Short-term memory manipulation ability training method and related product

Abstract

The invention discloses a short-term memory manipulation ability training method and related products, and the method comprises the steps: determining an initial training level based on the basic information of a target object; determining a material generation condition of current training of the target object based on the initial training level and the basic information; inputting the material generation condition into a large language model to obtain a training material; generating a training task based on the training material and the initial training level; displaying the training task to the target object, and obtaining feedback data of the target object on the training task; and training the target object for multiple times based on the feedback data and the basic information. By using the method, the challenging and interestingness of training are improved, the training effect and efficiency are further improved, and the training participation degree and the user experience are improved.

Description

Technical Field

[0001] The present invention relates to the field of computer technology, and particularly to a training method for short-term memory manipulation ability and related products. Background Art

[0002] Memory function is a complex cognitive process, which can be divided into short-term memory and long-term memory according to time length. Among them, short-term memory (also known as working memory) is the basis of long-term memory.

[0003] Currently, there are also some training websites or programs for short-term information manipulation ability on the market to help people train their own short-term information manipulation ability. Generally speaking, the existing training programs for short-term information manipulation ability are mainly based on the N-back task and the running memory task. Among them, the N-back task requires an individual to continuously compare the currently presented stimulus material with the previously presented stimulus material when observing a series of stimulus materials (such as letters, graphics, sounds, etc.) to determine whether the two are the same. By increasing the number of stimulus materials, the N-back can have different difficulties. For example, 1-back is to judge the similarity between the current stimulus material and the previous one, and 2-back is to judge the similarity between the current stimulus and the stimulus two positions before, and so on. The running memory task means presenting multiple stimuli to the subject in sequence. After the presentation of the stimuli is completed, the subject is required to recall the last M stimulus materials presented, and the number of stimuli presented in sequence needs to be greater than or equal to M. Since the number of stimuli presented is random and the subject cannot predict it in advance, the subject needs to continuously increase the memory length of the stimulus materials. When the number of stimulus materials exceeds M, if the trial has not ended, then the subject can forget the stimulus materials that appeared first and needs to remember the newly appeared stimulus materials.

[0004] However, for the training tasks in the existing training websites or programs, the stimulus materials are usually extracted from a fixed material library and the materials are separated from the living environment, resulting in the fixation of the stimulus materials. This will limit the user's transfer of the abilities obtained from training to other scenarios, and will also make the training lack challenges and boring, which will further affect the training effect and efficiency. At the same time, it will also lead to low user participation and poor user experience. Summary of the Invention

[0005] In order to solve the above problems existing in the prior art, the embodiments of the present application provide a training method for short-term memory manipulation ability and related products, which can generate the materials required for the training task in real time for the target object participating in the training to constitute the training task to train the target object, thereby enhancing the challenge and interest of the training, further improving the training effect and efficiency, and enhancing the training participation and user experience.

[0006] In a first aspect, an embodiment of the present application provides a training method for short-term memory manipulation ability, the method comprising:

[0007] Determining an initial training level based on the basic information of the target object;

[0008] Determining the material generation conditions for the current training of the target object based on the initial training level and the basic information;

[0009] Inputting the material generation conditions into a large language model to obtain training materials;

[0010] Generating a training task based on the training materials and the initial training level;

[0011] Showing the training task to the target object and obtaining feedback data of the target object on the training task;

[0012] Performing multiple trainings on the target object based on the feedback data and the basic information.

[0013] In a possible implementation manner, the determining the material generation conditions for the current training of the target object based on the initial training level and the basic information includes:

[0014] Obtaining the first material generation conditions corresponding to the initial training level, where the first generation conditions include at least one of the following condition factors: the number of pictures, the interference items of the pictures, and the recognizability of the pictures;

[0015] Determining the second material generation conditions based on the basic information, where the second material generation conditions include at least one of the following condition factors: style and type;

[0016] Taking the first material generation conditions and the second material generation conditions as the material generation conditions.

[0017] In a possible implementation manner, the generating a training task based on the training materials and the initial training level includes:

[0018] Identifying the complexity of each picture in the training materials to obtain the complexity of each picture;

[0019] Determining the display order of the pictures in the training materials based on the complexity of each picture and the initial training level;

[0020] Sequentially putting the pictures in the training materials into the task template corresponding to the initial training level to obtain the training task.

[0021] In a possible implementation, identifying the complexity of each picture in the training material to obtain the complexity of each picture includes:

[0022] Identifying the first quantity of named entities in each picture, where the named entities in each picture include target entities and interfering entities, the target entities are the entities that the target object memorizes during training, and the interfering entities are the entities that interfere with the target object's memory during training;

[0023] Determining the first color of the target entity in each picture, and the first type information of the target entity;

[0024] Determining the second color of the interfering entity in each picture, and the second type information of the interfering entity;

[0025] Based on the first quantity, the first color, the second color, the first type information, and the second type information, determining the complexity of each picture.

[0026] In a possible implementation, the complexity of each picture can be represented by formula ①:

[0027] F i = αS i 2 + βlog(1 + ΔE)+ γlog(1 + ΔX)………①

[0028] Where F i represents the complexity of picture i, S i represents the first quantity of named entities in picture i, ΔE represents the color difference between the first color and the second color of picture i, ΔX represents the type difference between the first type information and the second type information of picture i, and α, β, γ are hyperparameters.

[0029] In a possible implementation, putting each picture in the training material into the task template corresponding to the initial training level in sequence according to the display order to obtain the training task includes:

[0030] Based on the position of each picture in the task template, determining the manipulation mode of each picture, where the manipulation mode includes: update, sorting, calculation, and mental rotation, each position in the task template corresponds to a manipulation mode, and the manipulation mode corresponding to each position in the task template is determined by the training level of the task template.

[0031] In a possible implementation, presenting the training task to the target object and obtaining the feedback data of the target object on the training task includes:

[0032] Show a fixation point to the target object so that the target object fixates on the fixation point;

[0033] After a preset first time interval, show the pictures in the training task to the target object in sequence, where there is a time interval between the display of each picture, and only one picture is shown each time;

[0034] After a preset second time interval, show a question interface to the target object, where the question interface is used to display questions related to the pictures shown in the training task in sequence, and the questions are related to the manipulation mode corresponding to the shown pictures;

[0035] After a preset third time interval, show an answer interface to the target object, where the answer interface is used to display the answer options for the questions;

[0036] Receive the operation data of the target object on the answer options, and use the operation data and the time taken for this training as the feedback data.

[0037] In a second aspect, an embodiment of the present application provides a training device for short-term memory manipulation ability, including:

[0038] A determination module, configured to determine an initial training level based on the basic information of the target object;

[0039] A training module, configured to determine the material generation conditions for the current training of the target object based on the initial training level and the basic information, input the material generation conditions into a large language model to obtain training materials, and generate a training task based on the training materials and the initial training level;

[0040] Show the training task to the target object and obtain the feedback data of the target object on the training task;

[0041] Based on the feedback data and the basic information, conduct multiple trainings on the target object.

[0042] In a third aspect, an embodiment of the present application provides an electronic device, including: a processor, the processor is connected to a memory, the memory is used to store a computer program, and the processor is used to execute the computer program stored in the memory so that the electronic device executes the method as in the first aspect.

[0043] In a fourth aspect, an embodiment of the present application provides a computer-readable storage medium, the computer-readable storage medium stores a computer program, and the computer program enables a computer to execute the method as in the first aspect.

