maintain 5 times a week, 15 minutes each time. Task difficulty: gradually increase the amount of information, such as from 2 conditions to 3 conditions. Auxiliary mechanism: show the correct answer when there is an error to help learners understand the reason for the error. Level C, training intensity: increase the training frequency to 7 times a week, with a single session duration of 15 minutes. Task difficulty: maintain the baseline difficulty, increase prompts, such as showing logical relationships. Auxiliary mechanism: automatically replay the logical reasoning process when there is an error to help children understand the reasoning steps. Level D, training intensity: increase the training frequency to 7 times a week, and increase the single session duration to 20 minutes. Task difficulty: maintain the baseline difficulty, but increase prompts, such as showing logical relationships. Auxiliary mechanism: automatically replay the logical reasoning process when there is an error to help children understand the reasoning steps. Level E, training intensity: reduce the training frequency to 5 times a week, with a single session duration of 15 minutes. Task difficulty: reduce the baseline difficulty, reduce the amount of information, for example, from 3 conditions to 2 conditions. Auxiliary mechanism: increase immediate feedback and prompts, extend the task time, for example, increase to 40 seconds. Level F, training intensity: reduce the training frequency to 3 times a week, and increase the single session duration to 20 minutes. Task difficulty: significantly reduce the difficulty, only require completion of a simple logical reasoning task, such as identifying cause and effect relationships. Auxiliary mechanism: provide multi-sensory support, such as sound and visual aids, to ensure understanding of the task requirements.

[0040] Concentration: Level A, Training intensity: Maintain 3 times a week, 15 minutes each time. Task difficulty: Increase complexity, such as increasing the number of interference sources. Auxiliary mechanism: No additional assistance, appropriately increase the time limit of the task, for example, require the task to be completed within 30 seconds. Level B, Training intensity: Maintain 5 times a week, 15 minutes each time. Task difficulty: Gradually increase the amount of information, such as increasing from 3 interference sources to 4 interference sources. Auxiliary mechanism: The correct answer will be displayed when there is an error to help the learner understand the reason for the error. Level C, Training intensity: Increase the training frequency to 7 times a week, with a single duration of 15 minutes. Task difficulty: Maintain the baseline difficulty and increase prompts, such as highlighting the target stimulus. Auxiliary mechanism: Automatically reduce the number of interference sources when there is an error to help children concentrate better. Level D, Training intensity: Increase the training frequency to 7 times a week, and increase the single duration to 20 minutes. Task difficulty: Maintain the baseline difficulty, but increase the complexity of the interference sources. Auxiliary mechanism: Automatically extend the presentation time of the target stimulus when there is an error to help children capture the target. Level E, Training intensity: Reduce the training frequency to 5 times a week, with a single duration of 15 minutes. Task difficulty: Reduce the baseline difficulty and the number and complexity of interference sources. Auxiliary mechanism: Increase immediate feedback and prompts, and extend the task time, for example, increase to 40 seconds. Level F, Training intensity: Reduce the training frequency to 3 times a week, and increase the single duration to 20 minutes. Task difficulty: Greatly reduce the difficulty, only require to complete a simple concentration task, such as identifying a single target stimulus. Auxiliary mechanism: Provide multi-sensory support, such as sound and visual assistance, to ensure understanding of the task requirements.

[0041] Visual Perception Space: Level A, Training Intensity: Maintain 3 times a week, 15 minutes each time. Task Difficulty: Increase complexity, such as adding spatial rotation tasks. Auxiliary Mechanism: No additional assistance, appropriately increase the time limit of the task, for example, require the task to be completed within 30 seconds. Level B, Training Intensity: Maintain 5 times a week, 15 minutes each time. Task Difficulty: Gradually increase the amount of information, such as increasing from simple shapes to complex shapes. Auxiliary Mechanism: The correct answer will be displayed when there is an error to help the learner understand the reason for the error. Level C, Training Intensity: Increase the training frequency to 7 times a week, with the single session duration remaining 15 minutes. Task Difficulty: Maintain the baseline difficulty, increase prompts, such as showing the shape before rotation. Auxiliary Mechanism: Automatically replay the rotation process when there is an error to help children understand spatial relationships. Level D, Training Intensity: Increase the training frequency to 7 times a week, and increase the single session duration to 20 minutes. Task Difficulty: Maintain the baseline difficulty, but increase prompts, such as showing the shape before rotation. Auxiliary Mechanism: Automatically replay the rotation process when there is an error to help children understand spatial relationships. Level E, Training Intensity: Reduce the training frequency to 5 times a week, with the single session duration remaining 15 minutes. Task Difficulty: Reduce the baseline difficulty and reduce the amount of information, for example, reducing from complex shapes to simple shapes. Auxiliary Mechanism: Increase immediate feedback and prompts, and extend the task time, for example, increase to 40 seconds. Level F, Training Intensity: Reduce the training frequency to 3 times a week, and increase the single session duration to 20 minutes. Task Difficulty: Greatly reduce the difficulty, only require to complete a simple visual perception space task, such as recognizing basic shapes. Auxiliary Mechanism: Provide multi-sensory support, such as sound and visual aids, to ensure understanding of the task requirements. The method for constructing the strategy knowledge base provides personalized learning strategies for promoting the cognitive development of learners by designing graded training programs for different cognitive abilities. Its design adopts a systematic grading method, from low level to high level, gradually increasing the training intensity and task difficulty, and providing auxiliary mechanisms when necessary. As learners progress, the training program is dynamically adjusted to ensure that each learner can receive challenges and support at an appropriate difficulty level. It covers multiple cognitive dimensions to ensure comprehensive improvement, and helps learners who encounter difficulties through auxiliary mechanisms. The strategy knowledge base provides a structured and dynamic solution for personalized learning and has the potential to play an effective role in cognitive training.

