Training system for motion direction distinguishing ability

Abstract

The invention relates to the technical field of exercise direction distinguishing ability training, in particular to an exercise direction distinguishing ability training system which comprises a training parameter setting module, a training image generating module, a man-machine interaction module, a training process control module and a data recording and analyzing module. The training parameter setting module sets training related parameters; the training image generation module generates a training image; the man-machine interaction module presents a training image, obtains the identification condition of a training individual on the training image, monitors a training environment and a training state, and assists training through a low-frequency modulated intermediate-frequency current; the training process control module controls a training process; the data recording and analyzing module can record and analyze training data and evaluate the overall training process. Through targeted visual stimulation design, rigorous training process control and synchronous combination of training auxiliary equipment, training of the motion direction distinguishing ability is effectively achieved, and the motion direction distinguishing ability of a biological individual is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of training for the ability to distinguish movement directions, and particularly to a training system for the ability to distinguish movement directions. Background Art

[0002] The ability to distinguish movement directions is a visual function closely related to daily life. For example, when a bicycle falls and crashes into the roadside, pedestrians on the roadside need to judge whether they need to dodge; when a football flies towards the goal, the goalkeeper needs to judge the movement direction of the football to determine the method of blocking the goal; when driving, one needs to judge the movement direction of the vehicle in front to avoid collisions. Therefore, a good ability to distinguish movement directions helps to improve the quality of life and protect one's own health.

[0003] Scientific research shows that most functions of biological individuals can be effectively improved or enhanced through targeted training. In fact, this is also a kind of ability and manifestation form for biological individuals to adapt to the environment. However, there are almost no devices or systems for training the ability to distinguish movement directions at present. Summary of the Invention

[0004] The purpose of the present invention is to solve the deficiencies in the prior art and propose a training system for the ability to distinguish movement directions.

[0005] To achieve the above purpose, the present invention adopts the following technical solutions:

[0006] Design a training system for the ability to distinguish movement directions, including:

[0007] A training parameter setting module for setting parameters related to training, including the reference movement direction, movement speed, minimum change unit of the movement direction change amount, and the training amount for a single training;

[0008] A training image generation module that generates training images based on the parameters obtained by the parameter acquisition module, the built-in parameters of the system, and the movement direction change amount transmitted by the process control module;

[0009] A human-computer interaction module, including a display device, an information input device, a sound prompt device, and a monitoring device; the display device is used to present the generated training images to the training individual, the information input device is used to obtain the recognition situation of the training individual for the training images; the sound prompt device is used to feedback the correct / incorrect situation of the training image recognition to the training individual; the monitoring device is used to monitor the training environment and the training status of the training individual, and can remind the training individual to correct abnormal situations through the sound prompt device;

[0010] The training process control module is used for the sequential control between different trials during a single training session, specifically including: analyzing the training effectiveness of the previous completed trial based on the environmental and training status information obtained by the monitoring device, and combining the identification situation of the training individual for the training images, sending a control signal to the training image generation module to adjust the additional movement direction of the training images generated in the upcoming next trial; repeating this cycle until the training ends;

[0011] The data recording and analysis module is used for recording and analyzing the training data of this time, and can evaluate the overall training process based on all the training data recorded this time and previously.

[0012] Preferably, the training images generated by the training image generation module are moving grids formed by superimposing two moving gratings with a 90-degree difference in movement direction.

[0013] Preferably, after receiving the additional movement direction transmitted by the process control module, the visual stimulus generation module will randomly select one of the two situations: "the movement direction of the newly generated movement stimulus deflects the additional movement direction clockwise relative to the reference movement direction" and "the movement direction of the newly generated movement stimulus deflects the additional movement direction counterclockwise relative to the reference movement direction" to generate a movement stimulus.

[0014] Preferably, the monitoring device includes an environmental illuminance monitoring device, a training distance monitoring device, and an eye movement monitoring device;

[0015] The environmental illuminance monitoring device monitors the illuminance of the training environment in real time through an illuminance sensor; when the relative change amount of the environmental illuminance compared with the environmental illuminance at the start of training exceeds the first specified ratio, it sends a first prompt signal to the training process control module in real time, and sends a voice prompt of "correct the environmental illuminance" to the training individual through a voice prompt device;

[0016] The distance evaluation unit monitors the distance between the display device and the training individual in real time through a distance sensor; when the monitored distance exceeds the second specified ratio of the established training distance, it sends a second prompt signal to the training process control module in real time, and sends a voice prompt of "correct the training distance" to the training individual through a voice prompt device;

[0017] The eye movement monitoring device is used to monitor the line-of-sight direction of the trained individual when the training image appears; when it is found that the angle of the line-of-sight direction of the trained individual deviates from the training image by more than the first specified viewing angle, it sends a third prompt signal to the training process control module in real time, and sends a voice prompt of "correct the line-of-sight direction" to the training individual through a voice prompt device.