[0044] Fifth aspect, an embodiment of the present application provides a computer program product, which includes a non-transitory computer-readable storage medium storing a computer program, and the computer is operable to cause the computer to execute the method as in the first aspect.

[0045] Implementing the embodiments of the present application has the following beneficial effects:

[0046] In the embodiments of the present application, by obtaining the basic information of the target object, the initial training level suitable for the target object is determined, and then based on the initial training level and the basic information of the target object, the material generation conditions for the current training are determined. Then, the material generation conditions are input into the large language model to generate training materials that fit the current training level and the target object in real time, and corresponding training tasks are generated. Finally, the generated training tasks are presented to the target object, and feedback data of the target object on the training tasks is obtained. Then, based on the feedback data and the basic information, the target object is trained multiple times. Specifically, according to the feedback data and basic parameters of the target object in each training, the most suitable training level can be determined as the level for the next training until multiple trainings are completed. Thus, for the basic information of the target object participating in the training, the training level of the training task can be determined in real time, and the materials required for the training task can be generated to constitute the training task to train the target object, thereby enhancing the challenge and fun of the training. At the same time, based on the feedback data of the target object in the training, combined with its basic information, the next level is determined in real time, and then a more accurate training level is obtained to train the target object, so that each level received by the target object can play a good training role for the target object, thereby further improving the training effect and efficiency, and enhancing the training participation and user experience. BRIEF DESCRIPTION OF THE DRAWINGS

[0047] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the following will briefly introduce the drawings required for the description of the embodiments. Obviously, the following drawings are some embodiments of the present application. For those of ordinary skill in the art, without creative efforts, other drawings can also be obtained based on these drawings.

[0048] Figure 1 Schematic diagram of a training system for short-term memory manipulation ability provided by an embodiment of the present application;

[0049] Figure 2 Schematic flowchart of a training method for short-term memory manipulation ability provided by an embodiment of the present application;

[0050] Figure 3 Schematic diagram of a material picture with interference items provided by an embodiment of the present application;

[0051] Figure 4 A schematic diagram of a material picture under different recognizabilities provided by an embodiment of the present application;

[0052] Figure 5 A schematic diagram of a material picture generated based on material text provided by an embodiment of the present application;

[0053] Figure 6 Another schematic diagram of a material picture generated based on material text provided by an embodiment of the present application;

[0054] Figure 7 A schematic diagram of an updated manipulation training task provided by an embodiment of the present application;

[0055] Figure 8 A schematic diagram of a sorting manipulation training task provided by an embodiment of the present application;

[0056] Figure 9 A schematic diagram of a calculation manipulation training task provided by an embodiment of the present application;

[0057] Figure 10 A schematic diagram of a mental rotation material picture provided by an embodiment of the present application;

[0058] Figure 11 It is a functional module composition block diagram of a training device for short-term memory manipulation ability proposed by an embodiment of the present application;

[0059] Figure 12 It is a schematic structural diagram of an electronic device proposed by an embodiment of the present application. Specific embodiments

[0060] Next, the technical solutions in the embodiments of the present application will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present application. Obviously, the described embodiments are part of the embodiments of the present application, rather than all of the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present application without creative efforts shall fall within the protection scope of the present application.

[0061] The terms "first", "second", "third", "fourth", etc. in the specification, claims and drawings of the present application are used to distinguish different objects, rather than to describe a specific order. In addition, the terms "comprising" and "having" and any variations thereof are intended to cover non-exclusive inclusion. For example, a process, method, system, product or device that includes a series of steps or units is not limited to the listed steps or units, but optionally further includes steps or units not listed, or optionally further includes other steps or units inherent to these processes, methods, products or devices.

[0062] As used herein, the term "embodiment" means that a particular feature, result, or characteristic described in connection with an embodiment may be included in at least one embodiment of the present application. The phrase appears in various places in the specification and does not necessarily refer to the same embodiment, nor is it an independent or alternative embodiment mutually exclusive with other embodiments. Those skilled in the art will explicitly and implicitly understand that the embodiments described herein may be combined with other embodiments.

[0063] First, refer to Figure 1 , Figure 1 which is a schematic diagram of a training system for short-term memory manipulation ability provided for an embodiment of the present application.

[0064] Exemplarily, the training system may include a training device, a processing device, and a database. Among them, the training device may be an operable device with a display function such as a smart phone (such as an Android phone, an iOS phone, a Windows Phone), a tablet computer, a personal digital assistant, a laptop computer, a mobile Internet device MID (Mobile Internet Devices), a robot, or a wearable device, etc. The present application does not make specific limitations thereto. The processing device may be a server. For example, it may be an independent server or a cloud server providing basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communications, middleware services, domain name services, security services, content delivery network (CDN), and big data and artificial intelligence platforms. The present application does not make specific limitations thereto. The database may also be a server or a memory providing data storage services. For example: a read-only memory (ROM) or other types of static storage devices that can store static information and instructions, a random access memory (RAM) or other types of dynamic storage devices that can store information and instructions, or an electrically erasable programmable read-only memory (EEPROM), a compact disc read-only memory (CD-ROM), or other optical disc storage, optical disc storage (including compact discs, laser discs, optical discs, digital versatile discs, Blu-ray discs, etc.), magnetic disk storage media, or other magnetic storage devices, or any other medium that can be used to carry or store desired program code in the form of instructions or data structures and can be accessed by a computer. The present application also does not make specific limitations thereto.

[0065] Specifically, the trainee can initiate a training request through the training device. After receiving the training request, the training device obtains the identity identifier of the trainee. For example, the identity identifier of the trainee can be obtained through login ID, facial recognition, fingerprint recognition, etc. Then, the training device sends the identity identifier and the training request to the processing device. The processing device obtains the personal parameters of the trainee from the database according to the identity identifier, and determines the initial training level of the trainee based on the personal parameters. Then, the processing device obtains the corresponding materials from the database according to the initial training level, and generates a first training topic to be sent to the training device for display to the trainee through the training device. The trainee can train by operating the training device in response to the displayed first training topic. The training device collects the data of the trainee during the training process as the feedback data of the trainee on the first training topic. Then, the training device sends the feedback data to the processing device, and the processing device determines the next training level of the trainee. Thus, by repeating the above processes of obtaining materials, generating training topics, displaying training topics, obtaining feedback data, and determining the next training level, the trainee is trained multiple times.

[0066] Thus, through the personal parameters, the historical data of the people similar to the trainee in the previous training is obtained as a reference for determining the level. At the same time, combined with the feedback data, when determining the next level, the personal characteristics and group characteristics of the trainee can be taken into account simultaneously, and then a more accurate training level can be obtained to train the trainee, improving the training efficiency and training effect.

[0067] Secondly, it should be noted that the training method provided by the embodiment of the present application can be applied to scenarios where ability training can be carried out, such as short-term information manipulation ability, flexibility ability, understanding ability, etc. Hereinafter, taking the training scenario of short-term information manipulation ability as an example, a training method proposed by the present application will be described. The training methods in other scenarios are similar to those in the training scenario of short-term information manipulation ability and will not be elaborated here.

[0068] In this embodiment, the short-term information manipulation ability involves short-term information storage, which refers to a memory system that stores a small amount of information within a relatively short period of time. Its characteristics are: (1) The encoding is mainly verbal-auditory encoding, and there is also visual and semantic encoding; (2) The capacity is limited, generally 7±2 chunks; (3) The retention time is short. If the information is not rehearsed in time, it can only be maintained for about 5 seconds to 1 minute; (4) The extraction method of short-term memory is the way of complete serial scanning.

