JPH06301334A - Method and device for supporting memory - Google Patents

Abstract

PURPOSE:To hold the tested result of storage conditions in the past in the state of containing a time base and to provide the optimum storage effects of a test case for storage training while utilizing the tested result by extracting and presenting the test case for promoting new storage by analyzing the history information of the stored tested result. CONSTITUTION:Test data are presented to a user, and the tested result corresponding to the response of the user is held in a storage degree management table 13 as a means for holding the tested result. A test target extraction processing part 14 analyzes the storage degree management table 13, prepares the new test data from respective groups with an algorithm embedded in a system or a method designated by the user and presents those data to the user. Since the user answers the test, the storage management table 13 is updated. A training method advice processing part 15 advises which frequency of test training to be continued corresponding to the test data in the past, time advancement conditions and the target date and time of storage completion.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a method for supporting human memorization activities in education and the like, and in particular, those in which memorization or memorization is not confirmed by analyzing the memory status data of the human until then. The present invention relates to a memory support method for obtaining an optimum memory effect by selecting only one and performing a test (memorization check). In addition, by analyzing the progress of memory, a memory activity with a limited deadline (for example, learning Kanji for elementary school students and memorizing English words for examinees)
It is about how to support memory so that you can meet your daily goals.

[0002]

2. Description of the Related Art A person who wants to memorize himself / herself records the history of repeated memory activities (completed / unfinished) on a target paper (for example, a reference book or a blank space of a word book), I used it as a guide to grasp the memory situation and narrow down the memory training targets. Most commonly, a "positive" letter was constructed next to the subject each time a memory check failed, and the number of unmemorized points in the test was recorded.

[0003]

In the above-mentioned conventional method, the information on the memory status at the past test and the time history of the test history are insufficient, and it is impossible to extract only the unmemorized object at the present time. , It was not possible to realize efficient learning by re-testing the mixed objects with the already memorized objects. Also, because the stated position is fixed, the memory of the memory object before and after it is recollected from a different angle from the memory of itself, providing an environment in which true memory can be tested. I didn't arrive.

SUMMARY OF THE INVENTION The present invention is to provide a method for holding the results of past memory status tests with a time axis in between and using the results to achieve an optimum memory effect in a memory training test case. .

[0005]

[Means for Solving the Problems] Means for holding a result of a memory test and temporal information of the test (memory level management table), and a method prepared in advance or a user's intention using the information. According to the instructions, the frequency of test training depends on the means to analyze the information and extract the test object (test object extraction processing unit), the past test data, the time progress status and the target date and time of the memory completion. A means (training method advice processing unit) for advising whether or not to continue is provided, and the memory test contents and the test method advice are output according to the user's input or inquiry.

[0006]

[Function] At the very beginning of using the system, all the data to be stored are new target data that has not been stored. The system presents the test data to the user and holds the test result in response to the user's response in the memory degree management table which is a means for holding the test result. This table classifies the storage target data group into a new target data group, an unstored data group, a temporary storage data group, a pseudo storage data group, a temporary storage data group, and a fixed storage data group based on the test results, and stores information. keeping.

The test target extraction processing section analyzes the storage degree management table, creates new test data from each of the groups by an algorithm embedded in the system or a method specified by the user, and presents it to the user. The user answers the test to update the storage management table, and by repeating this operation, the fixed storage data group increases.

[0008] The training method advice processing unit presents the progress status of the memory by analyzing the data amount of each data group and the transition status between the groups in order to make the fixed storage or temporary storage of all the target data, The amount of effort required to reach the target on the completion deadline is advised based on the test frequency and daily test amount, concentration of weak data with a large number of tests is concentrated, and test extraction from each group is performed based on the residual data amount of the data group. Advise on tuning frequency.

As described above, efficient memory training can be completed by a desired deadline.

[0010]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS An embodiment of the present invention will be described with reference to FIGS.