[0042] As a preferred method, the collected data includes the spatio-temporal characteristics of the touch operation trajectory, the task completion progress and the number of interruptions, and the frequency statistics of incorrect triggering of interference items. The multi-dimensional feature fusion step specifically includes:

[0043] Perform wavelet transform on the spatio-temporal characteristics to extract the frequency domain characteristics of the operation trajectory;

[0044] Quantify the task completion progress into a time efficiency coefficient, and the calculation formula is:

[0045]

[0046] Where T 基准 is the norm average completion time, and T 实际 is the actual time consumed by the user; weighted fusion of multi-source features is performed to generate a comprehensive ability score.

[0047] The data acquisition and multi-dimensional feature fusion method is a system for comprehensively evaluating user capabilities. By collecting data and extracting features from different dimensions, a score reflecting the user's comprehensive capabilities is generated. The following are the specific steps and methods: Data acquisition: Spatiotemporal features of touch operation trajectories: Record all operation paths and time information when the user touches the screen. This includes the user's gestures, sliding speed, acceleration, etc. This can reflect the user's operation proficiency, coordination ability, and reaction speed. Task completion progress and number of interruptions: Monitor the progress of the user in completing a specific task, including the gap between the time required to complete the task and the expected time, and the number of interruptions during the task. This can evaluate the user's task processing efficiency and concentration. Frequency statistics of accidental triggering of interference items: Record the frequency of the user triggering options or functions unrelated to the current task during the operation. This can reflect the user's concentration and accuracy in a complex task environment. Feature extraction and processing: Perform wavelet transform on spatiotemporal features: Convert the collected spatiotemporal feature data from the time domain to the frequency domain. Wavelet transform can effectively capture local features in the signal, helping to analyze the rhythm and pattern in the user's operation. For example, the frequency features of different gestures such as fast sliding, tapping, and long pressing in the user's operation can be identified. Quantification of task completion progress: Convert the task completion progress into a time efficiency coefficient. This coefficient can quantify the efficiency of the user in completing the task, and the larger the value, the faster the task is completed and the higher the efficiency. Multi-dimensional feature fusion: Adopt an attention mechanism to dynamically assign weights to data in each dimension according to the importance of different features. The attention mechanism can learn which features are more critical for evaluating user capabilities, and thus give higher weights during the fusion process. This step can help the system better focus on important information and reduce interference from irrelevant information.

[0048] Generate a comprehensive ability score: Generate a comprehensive score by weighted fusion of the processed multi-dimensional data. This score can comprehensively reflect the user's comprehensive capabilities in terms of operation proficiency, task efficiency, concentration, etc., and provide a comprehensive evaluation result.

[0049] Through the above steps, the data acquisition and multi-dimensional feature fusion method can be effectively combined to form a highly comprehensive and intelligent user ability evaluation system. This not only helps to understand and evaluate the user's operation ability, but also can provide targeted improvement suggestions for the user, enhancing the user experience and operation efficiency.

[0050] As a preferred method, it also includes a data security module configured to:

[0051] Desensitize the user identity information to generate an anonymized unique identifier;

[0052] Use homomorphic encryption technology to encrypt and transmit the operation track data;

[0053] Establish a two-way verification channel between the local storage device and the cloud server.

[0054] A children's brain intelligence cognitive development ability evaluation system adopts the above-mentioned children's brain intelligence cognitive development ability evaluation method; including:

[0055] A user interface module, used to display game tasks and feedback reports, and support touch interaction;

[0056] A task evaluation module, used to execute game tasks and record behavior data, and the tasks are randomly retrieved according to the secondary dimension;

[0057] A data analysis module, used to calculate percentiles and levels according to the norm data table and generate multi-dimensional scores;

[0058] A report generation module, used to integrate radar charts, level marks and suggestions to generate a visual report;

[0059] A background management module, used to store user data, configure system parameters and manage the training task library.

[0060] The present invention has at least the following beneficial effects: It proposes an evaluation method and system for children's cognitive abilities. The system focuses on six core cognitive abilities, including memory, attention, self-control, reaction, thinking and space. In the gamified evaluation process, the system comprehensively covers these six cognitive fields, aiming to provide a scientific, comprehensive and interesting solution for the evaluation of children's cognitive development.

[0061] The present invention receives user identity information, including age, gender and education stage, to provide personalized basic data for subsequent evaluations. The model design can comprehensively and meticulously evaluate the brain intelligence cognitive development level of children. During the task execution process, game tasks are randomly selected from the secondary dimension for evaluation to collect children's behavior data during the tasks, including reaction time, accuracy and strategy selection, etc. This random selection and task design method not only ensures the comprehensiveness of the evaluation, but also increases the fun and interactivity of the evaluation, enabling children to participate more actively and proactively during the evaluation process.

[0062] Construct a norm data table based on the behavioral data of children of different age groups. This norm data table contains the mean and standard deviation of each dimension, providing a scientific basis for subsequent score conversion and grade division. Calculate the task score, primary dimension score, and total score based on the collected behavioral data, and convert the scores into percentiles based on the norm data table, thereby dividing into six grades. Such a scoring and grading method can intuitively and accurately reflect the brain and cognitive development level of children, providing a strong reference basis for parents and educators. BRIEF DESCRIPTION OF THE DRAWINGS

[0063] In order to reveal the technical details of the embodiments of the present invention, the drawings involved in the embodiments will be briefly introduced below. It should be emphasized that these drawings only present several embodiments of the present invention and should not be regarded as a definition of the scope of the invention. For those of ordinary skill in the art, other related drawings can still be derived based on these drawings without creative labor.

[0064] Figure 1 It is a schematic diagram of the evaluation structure;

[0065] Figure 2 It is a schematic diagram for constructing the norm data table;

[0066] Figure 3 It is an example diagram of the task content in the embodiment;

[0067] Figure 4 It is a heptagon radar chart. SPECIFIC EMBODIMENTS

[0068] The technical solution of the present invention will be further described in detail below with reference to the drawings, but the protection scope of the present invention is not limited to the following.

