A child AI non-decision thinking cultivation method and system based on guided interaction

Abstract

The present application belongs to the technical field of children's AI non-decision thinking cultivation method, and particularly relates to a children's AI non-decision thinking cultivation method and system based on guided interaction, which acquires interactive input information of children, the interactive input information including one or more of voice, text and action instructions; performs semantic analysis on the interactive input information and extracts core demands, matches guided dimensions and guided depths suitable for children's age and cognitive level in combination with a preset children's cognitive development model; generates stepwise guided questioning content based on the core demands and the matched guided dimensions and guided depths, the guided questioning content not directly giving answers to corresponding questions, but guiding children to independently think from exploration direction, analysis angle and logical arrangement level.

Description

Technical Field

[0001] This invention belongs to the technical field of methods for cultivating children's AI non-decision thinking, and particularly relates to a method and system for cultivating children's AI non-decision thinking based on guided interaction. Background Technology

[0002] Currently, AI products for children are widely used in various scenarios such as family companionship and early childhood education, becoming a common interactive medium in children's growth process. However, the core interaction mode of existing AI products for children has obvious design flaws. They mostly adopt decision-making interaction logic, that is, directly providing standard answers or making choices when children ask questions or make requests. Although this can quickly meet children's immediate information needs, it deprives children of the process of independent exploration, problem analysis, and independent thinking. Long-term use can easily lead to the domestication of children's thinking, making them dependent on AI and making it difficult to cultivate core thinking abilities such as logical analysis and divergent thinking. At the same time, these products have not been designed to differentiate themselves according to the cognitive development patterns of children of different ages. The uniform guidance mode cannot adapt to the different thinking characteristics of children of different ages, such as young children and school-age children, and the guidance effect is greatly reduced.

[0003] Furthermore, existing AI products for children suffer from problems such as ambiguous data sovereignty, insufficient accuracy in interactive recognition, and a lack of a closed loop for thinking development. Some products treat children's interactive data as commercial assets, posing a risk that the data may be accessed and sold by third parties, violating the principle of sovereignty over children's data. At the same time, they lack the ability to effectively identify the nature of children's speech, failing to distinguish between joking expressions and genuine needs, which can easily lead to targeted misguidance. Moreover, the interaction process lacks a thought review stage, making it impossible to organize and reinforce children's thinking process, and thus difficult to form a complete thinking development system. In addition, the lack of contextual adaptation capabilities and the disconnect between the guidance content and children's actual life scenarios further reduce the actual value of AI in children's thinking development, which also contradicts the original intention of the companion-oriented guidance design of AI products for children's digital personality inheritance. Summary of the Invention

[0004] In view of the aforementioned problems, and in conjunction with the first aspect of the present invention, embodiments of the present invention provide a method for cultivating children's non-decision-making thinking based on guided interaction, the method comprising:

[0005] Step 1: Obtain the child's interactive input information, which includes one or more of voice, text, and action commands;

[0006] Step 2: Semantically analyze the interactive input information and extract the core needs. Combine this with a preset child cognitive development model to match the guidance dimensions and depth that are appropriate for the child's age and cognitive level.

[0007] Step 3: Based on the core needs and the matched guidance dimensions and depth, generate tiered guidance questions. These questions do not directly provide answers to the corresponding questions, but rather guide children to think independently from the perspectives of exploration direction, analysis angle, and logical organization.

[0008] Step 4: Output the guiding questions to the child, obtain the child's feedback information in real time and perform semantic analysis, dynamically adjust the content and depth of the guiding questions according to the thinking level reflected in the feedback information, form a multi-round guided interaction, until the child completes independent thinking or actively terminates the interaction.

[0009] Step 5: After the interaction ends, generate a review of the child's thinking process based on the full-process data of multiple rounds of interactive feedback, and sort out the child's thinking nodes and logical structure;

[0010] Step Six: All data from this interaction will be encrypted and stored locally, and data viewing and management permissions will only be granted to children and their legal guardians. Third parties will not have the right to access the data.