[0018] Preferably, the training process control module analyzes the training effectiveness of the previous trial that has ended, and based on the recognition situation of the training individual for the training images, sends a control signal to the training image generation module to adjust the specific method of the additional motion direction of the training images generated in the upcoming next trial, including:

[0019] S1-1. If during the presentation of the training images in the previous trial, the training process control module receives the first prompt signal sent by the ambient illumination monitoring device, or the second prompt signal sent by the distance evaluation unit, or the third prompt signal sent by the eye movement monitoring device, it is determined that the training in the previous trial is "invalid", and a control signal is sent to the training image generation module to determine that in the next trial, the absolute value of the additional motion direction of the training images will not be adjusted and will continue to use the absolute value of the additional motion direction in the previous trial; otherwise, it is determined to be "valid", and proceed to S2-2;

[0020] S1-2. If in the previous trial, the training individual correctly recognized the stimulus motion direction, then in the next trial, the absolute value of the additional motion direction of the motion grid is reduced by 1 minimum change unit; if the recognition result in the previous trial is incorrect, then in the next trial, the absolute value of the additional motion direction is increased by the first specified number of minimum change units; where the first specified number is a positive integer greater than 1.

[0021] Preferably, the initial value of the minimum change unit of the additional motion direction is 5.0 degrees. After completing the measurement of the first 3 trials, it is adjusted to 1 / 5 of the initial value, that is, 1.0 degree. After completing the measurement of another 3 trials after the adjustment, it is further adjusted to 1 / 10 of the initial value, that is, 0.5 degree.

[0022] Preferably, the steps for the data recording and analysis module to analyze the training data of this time and evaluate the overall training process include:

[0023] S2-1. Based on the recorded data, list the absolute values of the additional motion directions used in all "valid" trials, as well as the correct / incorrect situation of the training individual's recognition of the training images;

[0024] S2-2. Based on the list sorted out in S2-1, select the values at the points where the change trend of the absolute value of the additional motion direction changes during the training process, and record them all as C cri ;

[0025] S2-3. Among all C cri in chronological order of appearance, eliminate the first 3 or the first 4 C cri , so that the number of the remaining C cri is an even number, and then take all the remaining C criThe arithmetic mean, and the result is the threshold of the motion direction discrimination ability in the current reference motion direction obtained from this training;

[0026] S2-4. List all the thresholds of the motion direction discrimination ability in this and previous records. If the quantity is not less than the second specified quantity, evaluate the overall training process; when evaluating, first perform data fitting using the following formula:

[0027] lg(threshold n ) = c × lg(n) + d (Formula 1)

[0028] lg(threshold n ) = a × 0.5 lg(n) + b (Formula 2)

[0029] Among them, threshold n is the threshold of the motion direction discrimination ability obtained during the nth training in this and previous multiple trainings; n is a positive integer greater than or equal to 1; a, b, c, and d are all parameters to be fitted;

[0030] S2-5. Evaluate the results of data fitting using Formula 1 and Formula 2 in S2-4; if the goodness of fit of the data obtained when fitting with Formula 1 is better than the goodness of fit of the data obtained when fitting with Formula 2, and the data fitting performed using Formula 1 is statistically significant, then determine that "the training process is in the first stage" and end the evaluation; if the goodness of fit of the data obtained when fitting with Formula 1 is better than the goodness of fit of the data obtained when fitting with Formula 2, but the data fitting performed using Formula 1 is not statistically significant, then determine that "the training process is in the third stage" and end the evaluation; if the goodness of fit of the data obtained when fitting with Formula 2 is better than the goodness of fit of the data obtained when fitting with Formula 1, and the data fitting performed using Formula 2 is statistically significant, then preliminarily determine that "the training process enters the second stage" and enter S2-6;

[0031] S2-6. Based on the latest obtained third specified quantity of thresholds of the motion direction discrimination ability, re-perform data fitting using the formula in S2-4. If the goodness of fit of the data obtained when fitting with Formula 1 is better than the goodness of fit of the data obtained when fitting with Formula 2 at this time, but the data fitting performed using Formula 1 is not statistically significant, then determine that "the training process enters the third stage"; otherwise, officially determine that "the training process enters the second stage".