[0069] In the following content, the embodiments of the present disclosure will be described in detail with the help of the drawings. However, it should be clear that the present disclosure is not limited to the specific forms shown here. On the contrary, it should be understood to cover various variations, equivalent forms, and / or alternative solutions of the embodiments of the present disclosure. In the process of explaining the drawings, the same reference numerals will be used to label similar components.

[0070] In each embodiment of the present disclosure, expressions such as "first", "second", "the first", or "the second" are used to modify different components, rather than indicating order and / or importance, nor imposing restrictions on the corresponding components. For example, the first user device and the second user device represent different user devices, although they both belong to the category of user devices. Similarly, the first component can be named the second component, and the second component can also be named the first component, which does not change their essential attributes within the scope of the present disclosure.

[0071] In this disclosure, terms are used to describe specific embodiments and do not constitute a limitation on this disclosure. In this context, the use of the singular form also covers the plural form unless otherwise clearly stated in the text. In the description process, it should be understood that terms such as "including" or "having" are intended to indicate the existence of features, quantities, steps, operations, structural components, parts, or combinations thereof, and do not preclude the possibility or addition of one or more other features, quantities, steps, operations, structural components, parts, or combinations thereof in advance.

[0072] It should be clear that although detailed specific details are provided in the following description to help comprehensively understand the exemplary embodiments, those skilled in the art should know that the exemplary embodiments can still be implemented even without these specific details. For example, the system can be presented in the form of a block diagram to avoid excessive details interfering with the clarity of the example. In other cases, in order to maintain the clarity of the example, unnecessary details of those well-known processes, structures, and technologies may be omitted.

[0073] A method for evaluating children's brain intelligence cognitive development ability includes:

[0074] Receiving user identity information, including age, gender, and education stage;

[0075] Constructing an evaluation model covering six first-level dimensions and eight second-level dimensions. The first-level dimensions include memory, self-control, thinking ability, reaction ability, attention, and spatial ability, and the second-level dimensions include short-term memory, working memory, inhibitory control, task switching, processing speed, logical thinking, concentration, and visual perception space;

[0076] Randomly extracting a target game task set from the task library, where the task set covers the evaluation requirements of at least three first-level dimensions; during the task execution process, randomly select game tasks from the second-level dimensions for evaluation, and collect the children's behavioral data during the task, including reaction time, accuracy rate, and strategy selection;

[0077] Constructing a norm data table based on the behavioral data of children of different ages, where the norm data table contains the mean and standard deviation of each dimension;

[0078] Calculate the task score, the first-level dimension score, and the total score based on the behavioral data, convert the scores into percentiles according to the norm data table, divide them into six levels (A to F), and generate a visualization report including a radar chart, level markers, and personalized suggestions. The beneficial effects of this evaluation method are mainly reflected in the following aspects: First, through a gamified evaluation method, it improves children's participation and interest; second, through a comprehensive evaluation model and detailed dimension division, it can accurately and comprehensively evaluate children's brain intelligence cognitive development level; third, by constructing a norm data table and performing score conversion and level division, it provides intuitive and scientific evaluation results and reference bases for parents and educators. These beneficial effects jointly promote the improvement of children's brain intelligence cognitive development ability and the scientific development of education.

[0079] In one embodiment:

[0080] Memory includes two dimensions: short-term memory and working memory; self-control includes inhibitory control and task switching; thinking ability corresponds to logical thinking; reaction ability corresponds to processing speed; attention corresponds to concentration; spatial ability corresponds to visual perception space;

[0081] Based on a pre-trained cognitive ability evaluation model, perform multi-dimensional feature fusion on the collected data to generate an ability vector including a memory index, a self-control index, a thinking ability index, a reaction ability index, an attention index, and a spatial ability index;

[0082] Compare and analyze the ability vector with the norm database, and output an evaluation report containing level symbol markers; the norm database is a standardized database constructed based on the behavioral data of children of different ages, and is used to measure whether children's performance in each cognitive dimension conforms to the average level of the same-age group; the norm database includes the following contents: age segmentation, the norm database is segmented according to the age of children, and each age group corresponds to different norm data; the mean and standard deviation of each dimension: for each first-level dimension (such as memory, self-control, etc.) and second-level dimension (such as short-term memory, working memory, etc.), the database stores the mean (average value) and standard deviation of this dimension at a specific age; recommend targeted training tasks according to the level and score, and the training tasks dynamically adjust the difficulty and are designed based on psychological experimental paradigms.

[0083] Ability vector: Through children's behavioral data (such as reaction time, accuracy rate, strategy selection, etc.), combined with a pre-trained cognitive ability evaluation model, generate an ability vector including a memory index, a self-control index, a thinking ability index, etc. Each index represents a child's performance in a certain dimension. Norm database: According to the age of children, query the mean (μ) and standard deviation (σ) data corresponding to the age group in the norm database.

[0084] The norm database is a standardized database constructed based on the behavioral data of children of different ages, and is used to measure whether the performance of children in various cognitive dimensions conforms to the average level of their peer groups. Specifically, the norm database includes the following: Age segmentation: The norm database is segmented according to the age of children, and different norm data correspond to each age group. Mean and standard deviation of each dimension: For each primary dimension (such as memory, self-control, etc.) and secondary dimension (such as short-term memory, working memory, etc.), the database stores the mean (average value) and standard deviation of this dimension at a specific age.

[0085] Behavioral data indicators: The behavioral data in the database includes reaction time (i.e., the time to complete a task), accuracy rate (the accuracy of task completion), strategy selection (the type of strategy used in the task), etc. Through such a structure, a standardized reference can be provided for children of different age groups to evaluate their cognitive abilities in specific dimensions.