[0011] Preferably, the children's cognitive development model described in step two is a graded model constructed based on the cognitive abilities and thinking characteristics of children of different ages. It is divided into three age groups: the lower age group (3-6 years old), the school age group (7-12 years old), and the adolescent age group (13-16 years old). Each age group corresponds to different guidance dimensions and depths. The lower age group is mainly guided by concrete and fun exploration, the school age group is mainly guided by logical analysis and problem decomposition, and the adolescent age group is mainly guided by dialectical thinking and multidimensional exploration.

[0012] Preferably, the step-by-step guided questioning in step three adopts a three-level design from simple to complex. The first level is basic exploratory questioning, which targets the superficial characteristics and intuitive perception of the problem; the second level is logical analysis questioning, which targets the internal connections and causes of the problem; and the third level is extended thinking questioning, which targets the extended scenarios and multiple solutions of the problem.

[0013] Preferably, step four also includes a boasting identification step: the nature of the child's feedback information is judged by a preset semantic feature database and behavioral feature database, distinguishing between the child's joking expression and real demand. If it is judged to be a joking expression, the child is guided to ask questions in a fun way without logical guidance on substantive questions; if it is judged to be a real demand, the content and depth of the guiding questions are dynamically adjusted according to the original logic.

[0014] Preferably, the criteria for judging the joking expression include the objective rationality of the expression content, the child's tone characteristics, and the fun characteristics of the interactive scene. When at least two of the three characteristics meet the preset joking characteristic threshold, it is judged as a joking expression.

[0015] Preferably, the review content described in step five is presented in a form that children can understand, including one or more of text, audio, and cartoon graphics, and the review content only summarizes the thinking process and does not evaluate the right or wrong of the thinking results.

[0016] Preferably, the local encrypted storage in step six adopts a hardware localized storage mode. During the data storage process, a DNA-bound permission verification mechanism is set up. The child's legal guardian can view and manage the data through daily verification methods of fingerprint + retina. DNA verification is only used for data inheritance, recovery and legal confirmation. The hardware localized storage mode has the characteristic of self-destruction upon disassembly, so that the data never leaves the domain.

[0017] Preferably, during the multi-round interactive feedback process described in step four, if the child is detected to have mental blocks or negative emotions such as irritability or resistance, the in-depth progression of guided questioning is paused, and the questioning is changed to a shallow level of encouragement and inspiration, or the child is proactively advised to pause the interaction.

[0018] Preferably, before generating the guiding question content in step three, an interactive scenario adaptation step is also included: adjusting the guiding question content according to the scenario information of the interaction, wherein the scenario information includes one or more of home, campus, and outdoor.

[0019] In step one, speech recognition, image recognition, and text parsing technologies are used to accurately capture interactive input information. In steps two and four, natural language processing technology is used to complete semantic parsing.

[0020] Furthermore, embodiments of the present invention also provide a guided interactive AI non-decision thinking training system for children, characterized in that it includes:

[0021] A processor; a machine-readable storage medium for storing machine-executable instructions of the processor; wherein the processor is configured to execute the above-described guided interaction-based method for cultivating children's non-decision-making thinking via executing the machine-executable instructions.

[0022] In another aspect, embodiments of the present invention also provide a computer program product, the computer program product including machine-executable instructions, the machine-executable instructions being stored in a computer-readable storage medium, a processor of a computer device reading the machine-executable instructions from the computer-readable storage medium, the processor executing the machine-executable instructions, causing the computer device to execute the above-described method for cultivating children's AI non-decision thinking based on guided interaction.