First, the overall flow and concept will be described, and then the details of each processing unit will be described. FIG. 1 shows the overall configuration of the present invention. The data to be stored may be externally input from the floppy disk device or the magnetic card 17 by the storage target input processing unit and set in the storage degree management table 13, or may be stored in advance in the storage degree management table of the system. May be. FIG. 2 shows the configuration of the storage degree management table (the substance of the data structure of the table will be described later with reference to FIG. 4). The constituent elements of this table are test target data (FIG. 3), and the storage status thereof is divided into levels according to past test results and held. The level is a new target data group that has not been tested yet, data that has not been stored yet, unstored data group and temporary storage data group according to the level, and pseudo according to the fixed degree of storage of the data determined to be stored. It is managed by a storage data group, a temporary storage data group, and a fixed storage data group. At the very beginning of using the system, all data is registered in the new data group.
The test target extraction processing unit 14 responds to the request of the user.
It is determined from which data group the test object is to be extracted, the test object is extracted by using each group test object extraction pointer 22 of the corresponding data group, and the user can display them on the screen, hard copy, or voice 19. Present. The user performs a memory test on it, and instructs the system of the test result by a button or key 18, etc., and the test result input processor 12 holds the memory status information of the data and the data to which the data belongs. The memory degree management table 13 and the progress status management table 16 are updated by the movement between the groups. As shown in Figure 2, the transition between data groups is fixed according to the latest test result (○: memory confirmation by test, ×: memory unfinished by test), one group each time a test can be made from the unstored data group. If a test is not possible, a transition is made to the data group, and if it is not possible to perform a test, transition is made from the three groups in the storage area to the virtual memory data group, and from the virtual memory data group to the non-memory data group. When the test is repeated, the storage degree management table 13 is composed of the six groups as described above according to the storage level, and the test target extraction unit 14 uses this group to store data in a more reliable storage level. Repeat the extraction of the test target so that it moves.

Training method advice processing unit 15
Refers to the progress status management table 16 and analyzes the progress status of the memory in response to an inquiry from the user, thereby presenting the information on the memory progress, and the estimated completion time when the current pace is used. In addition, support memory activities by giving advice such as the daily test amount required to reach the completion target time.

Next, the details of the operation of each processing unit will be described while presenting the processing flow and the specific configuration of the data structure of each table.

The storage target input processing unit 11 inputs data from each storage medium and registers all test target data in the new target data group of the storage degree management table 13. FIG. 4 shows a specific data structure of the storage degree management table. Each test target data is connected in a chain from the storage degree management table header 41 for each group. Here, the test target data is shown at 31 in FIG. If the question and answer 32 or fixed length is difficult to manage, it is managed in another place and a pointer to it, a pointer 33 for chain construction, an attribute 34 indicating when the data should be tested first, etc. , The test count information 35 of the data, the remaining count in each data group, and a finite number of latest test histories 37. The memory management table further includes a test target extraction pointer 42 for each group, and this pointer indicates a target for selecting a test question from each group, and each time the pointer is used, the pointer goes to the back chain. It is cyclically controlled so as to return to the beginning when the end of each data group is reached. The initial value points to the beginning of the chain of each data group. However, in the initial state, the test target data is connected only to the new target data group.

The test target extraction processing unit 14 determines a test target according to the processing flow shown in FIG. 5 in response to a test request from the user. In this system, a fixed addition mode that adds new data to the test target at a fixed time while managing the date and time when the new target data should be tested for the first time,
A variable addition mode that adds new data to the test object at any time according to the amount of unmemorized data at each time in order to store a certain amount of data at high speed depending on the effort of the user,
It has two modes (51). The former is assumed to be, for example, elementary school students' acquisition of new kanji, and the latter is assumed to be a candidate's memory of English words.

When the data addition mode is the fixed addition mode, it is determined whether or not the test request is the beginning of a time (day unit, week unit, month unit, etc.) set in advance by the system (5
2) If it is the first time, the relevant data in the new target data group in the storage degree management table is extracted by checking the attribute field 34 of the test target (FIG. 3) and transferred to the unstored data group, and , The new target pointer of each group test target pointer 42 is updated (53). This data movement and pointer updating is shown in FIG. The extracted data (X, Y in this example) is inserted immediately before the data (C in this example) pointed to by the extraction pointer of the unstored data group, the extraction pointer is set at the beginning of the additional data, and the data number field is set. It is updated (+ -2 in this example) according to the number of moving data. By this movement and updating of the pointer, the added data becomes a trigger for being selected first from the unstored data group.