[0032] Preferably, the human-computer interaction module further includes a low-frequency modulated medium-frequency current stimulation device, which can generate a low-frequency modulated medium-frequency current and act on the acupoints around the eyes during the presentation of the training image.

[0033] Preferably, when the medium-frequency current modulated by low frequency acts on the acupoints around the eyes, the output power is related to the additional movement direction in the training image, and the formula is:

[0034] P = P max -(P max -P min )×ΔD

[0035] where P min and P max are respectively the minimum and maximum values of the output power, and ΔD is the angle by which the movement direction of the movement stimulus deflects relative to the reference movement direction, that is, the additional movement direction.

[0036] Preferably, the medium-frequency current stimulation device can also give training feedback to the training individual through current stimulation, specifically: in each trial, when the training individual completes the identification of the training image, if the identification is correct, a current stimulation with a first specified power and a first specified duration is given to the training individual; otherwise, no current stimulation is given.

[0037] The training system for the ability to distinguish movement direction proposed by the present invention has the beneficial effect that: through targeted visual stimulus design, rigorous training process control, and synchronous combination of training auxiliary equipment, the training system for the ability to distinguish movement direction effectively realizes the training of the ability to distinguish movement direction and improves the ability of biological individuals to distinguish movement direction. Description of the Drawings

[0038] Figure 1 is an example diagram of a training system for the ability to distinguish movement direction proposed by the present invention.

[0039] Figure 2 is a schematic diagram of a training image (moving grating stimulus).

[0040] Figure 3 is a schematic diagram of the change in the ability to distinguish movement direction during a certain training. Detailed Embodiments

[0041] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments.

[0042] As Figure 1 shown, a training system for the ability to distinguish movement direction includes:

[0043] (1), a training parameter setting module for setting parameters related to training, including a reference movement direction, a movement speed, a minimum change unit of the movement direction change amount, and the training amount of a single training;

[0044] (2) A training image generation module that generates training images based on the parameters obtained by the parameter acquisition module, the built-in parameters of the system, and the amount of change in the movement direction (i.e., the additional movement direction) transmitted by the process control module.

[0045] (3) A human-computer interaction module, including a display device, an information input device, a sound prompt device, and a monitoring device; the display device is used to present the generated training images to the training individual, the information input device is used to obtain the recognition situation of the training individual for the training images; the sound prompt device is used to feedback the correct / incorrect situation of the training image recognition to the training individual; the monitoring device is used to monitor the training environment and the training status of the training individual, and can remind the training individual to correct abnormal situations through the sound prompt device.

[0046] (4) A training process control module that is used for the sequential control between different trials during a single training process, specifically including: analyzing the training effectiveness of the previous trial that has ended based on the environment and training status information obtained by the monitoring device, and combining the recognition situation of the training individual for the training images, sending a control signal to the training image generation module to adjust the additional movement direction of the training images generated in the upcoming next trial; repeating this cycle until the training ends.

[0047] (5) A data recording and analysis module that is used for the recording and analysis of the training data of this time, and can evaluate the overall training progress based on all the training data recorded this time and previously.

[0048] All kinds of parameters set in the training parameter setting module are related to the generation of training images, the control of the training process, and the training effect.

[0049] The reference movement direction obtained by the parameter acquisition module is the reference standard for movement direction discrimination. When the training individual completes the training task, it is necessary to distinguish the difference between the movement direction of the presented movement stimulus and the established reference movement direction. That is, to judge whether the movement direction of the just-appeared stimulus has a clockwise deflection or a counterclockwise deflection compared with the reference movement direction?

[0050] The reference movement direction can be any direction within 360 degrees of the circumference. However, considering the application situation of the movement direction discrimination ability in real life, the reference direction is mostly taken as one or more values among 0 degrees, 45 degrees, 90 degrees, 135 degrees, 180 degrees, 225 degrees, 270 degrees, and 315 degrees.