[0023] Based on the above, by constructing a full-process guided interaction system, the traditional decision-making interaction mode of children's AI is completely abandoned. A tiered questioning approach guides children to independently complete the entire process of exploration, analysis, and thinking, fundamentally avoiding the problem of mind training and effectively cultivating children's independent thinking, logical analysis, and divergent thinking abilities. Simultaneously, personalized guidance is achieved based on age-appropriate children's cognitive development models. Combined with a boast recognition module, it accurately distinguishes between children's joking expressions and genuine needs. Coupled with emotion perception and scenario-based adaptation design, the guided interaction not only fits the thinking characteristics and real-life scenarios of children of different ages but also avoids misguidance and thinking anxiety, significantly improving the targeting, effectiveness, and user-friendliness of AI in children's thinking development.

[0024] This invention achieves full ownership of children's interactive data through a hardware-localized encrypted storage mode combined with a DNA-bound permission verification mechanism. Only legal guardians can manage the data through exclusive verification. The self-destructive feature of the hardware ensures that the data never leaves the domain, technically eliminating the risk of third-party access to and sale of children's data, and guaranteeing data security and privacy compliance. Simultaneously, the included thought review process can organize and strengthen children's thinking nodes and logical connections, forming a complete closed loop of thinking cultivation: interaction triggering, cognitive matching, guided questioning, multi-round interaction, thought review, and secure storage. This extends thinking cultivation from single interactions to habit formation, realizing the long-term value of thinking cultivation. Furthermore, this method can be implemented in corresponding systems, with clear execution logic and strong stability, providing a feasible technical solution for AI products related to children's digital personality companionship. Attached Figure Description

[0025] Figure 1 This is a schematic diagram of the execution flow of the method for cultivating children's non-decision thinking based on guided interaction provided in the embodiments of the present invention.

[0026] Figure 2 This is a schematic diagram of exemplary hardware and software components of a guided interactive AI non-decision thinking training system for children provided in an embodiment of the present invention. Detailed Implementation

[0027] The technical solutions of the embodiments of this application will be clearly described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of this application. All other embodiments obtained by those skilled in the art based on the embodiments of this application are within the scope of protection of this application.

[0028] To achieve the above-mentioned objectives, this invention provides a method for cultivating children's non-decision-making thinking based on guided interaction. This method constructs a guided interaction system based on the laws of children's cognitive development, combines a boasting recognition module with a data security storage mechanism, and forms a complete non-decision-making thinking cultivation program, specifically including the following steps:

[0029] The interaction triggers the acquisition of children's interactive input information, which includes one or more of voice, text, and action commands. AI uses technologies such as speech recognition, image recognition, and text parsing to accurately capture the interactive input information, providing a foundation for subsequent semantic analysis.

[0030] Cognitive matching first performs semantic parsing on the interactive input information, extracting the core needs through natural language processing technology. These core needs include information retrieval, problem-solving, interest exploration, and emotional communication. Then, it combines a pre-set child cognitive development model to match guidance dimensions and depths appropriate to the child's age and cognitive level. The child cognitive development model is a tiered model built based on the cognitive abilities and thinking characteristics of children at different ages, divided into three age groups: lower age (3-6 years old), school age (7-12 years old), and adolescent age (13-16 years old). Lower age children primarily use concrete thinking, so guidance focuses on concrete and engaging exploration, with a shallower guidance depth. School age children have developed basic logical thinking, so guidance focuses on logical analysis and problem breakdown, with a moderate guidance depth. Adolescent children possess dialectical thinking abilities, so guidance focuses on dialectical thinking and multidimensional exploration, with a deeper guidance depth.

[0031] The guided question generation system is based on extracted core needs and matched guidance dimensions and depths, generating tiered guided question content. The core design principle of this system is not to directly provide answers to the questions, but rather to guide children's independent thinking from the perspectives of exploration, analysis, and logical organization. The tiered guided question content adopts a three-level design from simple to complex: the first level is basic exploration questions, targeting the superficial characteristics and intuitive perception of the problem, helping children build a basic understanding of the problem; the second level is logical analysis questions, targeting the internal connections and causes of the problem, guiding children to organize the logical relationships within the problem; the third level is extended thinking questions, targeting extended scenarios and multiple solutions to the problem, cultivating children's divergent thinking abilities. Simultaneously, before generating the guided question content, it is adapted to the interactive scenario information (home, school, outdoors, etc.), integrating the guided questions with children's actual life scenarios to enhance the relevance and fun of the interaction.