[0069] In this embodiment, the basic process includes three steps, namely, encoding, maintaining, and manipulating the stimulus materials. Generally, the ability to manipulate short-term information is trained. However, in existing programs for training the manipulation ability of short-term memory, usually only one type of manipulation is performed. The most common manipulation methods include updating (such as the N-back task and the active memory task). However, in this embodiment, in addition to this, there are also manipulations such as sorting the stimulus materials, performing calculations such as addition, subtraction, multiplication, and division on the number of stimulus materials, and manipulating the two-dimensional or three-dimensional rotation of the stimulus material objects by using imagination.

[0070] The existing short-term memory manipulation training programs have the following defects:

[0071] (1) In the existing short-term memory manipulation training programs, there is only one type of manipulation mode throughout the process. For example, in the Memory Serves game designed by the cognitive training company Lumosity, the only manipulation method for the subjects in this game is updating, and the subjects need to remember the colors of the handbags that have appeared. However, in daily life, the manipulation of temporarily stored things may be diverse. When shopping, when seeing several items to buy, by comparing the prices of the items, one needs to calculate the prices of the items in the mind, then sort them according to importance, and finally select the items with high priority for purchase. In this process, three manipulation methods, namely, updating, mathematical calculation, and sorting, are required. Therefore, a training program that is more adaptable to real-life scenarios should exercise different manipulation methods.

[0072] (2) The stimulus materials involved in the existing short-term memory manipulation training programs are relatively simple and do not change throughout the training process. When the stimulus materials in the entire training program are fixed, this will limit the transfer of the abilities obtained from training by the users to other real-life scenarios, and it will also make the training lack challenge and be boring, which will in turn affect the user participation. The richness of the stimulus materials improves the transfer range of the training effect. Multiple factors such as the complexity, abstractness, and background interference of the stimulus materials can affect the difficulty of encoding the materials, thereby affecting the training performance of the practitioners. For example, in the memory manipulation game Follow That Frog designed by the cognitive training company Lumosity, throughout the game, the stimulus material is a single frog jumping on different lotus leaves. Such fixed stimulus materials can easily lead to the loss of training interest of the users and are difficult to transfer to different scenarios in real life.

[0073] (3)In the existing short-term memory manipulation training programs, different difficulty levels are set for the training tasks. The higher the difficulty level of the training task, the more cognitive resources the user needs to consume. In the current difficulty level setting, since the stimulus materials in the training tasks are relatively single and contain too few elements, the available difficulty levels are correspondingly fewer. For example, in the memory manipulation game Follow That Frog designed by the cognitive training company Lumosity, the changes in different difficulty levels are reflected in two dimensions: the quantity to be remembered and the speed of stimulus presentation, without involving other elements.

[0074] (4)In the existing short-term memory manipulation training programs, users can only start from the same difficulty level and challenge level by level following the fixed difficulty levels, regardless of whether the practitioner can adapt to the next level or whether the practitioner is interested in the training content. For example, in the game Memory Serves designed by the cognitive training company Lumosity, the difficulty change in the game is mainly reflected in the types of colors of the handbags that the practitioner needs to remember and increases according to the fixed levels. It is impossible to dynamically adjust the difficulty level of the user's training task according to the user's specific performance in the test, and inappropriate difficulty is also difficult to produce the best training effect.

[0075] In this embodiment, based on the defects of the existing short-term memory manipulation training programs, a training method is designed. Specifically, this is a training method for manipulating short-term memory. This training method has the advantages of diverse manipulation methods, personalized stimulus materials, diverse stimulus materials, diversified difficulty level settings, and adaptive adjustment of difficulty levels, which will be described in detail below.

[0076] Refer to Figure 2 , Figure 2 FIG. is a schematic flowchart of a training method for short-term memory manipulation ability provided by an embodiment of the present application. This method is applied to the training system in the above embodiment. This training method may include the following steps:

[0077] 201: Determine an initial training level based on the basic information of the target object.

[0078] In this embodiment, when the target object participates in the training, an account needs to be logged in on the training device. The basic information for this login operation can be scanned by a corresponding identity card through the Near Field Communication (NFC) method, such as scanning an ID card, student ID card, employee ID card, etc. for login; or, it can be logged in by biometric recognition methods, such as facial recognition, fingerprint recognition, red membrane recognition, etc.; or, it can be logged in by the account password method. The present application does not limit this.

[0079] In this embodiment, after the login is completed, the training system will display a prompt message to the target user, prompting the target object that its basic information will be obtained for training level recommendation. If the target object confirms this information, it authorizes the training system to obtain the permission of its basic information. Subsequently, the training device can obtain the basic information corresponding to this account. This prompt message can be popped up only when it is obtained for the first time to reduce the operations of the target object and further improve the user experience.

[0080] In this embodiment, the basic information may include age, gender, preference information, basic cognitive ability, and historical training data. Among them, the preference information can be obtained by the target object participating in a preference test provided by the training system before the first training, and the basic cognitive ability can be measured by the performance of the target object in the N-back task provided by the training system before the first training. Of course, other methods in the art that can obtain the preference information and basic cognitive ability of the target object can be applied to this application, and this application does not limit this.

[0081] Thus, based on the age, gender, basic cognitive ability, and historical training data in the basic information, the initial training level of the target object can be determined. Exemplarily, the population corresponding to the target object can be determined based on age and gender, for example: young women, teenage boys, etc., and then the initial training level range of this population when participating in training can be obtained. Then, based on the basic cognitive ability and historical training data of the target object, a suitable training level is determined within the initial training level range as the initial training level of the target object.

[0082] In this embodiment, if the target object refuses to authorize or no account is logged in on the training device, the preset lowest level will be defaulted as the initial training level.

[0083] 202: Determine the material generation conditions for the current training of the target object based on the initial training level and basic information.

[0084] In this embodiment, in order to improve the training effect, the setting of the training level can be based on the following three principles:

[0085] 1) The number of training levels is appropriate. Design an appropriate number and type of training levels so that the target object can match different training levels according to its own level and gradually challenge its ability limit, thereby achieving continuous improvement of cognitive ability;

[0086] 2) The difficulty difference between training levels is obvious. Make there be an obvious difference between each training level, but also retain a certain relevance, so that the target object can clearly understand the content, requirements, and rewards involved in each training level and gradually complete the next level based on the original level;

[0087] 3) Gradual increase in the training level enables the target object to feel the increasing difficulty during the game progress, and gradually improves the ability of the target object in small steps.

[0088] Compared with the existing cognitive training tasks that mostly change the training level from a single dimension, the adjustment of the training level is rather rigid, resulting in that it is very difficult for the target object to make a breakthrough after quickly reaching a certain level. In this embodiment, the training level will be set from multiple conditional factors. Specifically, these multiple conditional factors are respectively the number of materials, the presentation time of information, the presence or absence of interference items, and the recognition degree of material pictures.

[0089] Specifically, the number of materials refers to the number of materials presented successively in a single training task. This number starts from 1 and gradually increases. According to the short-term memory storage capacity of the healthy population, the maximum number of materials is 12. Then this conditional factor has a total of 12 levels.