When the data addition mode is the variable addition mode, it is determined whether the addition condition is satisfied (54), and the number of data determined by the addition condition is changed from the new target data group to the unstored data group by the same procedure as 53. Add to. Here, the additional condition is that the improvement of memory is considered to be some criteria by utilizing the physiology that is achieved by repeating the training of unmemorized data to some extent frequently. For example, the number of data in the unstored data group is set to a constant value (or more), or the sum of the data numbers of the unstored data group and the temporary storage data group is set to a certain value (or more) as the sum of a plurality of data groups. Alternatively, a weight based on the frequency selected from each data group as a test target may be multiplied by the number of each data group to obtain a total number, which is a constant value (or more). The basic idea is to monitor the progress of memory and automatically add new data so that the amount of incomplete memory data is not too large at each time point.

Next, as a method of selecting the test target data, a data group designation mode for extracting a test target from a specific data group for the purpose of, for example, strengthening training of shallow memory and reconfirming already stored data, It has two modes, a data group variable mode in which a data group is dynamically selected according to the judgment of the system (55).

In the data group designation mode, the extraction pointer 42 of the data group designated by the user or specific data is extracted as a test target (56). Here, the selection of the specific data may be overcoming weak data that cannot be easily stored, which is shown in the operation of the training method advice processing unit 15 (see FIG. 1).
0).

In the data group variable mode, data is selected at a certain frequency from each data group according to the storage level. The basic idea is to extract data more frequently from a group of data that has little memory. For example, if 50% of the unstored data group, 30% of the temporary stored data group, 15% of the pseudo stored data group, 4% of the temporary stored data group, and 1% of the fixed stored data group are selected with respect to the number of tests. , 57, the counter C is incremented for each selection, the selection data group is determined from the remainder of 100, for example, and the extraction pointer (4 in FIG. 4) of the corresponding data group is determined.
The test target is extracted according to 2) (58). However, in this example, the selection ratio of each data group becomes a predetermined numerical value every 100 times, and it has the property of being continuously extracted from the same data group. If you want to avoid that, 5 part of 57
If the flow is changed to 0, it is possible to realize test case extraction of a predetermined ratio by random access between data groups using random numbers. Further, unlike the present example, the selection ratio is not fixed only according to the characteristics of the data group, but a modification in which the ratio is dynamically controlled depending on the data amount of each data group can be easily realized.
Further, as will be described later in the explanation of the training method advice processing unit 15, it is also possible for the user to set it according to the memory characteristic of the user according to the instruction of the user.

The test target extraction processing is completed by extracting the test target by the above procedure and presenting the extracted test target to the user (59).

Although the flow of the test target extraction processing section 14 has been described above, the operation of extracting one operation has been described in this example. However, if a plurality of test questions are to be extracted simultaneously, the above procedure may be repeated. .

According to the processing flow shown in FIG. 6, the test result input processing unit 12 inputs the test result from the user (61) and updates the storage degree management table 13 and the progress status management table 16.

First, the storage test result is recorded / updated in each field of the corresponding test target data in the storage degree management table (62). Specifically, test history 3
7 is added and the count data 35 is incremented. If the test history field is full, it has the latest finite number of histories by replacing it with the oldest data in it.

Next, according to the test result, the data group to be tested is moved (63).

When the test is successful, the test object is moved to the data group one higher than the corresponding data group except the fixed storage data group, and the extraction pointer is changed to the next data. The movement of the data and the movement of the pointer are shown in FIG. This movement is common to movement between arbitrary data groups other than the new target data group. The moving data (E in this example)
Is extracted from the data group of No. 1 and is inserted before the extraction pointer (K in this example) of the destination second data group, and the extraction pointer of the first data group points to the next data (F in this example). And the data count field of each group is updated according to the movement (+1 in this example). In addition, each time the data group is moved, if each data group residence frequency field (36 in FIG. 3) of the relevant test target data is incremented,
The number of stays in each group can be recorded. By this movement, the moved data is selected at the end of the data group of the movement destination, and the data which has not been tested for a longer time is preferentially selected.