[0032] The multi-round interactive feedback system first presents guiding questions to children in a child-understandable format, such as voice or text. It then captures and semantically analyzes the child's feedback in real time, dynamically adjusting the content and depth of the guiding questions based on the level of thought revealed in the feedback: if the child's feedback reaches the current level of guidance and demonstrates the ability to think further, it advances to the next level of guiding questions; if the child's feedback indicates a mental block, it reverts to the previous level of guiding questions or simplifies the questions. This process forms a multi-round guided interaction until the child completes independent thinking, arrives at their own conclusion, or voluntarily terminates the interaction. Simultaneously, this step includes a "bragging" identification step: using a pre-set semantic and behavioral feature database, the system judges the nature of the child's feedback, distinguishing between joking expressions and genuine needs. The criteria for judging joking expressions include the objectivity and rationality of the content, the child's tone, and the fun of the interactive scenario. If at least two of these three features meet a pre-set joking feature threshold, it is determined to be a joking expression. If the expression is judged to be joking, the response will be guided by fun questions, without delving into the logic of substantive issues. If the expression is judged to be a genuine request, the guidance questions will be dynamically adjusted according to the original logic. Furthermore, if negative emotions such as mental blocks, frustration, or resistance are detected during the interaction, the deepening of the guidance questions will be paused, replaced by encouraging and thought-provoking questions at a shallower level, or a suggestion to pause the interaction will be proactively made to avoid causing the child mental anxiety and to ensure a friendly approach to developing thinking skills.

[0033] After the interactive session, the review process generates a summary of the child's thinking process based on data from multiple rounds of feedback. This summary is presented in child-understandable language, including text, audio, and cartoon illustrations. Its core function is to organize the child's thought processes and logical connections, rather than evaluating the correctness of the results. Through this review, children gain a clearer understanding of their own thought processes, strengthening their independent thinking and upgrading their cognitive development from "passive guidance" to "active organization."

[0034] Secure data storage involves locally encrypting and storing all data from this interaction, including input information, guided questions, child feedback, and thought-reflection, abandoning centralized cloud storage to prevent third-party access. A DNA-bound access control mechanism is implemented during data storage, allowing only the child's legal guardian to view and manage the data via routine fingerprint and retinal verification. DNA verification is solely for data inheritance, recovery, and legal ownership confirmation, ensuring data sovereignty belongs to the child and their legal guardian. Furthermore, the data storage employs a localized hardware design, self-destructing upon data disassembly, ensuring data never leaves the local domain.

[0035] To make the technical solution of the present invention clearer and easier to understand, the following describes in detail the method for cultivating children's AI non-decision thinking based on guided interaction in conjunction with specific implementation scenarios.

[0036] Implementation Scenario 1: Information query interaction for young children (4 years old)

[0037] After a child sees the sky outdoors, they can ask the AI ​​via voice interaction, "Why is the sky blue?" The steps of the method of this invention are as follows:

[0038] Interactive trigger: AI captures the child's core voice input, "Why is the sky blue?", through speech recognition;

[0039] Cognitive matching: After semantic parsing, the core demand is extracted as "information query on natural phenomena". Combined with the characteristics of children aged 4, concrete and interesting guidance dimensions are matched, and the guidance depth is based on the exploration level.