[0090] The presentation time of information refers to the presentation time of each material in a single training task. This time starts from 1.5 seconds and decreases successively by 0.5 seconds each time. The shortest presentation time is 0.5 seconds. Then this conditional factor has a total of 3 levels.

[0091] The presence or absence of interference items refers to whether there are factors that interfere with the target object's memory of the materials in each material of a single training task. Exemplarily, for the materials with interference items, 3 objects of the same type but different contents will be generated during generation. Among them, the target object needs to ignore the objects on the left and right sides of the 3 objects and mainly remember the object in the middle. For example, as Figure 3 shown, there are three fruits in a material picture, namely a green pear, a red apple, and a green apple. Among them, the green pear and the green apple on the left and right sides are interference items, and the target object needs to ignore the green pear and the green apple on the left and right sides and mainly remember the red apple in the middle. Then this conditional factor has a total of 2 levels.

[0092] The recognition degree of material pictures refers to the contrast between the objects and the background in each material of a single training task. Specifically, the encoding situation of the material pictures will affect the subsequent manipulation situation. When the materials for memory and manipulation require more attention and visual processing, it will be more difficult for an individual to manipulate the materials. Based on this, when the contrast between the objects and the background in the materials is lower and the similarity between the objects and the background is higher, the recognition degree of the material pictures will decrease, and the target object will need to spend higher cognitive ability to remember them. Exemplarily, as Figure 4As shown, there are three pictures of red apples, but with white, pink, and red backgrounds respectively. Then, as the color of the background gets closer to the color of the object, that is, when the contrast decreases, the gap between the object and the background becomes more blurred, and then the recognition rate of the material pictures gradually decreases, while the cognitive ability required for memory gradually increases. Then this conditional factor has a total of 3 levels, namely high, medium, and low recognition rate.

[0093] Thus, in this embodiment, by combining the above 4 conditional factors with each other, a series of training levels can be set. Specifically, according to the number of conditional factors there are 12 levels, the presentation time of the conditional factor information has 3 levels, whether there are interference items in the conditional factor has 2 levels, and the recognition rate of the material pictures of the conditional factor has 3 levels, then 216 training levels can be combined. Among these training levels, each training level corresponds to a combined conditional factor. For example: for training level 1, the corresponding conditional factor is 3 materials, 1 - second presentation time, with interference items and medium recognition rate; for training level 1, the corresponding conditional factor is 5 materials, 1 - second presentation time, without interference items and high recognition rate.

[0094] Based on this, in this embodiment, first, the conditional factors related to the material pictures in the conditional factors corresponding to the initial training level can be obtained as the first material generation conditions. The first generation conditions can include at least one of the following conditional factors: the number of pictures, the interference items of the pictures, and the recognition rate of the pictures. Thus, based on the first material generation conditions, it can be determined how many material pictures need to be generated in this training task, whether each material picture contains interference items, and the recognition rate of each material picture.

[0095] Meanwhile, in this embodiment, a variety of selectable material themes are set for different target objects. Specifically, they can be divided into two major categories: figurative and abstract. Exemplarily, there are a total of 16 material themes for figurative materials, which can be further divided into 3 categories. Specifically: for the life category, it includes 8 material themes such as clothing, tools, daily necessities, food, toys, fruits and vegetables, flowers and plants, and sports supplies; for the travel category, it includes 4 material themes such as transportation tools, road signs, scenery, and buildings; for the natural category, it includes 4 material themes such as land animals, marine organisms, astronomical bodies, and natural phenomena. There are 3 material themes for abstract materials, including numbers, symbols, and geometric figures.

[0096] Thus, the materials of the training task are associated with the events in the scenarios contacted in real life, realizing the life - orientation and diversification of training, avoiding the problems in existing training such as single materials, being decoupled from real life, resulting in users losing interest in training, and being difficult to transfer in different scenarios of real life.

[0097] In this embodiment, in order to increase the emotional investment and cognitive participation of the target object in training, material pictures of topics that the target object is interested in can be generated according to the preferences of the target object for training. Exemplarily, if the target object is an astronomy enthusiast, astronomical star themes related to astronomical knowledge can be selected to generate training materials; if the target object is interested in humanistic landscapes, landscape themes and architectural themes can be selected to generate training materials.

[0098] Based on this, in this embodiment, the preference situation of the target object can be determined based on the preference information in the basic information as the second material generation condition, and the second material generation condition includes at least one of the following condition factors: style and type, that is, the picture style preferred by the target object, and the type of picture content, or the theme.

[0099] Finally, in this embodiment, the first material generation condition and the second material generation condition can be combined together as the final material generation condition. For example: the first material generation condition is: 3 materials, no interference items, and high recognition; the second material generation condition is: fruit and vegetable theme and realistic style. Then the combined material generation condition is: 3 materials, no interference items, high recognition, fruit and vegetable theme, and realistic style.

[0100] 203: Input the material generation condition into the large language model to obtain training materials.

[0101] In this embodiment, the material generation condition can be input into a preset paradigm text to obtain a material generation text, and then the material generation text can be input into a pre-trained large language model to obtain training materials.

[0102] Exemplarily, the paradigm text can be as follows:

[0103] Generate [mask] pictures of [mask] style, requiring the content of the pictures to be items of [mask] type, there are [mask] interference items in the pictures, and the recognition of the pictures is [mask].

[0104] Among them, [mask] respectively corresponds to a condition factor in the material generation condition, and the corresponding condition factor is filled into the corresponding [mask] in this method text to obtain the material generation text. Exemplarily, continuing with the above material generation condition, that is: 3 materials, no interference items, high recognition, fruit and vegetable theme, and realistic style, then the following material generation text can be obtained:

[0105] Generate 3 pictures of realistic style, requiring the content of the pictures to be items of fruit and vegetable type, there are no interference items in the pictures, and the recognition of the pictures is high.

[0106] Based on this, such conditions can generate as Figure 5The 3 material pictures shown

[0107] Exemplarily, if the material generation conditions are: 3 materials, with interference items, medium recognition, fruit and vegetable theme, and realistic style, the following material generation text can be obtained:

[0108] Generate 3 pictures in realistic style, requiring the content of the pictures to be fruit and vegetable items, with interference items in the pictures and medium recognition of the pictures.

[0109] Based on this, such 3 material pictures as shown Figure 6 can be generated under this condition.

[0110] 204: Generate a training task based on the training materials and the initial training level.

[0111] In this embodiment, multiple manipulation modes are set for the training task, which may include: update, sorting, calculation, and mental rotation. Since each manipulation method exercises different cognitive abilities, the target object can choose any one of the manipulation methods each time it conducts the training of this program. Each method is introduced by way of example below.

[0112] a) Update: The task instruction requires the target object to correctly recall the last N materials that appeared. Multiple materials are successively presented on the screen, and the number of presented materials is random, but equal to or more than N. Since the subject does not know how many materials will be presented, it is necessary to update the memory content, focusing on remembering the last N materials, and the materials before N can be forgotten. Exemplarily, as Figure 7 shown Figure 7 shows the process of a training task that requires positive recall of the last 2 pictures. In this training task, N is 2 pictures; first, a rule interface is shown to the target object, and relevant rules of the training and / or operation instructions of the user will be displayed in this interface. Then a fixation point interface is shown to the target object to remind the user to concentrate, and this interface can be shown for 1 s. Then, 3 material pictures will be successively shown to the target object, and a blank interface can be inserted as an interval between each material picture, and the display time of each material picture and the blank interface can also be 1 s. Then, a question interface is shown, such as Figure 7 the question shown in it can be "What are the last 2 pictures?", and then relevant options are shown, and the target object needs to make a judgment, and finally the feedback on the correctness of the judgment is shown. Among them, the display time of the question interface can be 1 s, the display time of the judgment interface can be 3 s, and the display time of the feedback interface can be 0.5 s.