[0086] In one embodiment, the training process of the pre-trained cognitive ability evaluation model includes: collecting an operation data set of historical users, where the data set is labeled with ability level labels evaluated by experts; constructing a deep neural network model, the input layer receives the coordinate sequence of the operation trajectory, and the hidden layer contains LSTM units for extracting temporal features; using a contrast loss function for model optimization, and the loss function calculates the difference degree between the predicted ability level and the expert annotation.

[0087] Specifically, in the data collection stage, collect historical user operation data: Through an application or experimental device, record the operation trajectory of the user when completing a specific task. These operation data include the touch coordinate sequence (such as the position and timestamp of gestures, swipes, clicks) and the task completion situation (such as completion time, number of interruptions, etc.). Data storage: Store the collected operation data in an appropriate format (such as CSV, JSON) for subsequent processing. Data annotation: Expert evaluation annotation: Invite domain experts (such as psychologists or education experts) to evaluate the operation data of users and assign an ability level label to each user. These labels can be categorical (such as A, B, C, etc.). Annotation consistency check: Ensure the consistency of evaluations by different experts, and the annotation quality can be verified by statistical indicators of the evaluation consistency between experts (such as Cohen's Kappa coefficient). Data preprocessing: Data cleaning: Remove noise data, outliers or missing values to ensure data quality. For example, check whether the coordinates in the operation trajectory exceed the screen range, or whether the task completion time is within a reasonable range. Data normalization / standardization: Normalize the data such as the coordinates of the operation trajectory to ensure the comparability of data at different scales in the model. Data segmentation: Divide the data set into a training set, a validation set and a test set according to a certain proportion, usually 70%, 15%, 15%.

[0088] Model Design: Input Layer: Design the input layer of the model to receive the coordinate sequence of the operation trajectory. For example, the input can be a sequence of two-dimensional coordinates (x, y) over time, with a shape of (number of time steps, 2). Hidden Layer: Use Long Short-Term Memory (LSTM) units to process temporal features because LSTM can effectively capture long-term dependencies in sequence data. The hidden layer can include multiple LSTM layers or be combined with other types of neural network layers (such as fully connected layers). Output Layer: Design the output layer and select an appropriate activation function according to the task type. For example, for a classification task, the Softmax activation function can be used to output the probability distribution of each ability level; for a regression task, a linear activation function can be used to output the predicted ability score.

[0089] Model Training: Define the Loss Function: Adopt a contrastive loss function such as cross-entropy loss (for classification) or mean squared error (for regression) to measure the difference between the predicted value and the true label. Select the Optimizer: Choose an appropriate optimization algorithm such as the Adam optimizer to adjust the model parameters to minimize the loss function. Training Process: Input the training data into the model, perform forward propagation to calculate the predicted output. Calculate the loss value between the predicted output and the true label. According to the loss value, perform backpropagation to adjust the model parameters. Repeat the above steps until the performance of the model on the validation set reaches the expectation or the training reaches the set number of times.

[0090] Model Evaluation: Validation Set Evaluation: During the training process, regularly use the validation set to evaluate the model performance, such as accuracy, F1 score, mean squared error, etc., to monitor overfitting. Test Set Evaluation: After completing the training, use an independent test set to conduct a final evaluation of the model to obtain the performance metrics of the model on unseen data.

[0091] Model Optimization and Hyperparameter Tuning: Hyperparameter Tuning: Adjust the hyperparameters of the model (such as the number of LSTM layers, the number of units, learning rate, batch size, etc.) through methods such as grid search, random search, or Bayesian optimization to obtain the best performance. Model Architecture Adjustment: According to the evaluation results and performance bottlenecks, adjust the structure of the model, such as increasing or decreasing the number of layers, changing the type of activation function, etc. Model Deployment: Save the Trained Model: Save the trained model parameters for loading and use in actual applications. Integrate into the Application: Integrate the trained model into the relevant application system for real-time user cognitive ability assessment.

[0092] Through the above steps, a pre-trained cognitive ability assessment model can be implemented. This model can predict the cognitive ability level of users based on their operation trajectories, providing an objective user ability assessment for the education field. In one embodiment, the personalized recommendation engine is configured to recommend targeted training tasks according to the level and score. The personalized recommendation engine is configured to: analyze the abnormal dimensions in the ability level marker; match the intervention plan from the policy knowledge base, where the policy knowledge base stores a list of training activities associated with each level interval; generate an executable plan including gamified training task recommendations and family education guidance.

[0093] In one embodiment, the dynamic adjustment of the difficulty of the training task includes: when it is detected that the user's operation accuracy rate is higher than the first threshold for three consecutive times, the task difficulty level is increased, and the increase includes increasing the moving speed of the stimulus or reducing the effective response time window; when it is detected that the user's error rate exceeds the second threshold, the auxiliary prompt mechanism is activated, and the mechanism includes highlighting the target area or extending the task time limit.

[0094] In one embodiment, an age-segmented norm database is constructed, and the database stores the average reaction time and standard deviation of users of each age group in a preset task; the Z algorithm is used to transform the user's original score, and the calculation formula is:

[0095]

[0096] where X is the user's score, μ is the norm mean of the corresponding age group, σ is the norm standard deviation; Z is the percentile (usually multiplied by 100), and the ability level marker is generated.

[0097] In one embodiment, the construction method of the policy knowledge base involves constructing a comprehensive ability training framework, establishing a dynamic mapping relationship between ability types and intervention measures, including multi-level and hierarchical training programs; this framework covers multiple dimensions such as short-term memory, working memory, inhibitory control, task switching, processing speed, logical thinking, concentration, and visual perception space, providing personalized training intensity, task complexity, and auxiliary mechanisms for different levels (level A to level F) to effectively improve cognitive function and optimize learning effects.