[0040] Guided question generation: Based on outdoor scene information, generate tiered basic exploration questions such as, "Baby, look, the sky is blue on a sunny day, so what color is the sky on a cloudy day?", without directly giving the scientific answer that the sky is blue;

[0041] Multi-round interactive feedback: The AI ​​outputs the above question in a childlike voice. The child responds, "The sky is gray on a cloudy day!" After semantic analysis, it is determined that the child has completed the basic exploration. The AI ​​dynamically adjusts and generates the next shallow-level guiding question: "Then look again, is the color of the sea similar to the color of the sky on a sunny day?" The child continues to respond, "Yes! They are both blue!" At this point, it is detected that the child is in an active thinking state. The AI ​​continues to generate guiding questions: "Then guess, is the blue color of the sky related to water?" The child says "I don't know" without showing any negative emotions. The AI ​​continues to guide: "Then let's go home and look at picture books together to find the answer, okay?" The child actively responds "Okay," and the interaction ends.

[0042] Thinking Review: After the interaction, the AI ​​generates a review in the form of cartoon images and text + voice: "Today, my child observed that the sky is blue on sunny days and gray on cloudy days. They also discovered that the sky and the sea are both blue. Your little eyes are so observant! Let's go home and find the answers together!" The review only summarizes the observation and thinking points and does not evaluate right or wrong.

[0043] Secure data storage: Data such as voice interaction, guided questions, children's feedback, and cartoon review will be stored locally in an encrypted manner. Only the child's parents can view this data through fingerprint verification.

[0044] Implementation Scenario 2: Problem-solving interactions for school-aged children (8 years old)

[0045] When a child is doing math problems, they can input the following text into the AI: "How can 10 bananas be divided fairly among 3 children?" The method of this invention executes the following steps:

[0046] Interactive trigger: AI captures the child's core input through text parsing: "How can 10 bananas be fairly divided among 3 children?"

[0047] Cognitive matching: After semantic parsing, the core demand is extracted as "solving mathematical problems". Combined with the characteristics of children aged 8, the guidance dimensions of logical analysis and problem decomposition are matched, and the guidance depth is the level of logical analysis.

[0048] Guided Question Generation: Based on family scenario information, generate tiered guided questions: "First think about it, in a fair way, shouldn't each child get the same number of bananas? If 10 bananas are divided among 3 children, how many bananas will each child get first, and how many will be left over?"

[0049] Multi-round interactive feedback: The AI ​​outputs the above question, and the child responds: "Each child gets 3 bananas, and there's 1 left!" After semantic analysis, it determines that the child has completed the basic breakdown and dynamically adjusts to generate the next logical analysis question: "So, how should we deal with the remaining banana to make the division still fair?" The child responds: "Cut the remaining banana into 3 small pieces, and each child gets 1 piece!" The AI ​​determines that the child has independently come up with a solution and continues to generate an extension question: "If the banana cannot be cut, are there other fair ways to divide it?" After thinking, the child says "I can't think of any," without any negative emotions, and the interaction ends.

[0050] Thinking review: The AI ​​generates the review content in the form of voice and text: "You first thought that the core of a fair distribution method is that the quantity is the same, calculated that 10 bananas are divided among 3 people, each person gets 3 bananas and there is 1 banana left over, and then you thought of cutting the remaining banana to achieve fairness. Your logical thinking is very clear. Next time we can think of other ways to distribute them!"

[0051] Secure data storage: Data such as text interactions, guided questions, children's feedback, and review content are encrypted and stored locally, with no third-party access rights.

[0052] Implementation Scenario 3: Interaction with children's playful expressions

[0053] When a child at home gives voice input to an AI, saying, "I can fly to the moon, I want to go to the moon to play!", the method of this invention executes the following steps:

[0054] Interactive trigger: AI captures the child's core voice input through voice recognition: "I can fly to the moon and I want to go to the moon to play."

[0055] Cognitive matching: After semantic parsing, the core demand is extracted as "fun expression". Combined with the characteristics of children aged 6, fun guidance dimensions are matched.

[0056] Guided Question Generation: Initially generated engaging guided questions, to be adjusted based on the results of the boasting identification.