[0113] b) Sorting: The task instruction requires the target object to correctly recall the last N materials that appeared and their presentation order; at the end of the training task, let the target object recall the materials in the forward order or in the reverse order according to the presentation order. Exemplarily, asFigure 8 As shown Figure 8 shows the process of a training task that requires positive recall of the order of the last 3 pictures. In this training task, N is 3 pictures; first, a rule interface is shown to the target object, and relevant rules of the training and / or operation instructions for the user will be displayed in this interface. Then, a fixation point interface is shown to the target object to remind the user to concentrate, and this interface can be shown for 1 s. Then, 3 material pictures will be shown to the target object in sequence, and a blank interface can be inserted as an interval between each pair of material pictures, and the display time of each material picture and the blank interface can also be 1 s. Then, a question interface is shown, such as Figure 8 The question shown in it can be "What is the order of the last 3 pictures?", then relevant options are shown, and the target object needs to make a judgment, and finally the feedback on the correctness of the judgment is shown. Among them, the display time of the question interface can be 1 s, the display time of the judgment interface can be 3 s, and the display time of the feedback interface can be 0.5 s.

[0114] c) Calculation: The task instructions require the target object to perform addition and subtraction operations on the N materials that appear. N material pictures are presented continuously on the screen, and there may be duplicate material pictures among them. At the end of the training task, the target object is required to calculate the number of different material pictures, such as Figure 9 As shown Figure 9 shows the process of a training task of a calculation operation. In this training task, N is 5 pictures; first, a fixation point interface is shown to the target object to remind the user to concentrate, and this interface can be shown for 1 s. Then, 5 material pictures will be shown to the target object in sequence, and a blank interface can be inserted as an interval between each pair of material pictures, the display time of each material picture can be 1 s or 0.5 s, and the display time of the blank interface can be 1 s. Then, the fixation point interface is shown to the target object again, and the display time of this interface can be 1 s. Then, a question interface is shown, such as Figure 9 The question shown in it can be "How many apples are there?", then relevant options are shown, and the target object needs to make a judgment, and finally the feedback on the correctness of the judgment is shown. Among them, the display time of the question interface can be 1 s or 0.5 s, the display time of the judgment interface can be 3 s, and the display time of the feedback interface can be 0.5 s.

[0115] d) Mental rotation: The task instructions require the target object to perform two-dimensional or three-dimensional rotation on the presented material pictures, such as Figure 10 As shown. At the end of the training task, the target object is required to select the correct rotated image.

[0116] For the above basic process, in this embodiment, for training tasks with different manipulation types and training levels, the presentation order of the materials in the tasks can be different. For example, for tasks with a higher training level, pictures with higher complexity can be presented first, so that the memory ability required for this picture is higher, because the earlier it is presented, the longer the time required for memory, and it will also be interfered by the pictures presented later, thus making it more difficult.

[0117] Based on this, in this embodiment, after obtaining the training materials, the complexity of each picture in the training materials can be identified first to obtain the complexity of each picture. Specifically, the first quantity of named entities in each picture can be identified. The named entity can refer to all entities in the picture, including target entities and interfering entities. Among them, the target entity can be understood as the entity that the target object in the training needs to remember, and the interfering entity can be understood as the entity that interferes with the target object's memory in the training. Taking Figure 3 the material picture shown as an example, which contains 3 entities: a green pear, a red apple, and a green apple, then these 3 entities are all the named entities of this picture, and the first quantity of named entities in this picture is 3. At the same time, in this picture, the target object is required to remember the red apple in the picture, so the target entity is the red apple in the middle, and the interfering entities are the green pear on the left and the green apple on the right.

[0118] Then, determine the first color of the target entity in each picture, as well as the first type information of the target entity. Among them, the first type information can be the classification to which the target entity belongs, or it can also be the name of the target entity. Exemplarily, continuing with the above Figure 3 example, the first color is red, and the first type information can be fruit, or it can also be apple.

[0119] Similarly, it is also necessary to determine the second color of the interfering entity in each picture, as well as the second type information of the interfering entity. Exemplarily, continuing with the above Figure 3 example, the second color is green, and the second type information can be fruit, or it can also be pear and apple.

[0120] In a possible implementation manner, when there is only a target entity in the picture, the background can be used as the interfering entity. At this time, the second color is the color of the background, and the second type information is "background".

[0121] Then, in this embodiment, the complexity of each picture can be determined based on the first quantity, the first color, the second color, the first type information, and the second type information. Exemplarily, in this embodiment, the complexity of a picture can refer to the memorization difficulty of the picture. The greater the memorization difficulty of the picture, the greater its complexity. Moreover, the more the number of named entities in a picture, that is, the greater the first quantity, the greater the difficulty of memorizing the picture. At the same time, if the colors of the target entities in the picture are more similar to the colors of the interfering entities and the types are more similar, then more effort is required to distinguish the interfering entities from the target entities, and the memorization difficulty of the picture is also greater, and the corresponding complexity is also greater.

[0122] Based on this, in this embodiment, a method for determining the complexity of each picture based on the first quantity, the first color, the second color, the first type information, and the second type information is provided. Specifically, the complexity of each picture can be expressed by the following formula ②:

[0123]

[0124] Where, F i represents the complexity of picture i, S i represents the first quantity of named entities in picture i, ΔE represents the color difference between the first color and the second color of picture i, ΔX represents the type difference between the first type information and the second type information of picture i, and α, β, γ are hyperparameters.

[0125] In this embodiment, the color difference ΔE between the first color and the second color of picture i can be based on the distance between the first color and the second color in the Commission Internationale de l'Eclairage Lab (CIELab) color space. Specifically, it can be expressed by the following formula ③:

[0126]

[0127] Where, (L i1 , a i1 , b i1 ) represents the color value of the first color of picture i in the Lab color space, and (L i2 , a i2 , b i2 ) represents the color value of the second color of picture i in the Lab color space.

[0128] ​Furthermore, in this embodiment, the type difference ΔX between the first type of information and the second type of information of picture i can be determined based on the distance between the first type of information and the second type of information in the word vector space. Specifically, first, the first type of information and the second type of information can be respectively subjected to word embedding processing to obtain the word vector corresponding to the first type of information: T = [t1, t2, …, tm, …, tn], and the word vector corresponding to the first type of information: U = [u1, u2, …, um, …, un], where m = 1, 2, …, n.

[0129] Then, since the closer the types of the target entity and the interfering entity are, the higher the difficulty of discrimination is, ΔX can be represented by the following formula ④:

[0130]

[0131] where TXU represents the inner product of the word vector T and the word vector U, and || is the modulus symbol, |T| represents the modulus of the word vector T, and |U| represents the modulus of the word vector U.

[0132] Thus, in this embodiment, after determining the complexity of each picture, the display order of the pictures in the training material can be determined based on the complexity of each picture and the initial training level. Then, the pictures in the training material are sequentially placed into the task template corresponding to the initial training level according to the display order to obtain a training task.