If the test cannot be performed, the data group is moved to a lower data group depending on the data group to which the test object belongs (65). In the case of the unstored data group, only the extraction pointer is moved to the next data (66). In the case of the temporary memory data group, the relevant test object is moved to the unmemorized data group, and the extraction pointer is moved to the next data (6
7). In the case of other data groups, the relevant test object is moved to the temporary memory data group, and the extraction pointer is moved to the next data (68). Here, as the procedure for moving between data groups, the procedure shown in FIG. 6B can be used as it is.

Finally, the progress management table 16 is updated by using the information of the test object and the data group (6
9).

First, the structure of the progress management table is shown in FIG.
Will be described with reference to FIG. The purpose of this table is not the storage status of individual data (this information is stored in the storage management table), but the progress of data over time and the characteristic information of the entire data To advise on progress. This table is used in the training method advice process 15 described later. FIG. 8 is a table for holding the progress of storage by date and time, which includes date and time 81, the number of tests and breakdowns 82 of the day, the breakdown of movement between data groups, and the number 84 of each data group of the day. In addition, FIG. 9 is a feature accumulation data table that holds the overall features of the test situation. Data 91 that has a large number of tests and data 92 that has remained in each group for a long period of time or a large number of times are owned by the same structure. Has been realized. These are used to find data that is difficult to remember. The structure holds the number of registrations, the minimum number of feature quantities of registration data, and pointers to the storage degree management table of each test target in the chain 94, and further has an extraction pointer 93 unique to this table. Therefore, it is possible to independently extract the test target using this table.

Now, returning to the processing of 69, the corresponding fields in FIG. 8 may be counted up or down for each test. However, 8 when you start using the day
It is assumed that 0, 2, and 83 have been cleared, and that 84 has been initialized such as copying the previous day's data. In addition, the feature amount cumulative data table (FIG. 9) is updated by using the number-of-times data 35 owned by the test target data (FIG. 3) and the number-of-residence data 36 of each data group If the number data 35 to be tested is larger than the minimum number of times in the data, a chain similar to the data chain 94 is created and connected immediately before the extraction pointer, and the registered number in 91 is set to 1 Just count up. In addition, data on over-residence of each group will be added as well.
Can be updated. Here, as the control method of this table,
There are two methods, one is to set the minimum number of times in advance and register those that exceed it, and the other is to set the upper limit of the number of registrations and gradually increase the minimum number in progress. .

Training method advice processing unit 15
FIG. 10 shows the progress management table 16 analyzed in accordance with the user's inquiry, the presentation of information in response to the inquiry request, and the extraction of the test target data from the storage degree management table 13 which is the memory special training. It is performed according to the processing flow.

First, the progress status management table is input (101). Then, in response to the inquiry of the user, desired information is presented, advice and special training are performed. In this embodiment, the progress advice (10
2), weakness training (103), effort advice, system tuning advice (105)
Showing the street.

The progress advice 102 displays the date, week, monthly test amount, storage amount, and data group transition amount by a numerical value or a graph in response to a user inquiry (106). This is based on the date / time progress data owned by the progress management table of FIG. 8, for example, the temporal transition of the number of data that can be brought over the pseudo data group, which is a tentative storage standard, and the transition of the number of tests performed every day. Alternatively, it is possible to extract the temporal change of the total amount of each data group, and it is easy to graph it.

The weakness training 103 implements a test for intensively extracting data that cannot be stored.
For example, this may be data that is frequently extracted in the tests so far, or data that rarely transits to the temporary / fixed storage data group. This can be done by extracting the test target using the extraction pointer of this data by using the test count upper data of the feature accumulation data of FIG. In accordance with the test result, the storage management table and the progress management table can be updated by this training by the same processing as the test result input processing (FIG. 6) described above. This test is repeated as many times as desired by the user to perform intensive training on certain specific weakness data (107).