[0098] In one embodiment, the collected data includes the spatio-temporal characteristics of the touch operation trajectory, the task completion progress and the number of interruptions, and the frequency statistics of the accidental triggering of interference items. The multi-dimensional feature fusion steps specifically include:

[0099] Perform wavelet transform on the spatio-temporal characteristics to extract the frequency domain characteristics of the operation trajectory;

[0100] Quantify the task completion progress into a time efficiency coefficient, and the calculation formula is:

[0101]

[0102] where T 基准 is the norm average completion time, and T 实际 is the actual time consumed by the user; the attention mechanism is used to perform weighted fusion on multi-source features to generate a comprehensive ability score.

[0103] The data collection and multi-dimensional feature fusion method is a system for comprehensively evaluating user capabilities. Through data collection and feature extraction in different dimensions, a score reflecting the user's comprehensive capabilities is generated. The following are the specific steps and methods: Data collection: Spatiotemporal features of touch operation trajectories: Record all operation paths and time information when the user touches the screen. This includes the user's gestures, sliding speed, acceleration, etc. This can reflect the user's operation proficiency, coordination ability, and reaction speed. Task completion progress and number of interruptions: Monitor the progress of the user in completing a specific task, including the gap between the time required to complete the task and the expected time, and the number of interruptions during the task. This can evaluate the user's task processing efficiency and concentration. Frequency statistics of accidental triggering of interference items: Record the frequency of the user triggering options or functions unrelated to the current task during the operation. This can reflect the user's concentration and accuracy in a complex task environment. Feature extraction and processing: Perform wavelet transform on spatiotemporal features: Convert the collected spatiotemporal feature data from the time domain to the frequency domain. Wavelet transform can effectively capture the local features in the signal and help analyze the rhythm and pattern in the user's operation. For example, the frequency features of different gestures such as fast sliding, tapping, and long pressing in the user's operation can be identified. Quantification of task completion progress: Convert the task completion progress into a time efficiency coefficient. This coefficient can quantify the user's task completion efficiency, and the larger the value, the faster the task is completed and the higher the efficiency. Multi-dimensional feature fusion: Use the attention mechanism to dynamically assign weights to the data in each dimension according to the importance of different features. The attention mechanism can learn which features are more critical for evaluating user capabilities, and thus give higher weights during the fusion process. This step can help the system better focus on important information and reduce the interference of irrelevant information.

[0104] Generate a comprehensive ability score: Generate a comprehensive score by performing weighted fusion on the processed multi-dimensional data. This score can comprehensively reflect the user's comprehensive capabilities in terms of operation proficiency, task efficiency, concentration, etc., and provide a comprehensive evaluation result.

[0105] Through the above steps, the data collection and multi-dimensional feature fusion method can be effectively combined to form a highly comprehensive and intelligent user ability evaluation system. This not only helps to understand and evaluate the user's operation ability, but also can provide targeted improvement suggestions for the user, improving the user experience and operation efficiency.

[0106] In one embodiment, it further includes a data security processing method, configured as follows:

[0107] Perform desensitization processing on user identity information to generate an anonymized unique identifier;

[0108] Use homomorphic encryption technology to encrypt and transmit operation trajectory data;

[0109] Establish a two-way verification channel between the local storage device and the cloud server.

[0110] A children's brain intelligence cognitive development ability evaluation system adopts the above-mentioned children's brain intelligence cognitive development ability evaluation method; it includes:

[0111] A user interface module, used to display game tasks and feedback reports, and support touch interaction;

[0112] A task evaluation module, used to execute game tasks and record behavior data, and the tasks are randomly retrieved according to the secondary dimension;

[0113] A data analysis module, used to calculate percentiles and levels according to the norm data table, and generate multi-dimensional scores;

[0114] A report generation module, used to integrate radar charts, level marks and suggestions to generate a visual report;

[0115] A background management module, used to store user data, configure system parameters and manage the training task library. In one embodiment, the construction of the norm data table includes: collecting the reaction time and accuracy of children of different ages from 3 to 12 years old in specified game tasks; calculating the mean and standard deviation of each secondary dimension by age stratification; associating the mean and standard deviation with the primary dimension to generate an age-dimension comparison table.

[0116] The specific level division includes:

[0117] Level A corresponds to a percentile ≥ 76, marked as ↑↑↑↑; Level B corresponds to 51 - 75, marked as ↑; Level C corresponds to 26 - 50, marked as —; Level D corresponds to 11 - 25, marked as ↓; Level E corresponds to 2 - 10, marked as ↓↓; Level F corresponds to ≤ 1, marked as ↓↓↓.

[0118] The design of the game tasks is based on psychological experimental paradigms, including:

[0119] The inhibitory control task is adapted based on the Flanker task or the Go / No-Go paradigm, the short-term memory task is designed based on spatial memory reproduction, and the logical thinking task is realized through path planning and problem solving;

[0120] The task adopts a cartoonized interface, touch screen interaction, and dynamic difficulty adjustment, where the dynamic difficulty adjustment includes increasing the task complexity or adjusting the operation speed in real time according to the child's performance.

[0121] The generation of the visual report includes: generating a heptagon radar chart based on the first-level dimension scores, and marking the grade marks of each dimension; outputting the development grade and the comparison description of the national level according to the total score; providing personalized training suggestions for the low-score dimensions, and the suggestions include specific activity strategies and recommended training tasks.

[0122] For the children's brain intelligence cognitive development ability evaluation system, the task evaluation module includes a dynamic difficulty adjustment unit, and the specific implementation method is: in the logical thinking task, increase or decrease the number of obstacles according to the correct rate of the child's path planning; in the processing speed task, shorten or extend the stimulus presentation interval according to the reaction time; in the inhibitory control task, increase or decrease the density of interference stimuli based on the number of consecutive correct times.

[0123] The data analysis module also includes an anomaly detection unit, which is used for: when the reaction time of the child in the task exceeds 3 times the standard deviation of the norm data table, triggering a data review mechanism; when the number of uncompleted tasks exceeds the threshold, automatically adjusting the task difficulty or reallocating the evaluation task combination.