[0133] In addition, while placing each picture into the corresponding position in the task template corresponding to the initial training level, the manipulation mode of the picture placed in this position can also be determined based on this position. As can be seen from the above, the manipulation modes include: update, sorting, calculation, and mental rotation. Therefore, when setting the task template corresponding to the training level, the manipulation mode corresponding to each position in the training template can be set at the same time. Then, when the picture is placed in this position, relevant questions can be generated based on the manipulation mode corresponding to this position to obtain a training task.

[0134] 205: Show the training task to the target object and obtain the feedback data of the target object on the training task.

[0135] In this embodiment, first, a fixation point can be shown to the target object to enable the target object to fixate on the fixation point. The fixation point can be displayed in the center of the interface to help the subject focus. After a preset first time interval, the pictures in the training task are sequentially shown to the target object. During the showing process, there is a time interval between the showings of each picture, and only one picture is shown each time. After a preset second time interval, a question interface is shown to the target object. The question interface is used to display questions related to the pictures shown in the training task in sequence. The questions are related to the manipulation mode corresponding to the shown pictures. After a preset third time interval, an answer interface is shown to the target object. The answer interface is used to display the answer options for the questions. Finally, the operation data of the target object on the answer options is received, and the operation data and the time taken for this training are used as feedback data to complete the training task at this training level.

[0136] It should be noted that in this embodiment, one training can include at least one trial. Among them, when one training includes one trial, the training process is as shown above. When one training includes multiple trials, corresponding training materials can be generated for each trial based on the material generation conditions, and then a training task can be further generated for each trial. Then, in one training, the target object can sequentially complete the training tasks of each trial, and the operation data of the target object on the answer options and the time taken for each trial are used as the feedback data for this training.

[0137] 206: Based on the feedback data and basic information, conduct multiple trainings on the target object.

[0138] In this embodiment, the feedback data can be the answering accuracy rate and time taken by the target object in the training task. Thus, the feedback data can indicate the performance of the target object at this training level. Therefore, based on the feedback data and the basic information of the target object, the next training level of the target object can be determined. Similarly, after the target object completes the training at the next training level, the next-next training level can also be determined based on the feedback data of the target object in the training task at the next training level and combined with its basic information. In this way, the target object is trained multiple times in a cycle until the target object reaches the highest training level or when the next training level cannot be determined, and the training ends.

[0139] Specifically, the i-th training can include the following processes:

[0140] First, based on the feedback data Cj and basic information, determine the target training level Dj. When j = 1, the feedback data C1 is the feedback data of the target object on the training task;

[0141] Then, based on the target training level Dj and the basic information, determine the material generation condition Gj. Input the generation condition Gj into the large language model to obtain the training material Hj. Generate the training task Kj based on the training material Hj and the target training level Dj, display the training task Kj to the target object, and obtain the feedback data Rj of the target object on the training task Kj;

[0142] Finally, use the feedback data Rj as the feedback data Cj+1 for the (j + 1)-th training, and conduct the (j + 1)-th training until multiple trainings are completed.

[0143] It should be noted that in this embodiment, other alternative solutions are also applicable to this application, and this application does not limit this.

[0144] Specifically, four manipulation methods are listed in this application, namely dynamically updating (expanding and reducing) memory content, sorting content, performing mathematical calculations on content, and performing mental rotation on content. There can also be other manipulation methods, such as establishing connections between information content relationships, such as matching, grouping, preference rating, and other advanced manipulation methods. Moreover, these manipulation methods in this application can be combined, for example, updating and calculating simultaneously; or updating and rotating simultaneously, etc.

[0145] The materials in this application take pictures as an example, but the stimulus materials can appear in various forms, such as sounds, etc. Moreover, the material library can also be increased, not limited to the libraries listed in the specification.

[0146] The program content of this application can also be: presenting pictures: the number of materials presented on the pictures can change arbitrarily without limitation. Interference information: It can be information unrelated to the picture materials to be memorized, such as text, mathematical calculations, sounds, images, etc., and the appearance location can be in the presented pictures or in the delay time. Delay time: It can be any time without limitation, etc. Judgment rules: The judgment rules can be changed, for example, judging that all content in the picture is incorrect, or only one content is incorrect while the others are correct.

[0147] The training program process in this application can have changes. For example, in the process design, two manipulation methods can appear, that is, first 3 pictures appear and need to be sorted, and then a geometric figure appears and needs to be mentally rotated, etc.

[0148] The difficulty level of this application is determined by condition factors such as manipulation methods, material libraries, automatically generated picture materials (quantity, interference, discrimination), and presentation time, but it is not limited to being determined only by these condition factors. The number of condition factors can be increased or decreased, and the levels of condition factors can also be increased or decreased.

[0149] In addition to the ACER algorithm and the conditional variational autoencoder, other machine learning algorithms and image generation algorithms can also be used. For example, the Q-learning algorithm can be used to adaptively design the difficulty level for each practitioner, and the conditional generative adversarial network CGVAN can be used to achieve the adaptation of image materials.

[0150] In summary, in the training method provided by the present invention, by obtaining the basic information of the target object, the initial training level suitable for the target object is determined, and then based on this initial training level and the basic information of the target object, the material generation conditions for the current training are determined. Then, the material generation conditions are input into the large language model to generate training materials that fit the current training level and the target object in real time, and corresponding training tasks are generated. Finally, the generated training tasks are presented to the target object, and the feedback data of the target object on the training tasks is obtained. Then, based on this feedback data and the basic information, the target object is trained multiple times. Specifically, the most suitable training level can be determined as the level for the next training according to the feedback data and basic parameters of the target object in each training until multiple trainings are completed. Thus, the training level of the training task can be determined in real time according to the basic information of the target object participating in the training, and the materials required for the training task can be generated to constitute the training task to train the target object, thereby enhancing the challenge and interest of the training. At the same time, based on the feedback data of the target object in the training, combined with its basic information, the next level is determined in real time, and then a more accurate training level is obtained to train the target object, so that each level received by the target object can play a good training role for the target object, thereby further improving the effect and efficiency of the training, and enhancing the participation and user experience of the training.

[0151] The above mainly introduces the solutions of the embodiments of the present application from the perspective of the method side. It can be understood that in order to implement the above functions, the training device includes the corresponding hardware structure and / or software module for executing each function. Those skilled in the art should easily realize that, combined with the units and algorithm steps of each example described in the embodiments disclosed in this article, the present application can be implemented in the form of hardware or a combination of hardware and computer software. Whether a certain function is executed in the form of hardware or computer software driving hardware depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered to exceed the scope of the present application.

[0152] The embodiments of the present application can divide the functional units of the terminal device according to the above method examples. For example, each functional module can be divided corresponding to each function, or two or more functions can be integrated into one functional module. The above integrated module can be implemented in the form of hardware or in the form of a software program module. It should be noted that the division of modules in the embodiments of the present application is illustrative, only a logical function division, and there may be other division methods in actual implementation.

[0153] In the case of adopting an integrated module, Figure 11 is a block diagram of the functional modules of a training device for short-term memory manipulation ability proposed by the embodiments of the present application. Among them, the training device 1100 includes a determination module 1101 and a training module 1102.

[0154] In this embodiment, the determination module 1101 and the training module 1102 can be modules for receiving and processing signals, information, etc. or determining a monitoring mechanism, and no specific limitation is made thereto.

[0155] In this embodiment, the training device 1100 may further include a storage module for the computer program code or instructions executed by the training device 1100. Among them, the storage module can be a memory.