In the effort amount advice process 104, if the memory completion predicted date and time is used at the current pace, and how much daily or weekly effort amount (test amount) is required to clear the target completion date and time. Is presented (10
8). An example of the principle and the realization method will be described with reference to FIG. Modeling that memory is fixed exponentially by the number of iterations, let M be the number that has reached temporary memory in a certain recent period (for example, one week), and let L be the number of tests performed in that period. And Using M and L, the memory model is expressed as in Equation 111. Here, a and b are numbers depending on the memory capacity of the individual user. M and L can be easily extracted from the progress management table 16 over the past several weeks. For example, (M1, L1), (M
2, L2) and applied to the model of 111, 112
The following two equations are obtained, and a and b representing the user's individual ability are obtained from these equations in the form of equations 113 and 114. In addition, the data group that has not been surely stored, for example, the total amount of data of four groups below the pseudo memory data group at the time of inquiry, is set as P, and the prediction completion time is the future week If the test amount is set to the average (L1 + l2) / 2 of the last two weeks, the predicted completion time (it takes several more weeks) is represented by formula 115 by using the obtained a and b and the formula 111. This is presented to the user as the predicted completion time. In addition, if the original goal achievement time E is input, for example, the remaining weeks up to one month before the test, the same amount of effort per week (number of memory tests) can be calculated by using formula 111 as X of formula 116. It should be presented to the user. In this example, a simple memory model has been described, but it is possible to calculate from the above data regardless of which complicated model is applied.

System tuning advice 105
Displays the transition of the amount of data of each data group by means of a progress status management table (FIG. 8) by means of numerical values and graphs (109). Due to individual differences, you can remember it quickly but forget it immediately (for example, temporary memory, pseudo memory data groups often return to temporary data group), or you can't remember it very much, but if you remember it (always remember the temporary memory and pseudo memory data group) Depending on individual memory characteristics such as a large amount, but a small amount of data returned from the temporary storage data group), the extraction ratio of each data group currently used by the system is displayed, and advice from the system or user's instruction , By changing the extraction ratio of each data group, the optimal memory training frequency for an individual is achieved.

The embodiment of the present invention has been described above.

In the above embodiment, it is not specified whether the user or the system determines whether it is possible / not possible (stored / not stored). It can be judged by the system by inputting the answer that the system has in the target data and the answer of the user depending on the object to be memorized or tested, or the user himself can display the answer from the system. This is because it is possible to make a judgment and input the correctness or the both. Especially in the latter case, an interactive operation with the system can be performed by using several kinds of input buttons, and for example, a pocket electronic word book can be realized.

Further, although only one storage object is described in the above embodiment, the individual storage degree management tables 13 for a plurality of storage objects are provided in the same system, and any one of them can be selected by the user. It suffices to perform an interactive process of memory training for the memory degree management table of. Therefore, for example, for questions and answers that can be replaced, the same initial input data 17 is used, and both Japanese and English word tests, and
You can easily build a system that allows you to test from two directions, kanji writing and reading. Further, it is also possible to own a plurality of completely different types of storage objects in the same system by having a plurality of storage degree management tables.

Next, an embodiment for expanding the application and effect of the system of the present invention will be described.

FIG. 12 shows the configuration of the apparatus of one embodiment of the present invention.

The storage support device comprises a processing device 200, a learning content storage device 202, and a progress situation storage device 204.

The processing device 200 comprises a display screen 208, a button 18, an electronic pen 201, a voice output terminal 206, and storage device mounting slots 203 and 205.

The learning content storage device 202 is a device for storing learning contents, and is not particularly limited, but for example, R
It is configured by an external storage device such as an OM, a CD-ROM, a floppy disk, a magnetic card.

The progress status storage device 204 is a device for storing information such as the storage degree management table 13 and the progress status management table 16 which are updated according to the progress of the user. Such non-volatile writable storage means is constituted by a removable device such as a memory card.

The button 18 and the electronic pen 201 are used by the user to input an answer to the test to the processing device 200. The input means is not particularly limited to these and may be a pointing device such as a mouse or a joystick.

In the above embodiment, the learning content storage device 202 and the progress management storage device 204 are independent of each other. However, the invention is not limited to this. Management storage device 2
04 and 04 may be stored in one card.

The above embodiments are most effective for a device for memory support, but are not limited to memory support, but present a problem to the user and solve the problem of inputting or selecting the solution. It can also be applied to learning support. For this purpose, the storage degree management table 13 may be used as a correctness degree management table.

[0049]

According to the present invention, the situation of memory is grasped assuredly and objectively at any time, and the optimum memory training object (test case) at that time is provided on the basis of it, thereby providing a complete memory in a minimum time. Can be realized. In addition, the physiological effects of fading memory can be monitored by the system itself, and more accurate training can be fixed to actual memory. In addition, by instructing the amount of effort required for completion by the time the target is reached, it is possible to support memory activity while making the user aware of deadline management. Since the device of this system can be carried as a substitute for a word book and can be used even in transportation facilities because it is compact, it is possible to make effective use of time.