[0124] The report generation module uses the following steps to generate personalized suggestions: extract the dimensions with the total score lower than the preset threshold as the key improvement items; associate the game tasks corresponding to the dimensions in the training task library to generate a recommended task list; combine the child's age and the norm data to provide a family activity plan and educational strategies.

[0125] The background management module supports: customizing the evaluation task combination, configuring the number of tasks according to the weight of the first-level dimension; exporting the age-stratified statistical results of the norm data table; batch generating group screening reports, marking the list of children with abnormal dimensions and the intervention priorities.

[0126] In one embodiment:

[0127] (1) Evaluation model construction

[0128] A. Evaluation structure

[0129] The evaluation tasks cover eight secondary dimensions (short-term memory, working memory, inhibitory control, task switching, processing speed, logical thinking, concentration, visual perception space), which belong to six primary dimensions (memory, self-control, thinking ability, reaction ability, attention, spatial ability). Each evaluation randomly selects game tasks from the specified secondary dimensions to ensure the diversity and comprehensiveness of the evaluation. For the evaluation structure, see Figure 1 .

[0130] B. Norm data establishment

[0131] Collect key indicators (such as reaction time, accuracy) of children of different ages during the tasks, calculate the average value and standard deviation of each dimension, and construct a norm data table (see Figure 2 ).

[0132] C. Method for grade classification

[0133] a. Score calculation

[0134] Task score: After each game task is completed, record the score according to the performance. If the task is not completed, the default score is 50 points.

[0135] Score of the first-level dimension: Average the scores of all tasks in the same first-level dimension and round to the nearest integer. Total score: Average the scores of all first-level dimensions and round to the nearest integer.

[0136] b. Percentile calculation and conversion

[0137] Based on the norm table, convert the task score into a percentile to represent the relative position of the child in the norm group.

[0138] c. Specific percentile grade classification

[0139] Classify into six grades according to the percentile:

[0140] A (excellent): Percentile ≥ 76;

[0141] B (safe - good): Percentile 51 to 75;

[0142] C (safe - basically normal): Percentile 26 to 50; D (deviation): Percentile 11 to 25;

[0143] E (crisis - poor): Percentile 2 to 10;

[0144] F (crisis - lagging): Percentile ≤ 1.

[0145] d. Visual marking

[0146] To visually present the grade of each ability, the following marking symbols are used for grades in the report:

[0147] A: ↑↑↑↑

[0148] B: ↑

[0149] C: —

[0150] D: ↓

[0151] E: ↓↓

[0152] F: ↓↓↓

[0153] (2) Scientific basis for game task design

[0154] The game tasks are designed based on psychological experimental paradigms and cognitive development theories, combined with the scientific assessment requirements of six core cognitive abilities, ensuring the scientific nature and interest of the tasks. Specifically:

[0155] A. Scientific design principles

[0156] The tasks involve following psychological experimental paradigms or theoretical frameworks. For example:

[0157] a. Inhibitory control: "Swimming Around" is adapted based on the Flanker task, requiring children to ignore interfering information and focus on the judgment of target stimuli; "Star Lighting" incorporates the Go / No-Go paradigm to exercise children's rapid response and self-inhibition abilities to specific targets.

[0158] b. Short-term memory: "Icy Memory" is designed based on spatial memory tasks, requiring children to remember the positions of specific objects and reproduce them; "Clever Rabbit" simulates real-life scenarios and strengthens the application of short-term memory by remembering the fruits ordered by guests.

[0159] c. Logical thinking: "Swamp Adventure" and "Breaking Through the Encirclement" exercise children's reasoning and spatial planning abilities through route planning and problem-solving respectively.

[0160] d. Processing speed: "Balanced Nutrition" and "Ice World" design complex stimuli to evaluate children's ability to quickly identify and respond in visual information. See the following for examples of task content Figure 3 .

[0161] B. Interest and interactivity

[0162] a. Cartoon design

[0163] The tasks attract children's attention and stimulate their interest through vivid colors and dynamic effects.

[0164] b. Simple touch screen operation

[0165] The tasks enable children to easily complete the tasks and experience the fun of interaction through interactive methods such as swiping, clicking, and dragging.

[0166] c. Dynamic difficulty adjustment

[0167] The difficulty of the tasks is adjusted in real time according to children's performance. For example, the types of accessories and the conveyor belt speed in "Rainbow Cake" gradually increase with performance, ensuring that the tasks adapt to different ability levels.

[0168] The feedback mechanism of the tasks ensures that children receive timely incentives. For example, after completing "The Ultimate Goal", the correct route is displayed to strengthen memory.

[0169] 3. Application scenarios

[0170] (1) Group screening: This system can be used to screen the cognitive abilities of a large-scale student group and identify students who need additional support or intervention.

[0171] (2) Early intervention and personalized education: It can help identify early cognitive development problems, provide personalized educational intervention programs for children, and promote their all-round development.

[0172] (3) Family education support: Provide scientific cognitive development reports for parents to help them understand their children's strengths and weaknesses and optimize family education strategies.

[0173] (4) Extracurricular activities and gamified learning: Through the interesting game tasks in the system, encourage children to improve their cognitive abilities in an informal learning environment, combining daily learning with extracurricular activities.

[0174] (5) Career development and selection: It can be used for the cognitive ability assessment of specific groups to assist in talent selection and career planning.

[0175] The implementation methods of the present invention include the specific implementation steps of system architecture design, data processing flow, task execution, and scoring mechanism. Through artificial intelligence algorithms to analyze task data, the system can automatically generate personalized assessment reports and provide ability improvement suggestions. The following are the specific implementation methods of this system:

[0176] 1. System architecture and module design

[0177] The hardware platform of the system includes intelligent mobile devices (such as tablets, smartphones) and PC terminals, which are required to support touch operations, high processing capabilities, stable network connections, and graphics and sound outputs.