[0057] Multi-round interactive feedback: The boasting recognition process begins. The AI ​​determines that the child's statement "I can fly to the moon" is objectively unreasonable, the tone is cheerful and joking, and the interaction scenario is a fun family conversation. Two of these three characteristics meet the joke threshold, thus classifying it as a joking expression. The AI ​​then provides a fun guiding question: "Wow, baby wants to fly to the moon! What kind of transportation do you think you need to fly to the moon?" The child replies: "A rocket!" The AI ​​continues with fun guidance: "What do you need to prepare for your rocket to play on the moon?" This creates a fun, multi-round interaction, until the child voluntarily terminates the interaction.

[0058] Mindset debriefing: AI generates debriefing content in the form of cartoon voice: "Baby's rocket is going to play on the moon! Baby's imagination is so rich!"

[0059] Secure data storage: All data from this interactive activity will be encrypted and stored locally.

[0060] Based on the same inventive concept, please refer to Figure 2 The diagram shows a schematic block diagram of a guided interaction-based non-decision thinking training system 100 for performing the above-described guided interaction-based non-decision thinking training method for children's AI. The guided interaction-based non-decision thinking training system 100 may include a communication unit 110, a machine-readable storage medium 120, and a processor 130.

[0061] In this embodiment, both the machine-readable storage medium 120 and the processor 130 are located within the guided interaction-based AI non-decision-making thinking development system 100 for children and are separately configured. However, it should be understood that the machine-readable storage medium 120 may also be independent of the guided interaction-based AI non-decision-making thinking development system 100 for children and may be accessed by the processor 130 via a bus interface. Alternatively, the machine-readable storage medium 120 may also be integrated into the processor 130 and may communicate and interact with external systems via the communication unit 110.

[0062] The processor 130 is the control center of the guided interactive AI non-decision thinking training system 100 for children. It connects various parts of the system via various interfaces and lines, and performs overall monitoring of the system by running or executing software programs and / or modules stored in the machine-readable storage medium 120 and accessing data stored in the medium. Optionally, the processor 130 may include one or more processing cores; for example, it may integrate an application processor and a modem processor, where the application processor primarily handles the operating system, user interface, and applications, and the modem processor primarily handles wireless communication. It is understood that the modem processor may also not be integrated into the processor. The machine-readable storage medium 120 is used to store machine-executable instructions for executing the scheme of this application, and the processor 130 is used to execute the machine-executable instructions stored in the machine-readable storage medium 120 to realize the method for cultivating children's AI non-decision thinking based on guided interaction provided in the aforementioned method embodiments.

[0063] It should be noted that, in order to simplify the description of the present invention and thus help to understand one or more embodiments of the invention, multiple features may sometimes be grouped into one embodiment, drawing or description thereof in the foregoing description of the embodiments of the present invention.

[0064] The embodiments of this application have been described above with reference to the accompanying drawings. Unless otherwise specified, the embodiments and features in the embodiments of this application can be combined with each other. This application is not limited to the specific embodiments described above. The specific embodiments described above are merely illustrative and not restrictive. Those skilled in the art can make many other forms under the guidance of this application without departing from the spirit and scope of the claims, and all of these forms are within the protection scope of this application.

Claims

1. A method for cultivating children's non-decision-making thinking based on guided interaction, characterized in that: Includes the following steps: Step 1: Obtain the child's interactive input information, which includes one or more of voice, text, and action commands; Step 2: Semantically analyze the interactive input information and extract the core needs. Combine this with a preset child cognitive development model to match the guidance dimensions and depth that are appropriate for the child's age and cognitive level. Step 3: Based on the core needs and the matched guidance dimensions and depth, generate tiered guidance questions. These questions do not directly provide answers to the corresponding questions, but rather guide children to think independently from the perspectives of exploration direction, analysis angle, and logical organization. Step 4: Output the guiding questions to the child, obtain the child's feedback information in real time and perform semantic analysis, dynamically adjust the content and depth of the guiding questions according to the thinking level reflected in the feedback information, form a multi-round guided interaction, until the child completes independent thinking or actively terminates the interaction. Step 5: After the interaction ends, generate a review of the child's thinking process based on the full-process data of multiple rounds of interactive feedback, and sort out the child's thinking nodes and logical structure; Step Six: All data from this interaction will be encrypted and stored locally, and data viewing and management permissions will only be granted to children and their legal guardians. Third parties will not have the right to access the data.