[Brief description of drawings]

FIG. 1 is an overall configuration.

FIG. 2 is a structural concept of a storage management table.

FIG. 3 is a data structure of test target data.

FIG. 4 is a data structure of a storage degree management table.

FIG. 5 is a test target extraction processing flow.

FIG. 6 is a test result input processing flow.

FIG. 7 shows data transition between data groups.

FIG. 8 is a data structure of a progress management table (date / time progress data).

FIG. 9 is a data structure of a progress management table (feature cumulative data).

FIG. 10 is a training method advice processing flow.

FIG. 11 is an effort amount advice method.

FIG. 12 is a schematic configuration diagram of an apparatus for carrying out the present invention.

[Explanation of symbols]

1 ... Overall configuration of system, 11 ... Storage target input processing unit,
12 ... Test result input processing unit, 13 ... Memory level management table, 14 ... Test target extraction processing unit, 15 ... Training method advice processing unit, 16 ... Progress status management table.

Claims (15)

[Claims] 1. In a system for supporting human memory activity, a memory status for a test case given by the system is input to the system, history information of test results of the memory is retained, and the information is analyzed and newly updated. A method for supporting memory, which comprises extracting and presenting test cases for promoting effective memory. 2. A group of three or more, such as a new object which has not been tested yet, a tested unmemorized object, and a tested memory object, based on the memory level of the memory object based on past test history information. The method for supporting memory according to claim 1, wherein a new test case is extracted from each group with a certain probability from each group. 3. Based on the past test history information, the storage target is changed from the level of the storage to a new target that has not been tested yet, a tested non-stored target, a tested storage target, and further according to the test history. The storage support method according to claim 1, wherein the storage target and the storage target are each divided into a plurality of small groups, and a new test case is extracted from each group with a certain probability. 4. A method of dividing a memory object into a plurality of small groups based on past test history information based on the level of the memory and extracting a new test case from each group with a certain probability. The method for supporting memory according to claim 2, characterized in that the last tested one is preferentially extracted. 5. A method of dividing a memory object into a plurality of small groups based on the past test history information based on the level of the memory and extracting a new test case from each group. The memory support method according to claim 2, wherein the selection probability from each group is adjusted so that the group does not blister. 6. A storage object is divided into a plurality of small groups based on the past test history information based on the level of the storage, and a new test case is extracted from a specified specific group. A method for supporting memory according to claim 2. 7. A method of dividing a memory object into a plurality of small groups based on the past test history information from the level of the memory and updating the belonging data of the small groups at any time according to the continuous test results. A second aspect of the present invention is characterized in that means is provided for collectively outputting group membership data.
A method of supporting memory according to the item. 8. An object for memory training is sequentially added from the entire memory object, but a new object is tested based on the completion status of the memory so far so as not to have too many unmemorized objects. The method for supporting memory according to claim 2, wherein the method is added to the case. 9. A new object for memory training is sequentially added from the whole of the memory objects, and the new object is added to the test case at a predetermined time. The method for supporting memory according to item 2. 10. The memory test history information is analyzed to give advice such as memory training progress data, estimated estimated memory completion date and time, and memory training frequency (necessary test frequency) for ending within a specified period. The method for assisting memory according to claim 1, wherein: 11. The memory support according to claim 1, wherein the history information of the memory test is analyzed to detect data in which the memory does not proceed smoothly, and the data is extracted as the next test case. Method. 12. A device for supporting human learning, a computer, a first storage device for storing a learning content and a computer program for interactively supporting learning, and a learning achievement status of a user. And a second storage device for storing the memory. 13. The memory support device according to claim 12, wherein the computer has an input key or a pointing device for operating a computer program for the user to interactively support learning. 14. The memory support device according to claim 12, wherein the learning content includes pronunciation of a word, and the computer has a voice output unit for reproducing the pronunciation of the word. 15. A first storage device for storing a learning content and a computer program for interactively supporting learning, and a second storage device for storing a learning achievement status of a user. A memory support device, wherein the device and the device are configured by an integrated external storage device.