[0178] The system architecture can be divided into the following modules:

[0179] (1) User interface module: Used to display game tasks, feedback, and reports.

[0180] (2) Task assessment module: Responsible for executing various game tasks and recording children's performances.

[0181] (3) Data processing and analysis module: Collect various data in the tasks, conduct data analysis, and calculate scores. (4) Report generation module: Generate personalized assessment reports according to the analysis results, providing detailed ability analysis and development suggestions.

[0182] (5) Background management module: Used to manage and view user data, perform data storage, and system configuration.

[0183] 2. Evaluation process and methods

[0184] This evaluation system designs a standardized evaluation process as follows:

[0185] (1) Input information

[0186] Collect basic information of children, including name, gender, age, grade, etc.

[0187] (2) Evaluation tasks

[0188] The tasks involve six core cognitive domains, each of which contains multiple subtasks, and can effectively evaluate children's performance in each cognitive dimension.

[0189] (3) Data collection and analysis

[0190] The system automatically records key data such as task completion time and accuracy rate. Based on dimensions such as task completion accuracy, speed, and strategy selection, combined with a big data model, a standardized score is given.

[0191] (4) Scoring mechanism

[0192] The scores of each domain are calculated in a standardized way and normalized to a 100-point system.

[0193] (5) Report generation

[0194] According to the evaluation results, the system will automatically generate a cognitive ability evaluation report, including the overall score, analysis of each domain, ability radar chart, and personalized development suggestions based on the evaluation results.

[0195] (6) Sample output report

[0196] During the report generation process, the system will provide visual feedback according to the children's evaluation results and give specific suggestions to parents or educators, specifically including:

[0197] A. Basic information

[0198] The report starts with the child's name, gender, age, grade, test time, and test duration.

[0199] B. Display of evaluation results

[0200] a. Overall score and development level: Such as "In this evaluation, the overall cognitive development score is 67 points (out of 100), and the development level is B (↑), with good performance and being at a relatively high level in the country".

[0201] b. Radar chart: The heptagon radar chart (see Figure 4 ) intuitively shows the scores of the six core domains and the balance of overall cognitive ability.

[0202] c. Performance in each domain: Details of the scores and ability characteristics in each domain are described as follows:

[0203] Self-control score: 75 points (out of 100), development level is B, showing good performance and being at a relatively high level in the country.

[0204] Feature description: Can focus on the current task without being disturbed by irrelevant factors. In a complex environment, can reach the goal quickly and accurately. Has strong emotional and thinking control abilities, showing good behavioral self-control and stress-coping abilities.

[0205] Parent suggestion: Maintain a good self-management state and at the same time give children more room for independent development. Personalized development suggestions

[0206] C. Personalized development suggestions

[0207] If the score in a certain area is low, it will prompt parents about which activities or strategies can be used to help children improve their abilities in that area, and recommend relevant learning resources or training activities based on the children's performance to promote the overall improvement of children's cognitive abilities.

[0208] Although the preferred embodiments of the present invention have been described in detail, those skilled in the art may still make further adjustments and improvements to these embodiments after understanding its basic innovative concept. Therefore, the appended claims are intended to cover these preferred embodiments, as well as all changes and modifications that fall within the scope of the present invention. The above content is only an example of the preferred embodiments of the present invention and does not constitute a limitation thereto. It should be clear that any modifications, equivalent replacements, or improvements made under the guidance of the spirit and principles of the present invention should be considered as being included within the protection scope of the present invention.

Claims

1. A method for evaluating the cognitive development ability of children's brains and minds, characterized in that, Including: Receiving user identity information, including age, gender, and education stage; Constructing an evaluation model covering six first-level dimensions and eight second-level dimensions. The first-level dimensions include memory, self-control, thinking ability, reaction ability, attention, and spatial ability. The second-level dimensions include short-term memory, working memory, inhibitory control, task switching, processing speed, logical thinking, concentration, and visual perception space; Randomly extracting a target game task set from the task library. The task set covers the evaluation requirements of at least three first-level dimensions; during the task execution, randomly select game tasks from the second-level dimensions for evaluation, and collect the behavioral data of children during the task, including reaction time, accuracy, and strategy selection; Constructing a norm data table based on the behavioral data of children of different ages. The norm data table contains the mean and standard deviation of each dimension; Calculating the task score, first-level dimension score, and total score based on the behavioral data, and converting the scores into percentiles based on the norm data table, and dividing them into six grades.

2. The method for evaluating the cognitive development ability of children's brain intelligence according to claim 1, characterized in that Memory includes two dimensions: short-term memory and working memory; self-control includes inhibitory control and task switching; thinking ability corresponds to logical thinking; reaction ability corresponds to processing speed; attention corresponds to concentration; spatial ability corresponds to visual perception space; Based on a pre-trained cognitive ability evaluation model, perform multi-dimensional feature fusion on the collected data to generate an ability vector including memory index, self-control index, thinking ability index, reaction ability index, attention index, and spatial ability index; Comparing and analyzing the ability vector with the norm database, and outputting an evaluation report with grade symbol markings; the norm database is a standardized database constructed based on the behavioral data of children of different ages, and is used to measure whether the performance of children in each cognitive dimension meets the average level of the same-age group; the norm database includes the following contents: age segmentation, the norm database is segmented according to the age of children, and each age group corresponds to different norm data; the mean and standard deviation of each dimension: for each first-level dimension and / or second-level dimension, the database stores the mean and standard deviation of this dimension at a specific age; recommending targeted training tasks according to the grade and score; The norm database is a standardized database constructed based on the behavioral data of children of different ages, and is used to measure whether the performance of children in each cognitive dimension meets the average level of the same-age group; the norm database includes the following contents: age segmentation: the norm database is segmented according to the age of children, and each age group corresponds to different norm data; the mean and standard deviation of each dimension: for each first-level dimension and / or second-level dimension, the database stores the mean and standard deviation of this dimension at a specific age.