[0156] In this embodiment, the training device 1100 can be a chip or a chip module.

[0157] In this embodiment, the determination module 1101 and the training module 1102 can be integrated in a communication module. Among them, the communication module can be a communication interface, a transceiver, a transceiver circuit, etc.

[0158] In this embodiment, the determination module 1101 and the training module 1102 can be integrated in a processor.

[0159] It should be noted that the processor can be a baseband processor, a baseband chip, a central processing unit (CPU), a general-purpose processor, a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field programmable gate array (FPGA), or other programmable logic devices, transistor logic devices, hardware components, or any combination thereof. It can implement or execute various exemplary logical blocks, modules, and circuits described in connection with the disclosure of the present application. The processing module can also be a combination that implements computing functions, such as a combination including one or more microprocessors, a combination of a DSP and a microprocessor, etc.

[0160] In this embodiment, the training device 1100 is used to execute any step performed by a terminal device / chip / chip module, etc. in the above method embodiment.

[0161] Specifically, the determination module 1101 and the training module 1102 are used to execute any step in the above method embodiment, and when performing actions such as sending, other modules can be selectively called to complete the corresponding operations. Details are described below.

[0162] The determination module 1101 is used to determine an initial training level based on the basic information of the target object;

[0163] The training module 1102 is used to determine the material generation conditions for the current training of the target object based on the initial training level and the basic information, input the material generation conditions into a large language model to obtain training materials, and generate a training task based on the training materials and the initial training level;

[0164] Display the training task to the target object and obtain feedback data of the target object on the training task;

[0165] Based on the feedback data and the basic information, perform multiple trainings on the target object.

[0166] In this embodiment, in terms of determining the material generation conditions for the current training of the target object based on the initial training level and the basic information, the training module 1102 is specifically used for:

[0167] Obtain the first material generation conditions corresponding to the initial training level, where the first generation conditions include at least one of the following condition factors: the number of pictures, the interference items of the pictures, and the recognizability of the pictures;

[0168] Determine the second material generation conditions based on the basic information, where the second material generation conditions include at least one of the following condition factors: style and type;

[0169] Use the first material generation conditions and the second material generation conditions as the material generation conditions.

[0170] In this embodiment, in terms of generating a training task based on the training material and the initial training level, the training module 1102 is specifically configured to:

[0171] Identify the complexity of each picture in the training material to obtain the complexity of each picture;

[0172] Determine the display order of the pictures in the training material based on the complexity of each picture and the initial training level;

[0173] Put the pictures in the training material into the task template corresponding to the initial training level in sequence according to the display order to obtain the training task.

[0174] In this embodiment, in terms of identifying the complexity of each picture in the training material to obtain the complexity of each picture, the training module 1102 is specifically configured to:

[0175] Identify the first quantity of named entities in each picture, where the named entities in each picture include target entities and interfering entities, the target entities are the entities that the target object in training memorizes, and the interfering entities are the entities that interfere with the target object's memory in training;

[0176] Determine the first color of the target entity in each picture, and the first type information of the target entity;

[0177] Determine the second color of the interfering entity in each picture, and the second type information of the interfering entity;

[0178] Determine the complexity of each picture based on the first quantity, the first color, the second color, the first type information, and the second type information.

[0179] In this embodiment, the complexity of each picture can be represented by the following formula ⑤:

[0180] F i =αS i 2 +βlog(1 + ΔE)+γlog(1 + ΔX)………⑤

[0181] where F irepresents the complexity of picture i, S i represents the first quantity of named entities in the picture i, ΔE represents the color difference between the first color and the second color of the picture i, ΔX represents the type difference between the first type information and the second type information of the picture i, and α, β, γ are hyperparameters.

[0182] In this embodiment, in the aspect of putting each picture in the training material into the task template corresponding to the initial training level in the display order to obtain the training task, the training module 1102 is specifically configured to:

[0183] Based on the position of each picture in the task template, determine the manipulation mode of each picture, where the manipulation mode includes: update, sorting, calculation, and mental rotation, each position in the task template corresponds to a manipulation mode, and the manipulation mode corresponding to each position in the task template is determined by the training level of the task template.

[0184] In this embodiment, in the aspect of presenting the training task to the target object and obtaining the feedback data of the target object on the training task, the training module 1102 is specifically configured to:

[0185] Present a fixation point to the target object so that the target object fixates on the fixation point;

[0186] After a preset first time interval, present the pictures in the training task to the target object in sequence, where there is a time interval between the presentations of each picture, and only one picture is presented each time;

[0187] After a preset second time interval, present a question interface to the target object, where the question interface is used to display questions related to the pictures presented in the training task in sequence, and the questions are related to the manipulation mode corresponding to the presented pictures;

[0188] After a preset third time interval, present an answer interface to the target object, where the answer interface is used to display the answer options for the questions;

[0189] Receive the operation data of the target object on the answer options, and use the operation data and the time used for this training as the feedback data.

[0190] Refer to Figure 12 , Figure 12 is a schematic structural diagram of an electronic device proposed in an embodiment of the present application. Among them, the electronic device 1200 may include a processor 1210, a memory 1220, and a communication bus for connecting the processor 1210 and the memory 1220.

[0191] Optionally, the memory 1220 includes, but is not limited to, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM), or a compact disc read-only memory (CD-ROM). The memory 1220 is used to store the program code executed by the electronic device 1200 and the transmitted data.

[0192] In this embodiment, the electronic device 1200 further includes a communication interface, which is used to receive and send data.

[0193] In this embodiment, the processor 1210 can be one or more CPUs. When the processor 1210 is a single CPU, the CPU can be a single-core CPU or a multi-core CPU.

[0194] In this embodiment, the processor 1210 can be a baseband chip, a chip, a CPU, a general-purpose processor, a DSP, an ASIC, an FPGA, or other programmable logic devices, transistor logic devices, hardware components, or any combination thereof.

[0195] Specifically, the processor 1210 in the electronic device 1200 is used to execute the computer program or instruction 1221 stored in the memory 1220 and perform the following operations:

[0196] Based on the basic information of the target object, determine the initial training level;

[0197] Based on the initial training level and the basic information, determine the material generation conditions for the current training of the target object;

[0198] Input the material generation conditions into the large language model to obtain training materials;

[0199] Based on the training materials and the initial training level, generate a training task;

[0200] Display the training task to the target object and obtain the feedback data of the target object on the training task;

[0201] Based on the feedback data and the basic information, conduct multiple trainings on the target object.

[0202] It should be noted that Figure 12 For the specific implementation of each operation in the above embodiment, reference can be made to the description in the method embodiment shown above, and details will not be repeated here.

[0203] Embodiments of the present application also provide a computer-readable storage medium storing a computer program or instructions, and when the computer program or instructions are executed, the steps described in the above method embodiments are implemented.

[0204] Embodiments of the present application also provide a computer program product including a computer program or instructions, and when the computer program or instructions are executed, the steps described in the above method embodiments are implemented.

[0205] It should be noted that for the above various embodiments, for simplicity of description, they are all expressed as a series of action combinations. Those skilled in the art should know that the present application is not limited by the described order of actions, because some steps in the embodiments of the present application can be performed in other orders or simultaneously. In addition, those skilled in the art should also know that the embodiments described in the specification are all preferred embodiments, and the actions, steps, modules or units involved are not necessarily essential to the embodiments of the present application.