2. The method for cultivating children's non-decision-making thinking based on guided interaction according to claim 1, characterized in that: The cognitive development model for children described in Step Two is a graded model constructed based on the cognitive abilities and thinking characteristics of children of different ages. It is divided into three age groups: the lower age group (3-6 years old), the school age group (7-12 years old), and the adolescent age group (13-16 years old). Each age group corresponds to different guidance dimensions and depths. The lower age group is mainly guided by concrete and fun exploration, the school age group is mainly guided by logical analysis and problem decomposition, and the adolescent age group is mainly guided by dialectical thinking and multidimensional exploration.

3. The method for cultivating children's non-decision-making thinking based on guided interaction according to claim 1, characterized in that: The step-by-step guided questioning in step three adopts a three-level design from simple to complex. The first level is basic exploratory questioning, which points to the superficial characteristics and intuitive perception of the problem. The second level is logical analysis questioning, which focuses on exploring the internal connections and causes of the problem; The third level involves expanding thinking and asking questions, which points to extended scenarios and multiple solutions to the problem.

4. The method for cultivating children's non-decision-making thinking based on guided interaction according to claim 1, characterized in that: Step four also includes a boasting identification step: the nature of the child's feedback is judged by a preset semantic feature database and behavioral feature database, distinguishing between the child's joking expression and real needs. If it is judged to be a joking expression, the child will respond with fun questions without logical guidance on substantive issues; if it is judged to be a real need, the content and depth of the guiding questions will be dynamically adjusted according to the original logic.

5. A method for cultivating children's non-decision-making thinking based on guided interaction according to claim 4, characterized in that: The criteria for judging a joking expression include the objectivity and rationality of the content, the child's tone, and the fun of the interactive scene. When at least two of the three characteristics meet the preset joking characteristic threshold, it is judged as a joking expression.

6. A method for cultivating children's non-decision-making thinking based on guided interaction according to claim 1, characterized in that: The review content described in step five is presented in a form that children can understand, including one or more of text, audio, and cartoon graphics. The review content only summarizes the thinking process and does not evaluate the right or wrong of the thinking results.

7. A method for cultivating children's non-decision-making thinking based on guided interaction according to claim 1, characterized in that: The local encrypted storage described in step six adopts a hardware localized storage mode. During the data storage process, a DNA-bound permission verification mechanism is set up. The child's legal guardian can view and manage the data through daily verification methods of fingerprint + retina. DNA verification is only used for data inheritance, recovery and legal confirmation. The hardware localized storage mode has the characteristic of self-destruction upon disassembly, so that the data never leaves the domain.

8. A method for cultivating children's non-decision-making thinking based on guided interaction according to claim 1, characterized in that: In the multi-round interactive feedback process described in step four, if the child is detected to have mental blocks or negative emotions such as irritability or resistance, the in-depth progression of guided questioning should be paused, and the questioning should be changed to an encouraging and inspiring shallow level, or the child should be proactively advised to pause the interaction.

9. A method for cultivating children's non-decision-making thinking based on guided interaction according to claim 1, characterized in that: Before generating the guiding questions in step three, there is also an interactive scenario adaptation step: the guiding questions are adjusted according to the scenario information of the interaction, which includes one or more of the following: home, campus, and outdoors. In step one, speech recognition, image recognition, and text parsing technologies are used to accurately capture interactive input information. In steps two and four, natural language processing technology is used to complete semantic parsing.

10. A child AI non-decision thinking training system based on guided interaction, characterized in that, include: processor; A machine-readable storage medium for storing machine-executable instructions of the processor; The processor is configured to execute the method for cultivating children's non-decision thinking based on guided interaction as described in any one of claims 1 to 9 by executing the machine-executable instructions.

Images