3. A method for evaluating the cognitive development ability of children's brains and minds according to claim 2, characterized in that, The training process of the pre-trained cognitive ability evaluation model includes: collecting the operation data set of historical users, and the data set is labeled with ability level labels; constructing a deep neural network model, the input layer receives the coordinate sequence of the operation trajectory, and the hidden layer contains LSTM units for extracting temporal features; using a contrast loss function to optimize the model, and the loss function calculates the difference between the predicted ability level and the expert annotation.

4. The method for evaluating the cognitive development ability of children's brain intelligence according to claim 2, characterized in that, The recommendation of targeted training tasks according to the level and score is completed by a personalized recommendation engine, which is configured to: analyze the abnormal dimensions in the ability level mark; match the intervention plan from the policy knowledge base, where the policy knowledge base stores a list of training activities associated with each level interval; generate an executable plan including gamified training task recommendations and family education guidance.

5. A method for evaluating the cognitive development ability of children's brain intelligence according to claim 2, characterized in that The dynamic adjustment of the difficulty of the training task includes: when it is detected that the user's operation accuracy rate is higher than the first threshold for three consecutive times, the task difficulty level is increased, and the increase includes increasing the moving speed of the stimulator or reducing the effective response time window; when it is detected that the user's error rate exceeds the second threshold, the auxiliary prompt mechanism is activated, and the mechanism includes highlighting the target area or extending the task time limit.

6. The method for evaluating the cognitive development ability of children's brain intelligence according to claim 1, characterized in that Construct an age-segmented norm database, which stores the average reaction time and standard deviation of users of each age group in the preset task; use the Z algorithm to transform the user's original score, and the calculation formula is: where X is the user's score, μ is the norm mean of the corresponding age group, σ is the norm standard deviation; Z is the percentile, and the ability level mark is generated.

7. The method for evaluating the cognitive development ability of children's brain intelligence according to claim 4, characterized in that The construction method of the policy knowledge base involves constructing a comprehensive ability training framework, establishing a dynamic mapping relationship between ability types and intervention measures, including multi-level and hierarchical training programs; this framework covers multiple dimensions such as short-term memory, working memory, inhibitory control, task switching, processing speed, logical thinking, concentration, and visual perception space, and provides personalized training intensity, task complexity, and auxiliary mechanisms for different levels to effectively improve cognitive function and optimize learning effects.

8. The method for evaluating the cognitive development ability of children's brain intelligence according to claim 2, characterized in that The collected data includes the spatio-temporal characteristics of the touch operation trajectory, the task completion progress and the number of interruptions, and the frequency statistics of the mis-triggering of interference items. The multi-dimensional feature fusion steps specifically include: Perform wavelet transform on the spatio-temporal characteristics to extract the frequency domain characteristics of the operation trajectory; Quantify the task completion progress into a time efficiency coefficient, and the calculation formula is: where T 基准 is the norm average completion time, and T 实际 is the actual time consumed by the user; weighted fusion of multi-source features is performed to generate a comprehensive ability score; The collected data and multi-dimensional feature fusion is a method for comprehensively evaluating the user's ability. Through data collection and feature extraction from different dimensions, a score reflecting the user's comprehensive ability is generated; specifically: Data collection: The spatio-temporal characteristics of the touch operation trajectory record all the operation paths and time information of the user when touching the screen; including the user's gestures, sliding speed, acceleration, etc., reflecting the user's operation proficiency, coordination ability, and reaction speed; Task completion progress and number of interruptions: Monitor the progress of the user in the process of completing a specific task, including the gap between the time required to complete the task and the expected time, and the number of interruptions during the task, to evaluate the user's task processing efficiency and concentration; Frequency statistics of interference item mis-triggering: Record the frequency of the user triggering options or functions that are irrelevant to the current task during the operation process, reflecting the user's concentration and accuracy in a complex task environment; Feature extraction and processing: Perform wavelet transform on spatio-temporal features: Convert the collected spatio-temporal feature data from the time domain to the frequency domain; Help analyze the rhythm and pattern in the user's operation; Quantification of task completion progress: Convert the task completion progress into a time efficiency coefficient; This coefficient can quantify the efficiency of the user in completing the task, and the larger the value, the faster the task is completed and the higher the efficiency; Multi-dimensional feature fusion: Adopt an attention mechanism to dynamically allocate weights to the data in each dimension according to the importance of different features, helping the system to better focus on important information and reduce the interference of irrelevant information; Generate a comprehensive ability score: Generate a comprehensive score through weighted fusion of the processed multi-dimensional data; This score can comprehensively reflect the user's comprehensive ability in terms of operation proficiency, task efficiency, concentration, etc., and provide a comprehensive evaluation result.

9. The method for evaluating the cognitive development ability of children's brain intelligence according to claim 1, characterized in that, It also includes a data security module, configured to: Perform desensitization processing on the user identity information to generate an anonymized unique identifier; Adopt homomorphic encryption technology to encrypt and transmit the operation trajectory data; Establish a two-way verification channel between the local storage device and the cloud server.

10. A children's brain intelligence cognitive development ability evaluation system, characterized in that, Adopt a method for evaluating the cognitive development ability of children's brain intelligence according to any one of claims 1-9; including: A user interface module for displaying game tasks and feedback reports and supporting touch interaction; A task evaluation module for executing game tasks and recording behavior data, and the tasks are randomly retrieved according to the secondary dimension; A data analysis module for calculating percentiles and grades according to the norm data table and generating multi-dimensional scores; A report generation module for integrating radar charts, grade marks and suggestions to generate a visual report; A background management module for storing user data, configuring system parameters and managing the training task library.

Images