[0206] In the above embodiments, the descriptions of the various embodiments of the present application have their own emphases. For parts not detailed in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0207] The steps of the method or algorithm described in the embodiments of the present application can be implemented in a hardware manner or by a processor executing software instructions. The software instructions can be composed of corresponding software modules, and the software modules can be stored in RAM, flash memory, ROM, EPROM, electrically erasable programmable read-only memory (EEPROM), registers, hard disk, removable hard disk, compact disc read-only memory (CD-ROM) or any other form of storage medium well known in the art. An exemplary storage medium is coupled to the processor so that the processor can read information from the storage medium and write information to the storage medium. Of course, the storage medium can also be a component of the processor. The processor and the storage medium can be located in an ASIC. In addition, the ASIC can be located in a terminal device or a management device. Of course, the processor and the storage medium can also exist as discrete components in a terminal device or a management device.

[0208] Those skilled in the art should be able to realize that in one or more of the above examples, the functions described in the embodiments of the present application can be implemented in whole or in part by software, hardware, firmware, or any combination thereof. When implemented using software, it can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, the processes or functions described in the embodiments of the present application are generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable devices. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, the computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center in a wired (such as coaxial cable, optical fiber, digital subscriber line (DSL)) or wireless (such as infrared, wireless, microwave, etc.) manner. The computer-readable storage medium can be any available medium that the computer can access or a data storage device such as a server or data center that includes one or more available media integrated. The available medium can be a magnetic medium (such as a floppy disk, hard disk, magnetic tape), an optical medium (such as a digital video disc (DVD)), or a semiconductor medium (such as a solid state disk (SSD)), etc.

[0209] Each device and product described in the above embodiments, and each module / unit included therein, can be a software module / unit, a hardware module / unit, or partly a software module / unit and partly a hardware module / unit. For example, for each device and product applied to or integrated into a chip, each module / unit included therein can be implemented in the form of hardware such as circuits, or at least some of the modules / units can be implemented in the form of a software program that runs on a processor integrated inside the chip, and the remaining (if any) part of the modules / units can be implemented in the form of hardware such as circuits; for each device and product applied to or integrated into a chip module, each module / unit included therein can be implemented in the form of hardware such as circuits, and different modules / units can be located in the same component (such as a chip, a circuit module, etc.) or different components of the chip module, or at least some of the modules / units can be implemented in the form of a software program that runs on a processor integrated inside the chip module, and the remaining (if any) part of the modules / units can be implemented in the form of hardware such as circuits; for each device and product applied to or integrated into a terminal device, each module / unit included therein can be implemented in the form of hardware such as circuits, and different modules / units can be located in the same component (such as a chip, a circuit module, etc.) or different components inside the terminal device, or at least some of the modules / units can be implemented in the form of a software program that runs on a processor integrated inside the terminal device, and the remaining (if any) part of the modules / units can be implemented in the form of hardware such as circuits.

[0210] The above-described specific embodiments have further elaborated on the purpose, technical solutions, and beneficial effects of the embodiments of the present application. It should be understood that the above is only the specific embodiments of the embodiments of the present application and is not used to limit the protection scope of the embodiments of the present application. Any modifications, equivalent replacements, improvements, etc. made on the basis of the technical solutions of the embodiments of the present application shall be included in the protection scope of the embodiments of the present application.

Claims

1. A training method for short-term memory manipulation ability, characterized in that The method includes: Determining an initial training level based on the basic information of the target object; Determining the material generation conditions for the current training of the target object based on the initial training level and the basic information; Inputting the material generation conditions into a large language model to obtain training materials; Generating a training task based on the training materials and the initial training level; Showing the training task to the target object and obtaining feedback data of the target object on the training task; Performing multiple trainings on the target object based on the feedback data and the basic information.

2. The method according to claim 1, wherein The determining the material generation conditions for the current training of the target object based on the initial training level and the basic information includes: Obtaining the first material generation conditions corresponding to the initial training level, where the first generation conditions include at least one of the following condition factors: the number of pictures, the interference items of the pictures, and the recognition degree of the pictures; Determining the second material generation conditions based on the basic information, where the second material generation conditions include at least one of the following condition factors: style, type; Taking the first material generation conditions and the second material generation conditions as the material generation conditions.

3. The method according to claim 1 or 2, characterized in that, The generating a training task based on the training materials and the initial training level includes: Identifying the complexity of each picture in the training materials to obtain the complexity of each picture; Determining the display order of the pictures in the training materials based on the complexity of each picture and the initial training level; Sequentially putting each picture in the training materials into the task template corresponding to the initial training level to obtain the training task.

4. The method according to claim 3, characterized in that, The identifying the complexity of each picture in the training materials to obtain the complexity of each picture includes: Identifying the first number of named entities in each picture, where the named entities in each picture include target entities and interference entities, the target entities are the entities that the target object memorizes during training, and the interference entities are the entities that interfere with the target object's memory during training; Determining the first color of the target entity in each picture and the first type information of the target entity; Determining the second color of the interference entity in each picture and the second type information of the interference entity; Determining the complexity of each picture based on the first number, the first color, the second color, the first type information, and the second type information.

5. The method according to claim 4, wherein The complexity of each picture is represented by the following formula: F i = αS i 2 + β log(1 + ΔE) + γ log(1 + ΔX) Among them, F i represents the complexity of picture i, S i represents the first quantity of named entities in the picture i, ΔE represents the color difference between the first color and the second color of the picture i, ΔX represents the type difference between the first type information and the second type information of the picture i, and α, β, γ are hyperparameters.

6. The method according to any one of claims 3-5, characterized in that The sequentially putting each picture in the training materials into the task template corresponding to the initial training level to obtain the training task includes: Determining the manipulation mode of each picture based on the position of each picture in the task template, where the manipulation mode includes: update, sorting, calculation, and mental rotation, each position in the task template corresponds to a manipulation mode, and the manipulation mode corresponding to each position in the task template is determined by the training level of the task template.

7. The method according to any one of claims 3 to 5, characterized in that Presenting the training task to the target object and obtaining feedback data of the target object on the training task includes: Presenting a fixation point to the target object so that the target object fixates on the fixation point; After a preset first time interval, presenting the pictures in the training task to the target object in sequence, where there is a time interval between the presentations of each picture, and only one picture is presented each time; After a preset second time interval, presenting a question interface to the target object, where the question interface is used to display questions related to the pictures presented in sequence in the training task, and the questions are related to the manipulation mode corresponding to the presented pictures; After a preset third time interval, presenting an answer interface to the target object, where the answer interface is used to display answer options for the questions; Receiving the operation data of the target object on the answer options, and using the operation data and the time used for this training as the feedback data.

8. A training device for short-term memory manipulation ability, characterized in that, The device includes: A determination module, configured to determine an initial training level based on the basic information of the target object; A training module, configured to determine the material generation conditions for the current training of the target object based on the initial training level and the basic information, input the material generation conditions into a large language model to obtain training materials, and generate a training task based on the training materials and the initial training level; Presenting the training task to the target object and obtaining feedback data of the target object on the training task; Performing multiple trainings on the target object based on the feedback data and the basic information.

9. An electronic device, characterized in that, Including a processor, a memory, a communication interface, and one or more programs, where the one or more programs are stored in the memory and are configured to be executed by the processor, and the one or more programs include instructions for performing the steps in the method according to any one of claims 1-7.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program, and the computer program is executed by a processor to implement the method according to any one of claims 1-7.

Images