[0051] The minimum change unit of the amount of change in the movement direction (i.e., the additional movement direction) set in the training parameter acquisition module is related to the change in the difficulty of stimulus identification during the training process. To reach the current identification threshold of the training individual, the difficulty of the stimuli presented to the training individual during the training process usually becomes greater and greater (i.e., the additional movement direction becomes smaller and smaller). The change amplitude of the difficulty of identification each time is directly related to the currently set minimum change unit of the additional movement direction. According to the process control rules applied in this system, the smaller the minimum change unit, the longer the process of the difficulty of stimulus identification changing from the initial level to the level corresponding to the identification threshold of the training individual (see Figure 3 "Phase 1" in

[0052] . The more trial times consumed in Phase 1, the fewer trial times in the subsequent Phase 2 that actually exerts the training effect. Therefore, there is no need to overly pursue a small minimum change unit. On the other hand, the minimum change unit should not be too large either. Because when the minimum change unit is large, although "Phase 1" can be quickly passed, in the subsequent "Phase 2", the fluctuation degree of the difficulty of stimulus identification will become larger, resulting in the difficulty of some stimuli deviating far from the threshold level, thus reducing the training effect of "Phase 2".

[0053] In general, the minimum change unit should be determined before training and remain stable during the training process. A stable minimum change unit means that the gap between two adjacent difficulty levels defined by the minimum change unit is fixed. This helps to control errors and ensure the training effect.

[0053] In addition, to quickly pass through "Phase 1" (see Figure 3 ) during each training, a larger minimum change unit can be selected in the first few trial times after the start of each training. Subsequently, the minimum change unit is gradually reduced. Preferably, the initial value of the minimum change unit of the additional movement direction can be set to 5.0 degrees. After completing the training of the first 3 trial times, it is adjusted to 1 / 5 of the initial value, that is, 1.0 degree. After completing the training of another 3 trial times after the adjustment, it is further adjusted to 1 / 10 of the initial value, that is, 0.5 degree.

[0054] The training amount set in the training parameter setting module is directly related to the time consumed in a single training. Generally, the value range of the training amount is 300 - 600 trial times. Among them, one trial time represents "presenting a stimulus once and receiving the recognition situation of the training individual for the presented stimulus this time". Assuming that the completion time of one trial time is 3 seconds, the training time corresponding to the above training amount is 15 - 30 minutes. Of course, if the necessary rest time during the training process is considered, the time consumed in a single training corresponding to the above training amount is approximately 20 - 40 minutes.

[0055] The training volume set in the training parameter acquisition module is also related to the training effect. According to the general law of biological individuals adapting to the environment, when a biological individual receives a certain degree of external stimulation, and the stimulation intensity and cumulative amount reach a certain level but do not cause serious damage to the body, the biological individual will strengthen the parts or tissues related to the stimulation reception or response after the stimulation ends, so as to enhance the response ability to the stimulation. At the behavioral level, this is manifested as "Practice makes improvement". Therefore, a certain amount of training volume and necessary training difficulty are required during training, otherwise the body cannot be promoted to self-enhance after being stimulated. When the training volume set in the training parameter acquisition module is insufficient, not only is the training intensity of a single training insufficient, but it will also affect the training difficulty. This is because, if the starting difficulty of stimulus identification is relatively low and the training volume is small, it is very likely that the stimulus identification difficulty has not reached near the identification threshold at the end of the training. Generally, the threshold difficulty can be simply understood as the ability limit that a biological individual can reach. Not reaching the threshold difficulty means that the average difficulty of the stimulus during the training process does not meet the specified requirements. In this case, even if the training volume is relatively large, it is difficult to promote the body's self-enhancement. In real life, it is the same principle that ordinary people can hardly exercise their sprint speed or long-distance running endurance through walking training.

[0056] Optionally, the parameter setting module can also set parameters related to the motion stimulus attributes, including the size of the stimulus. The size of the stimulus is mostly marked by the diameter, and the unit is the visual angle. The macular area of the human eye generally does not exceed 6 degrees of visual angle, so the value range of the stimulus size is generally 2.5 - 6.0 degrees of visual angle. In addition, in actual applications, to reduce the difference between the stimulus edge and the background, the stimulus edge is often blurred using a specific algorithm. For the stimulus after edge blurring, it is necessary to ensure that the size of the non-edge area that has not been blurred meets the test requirements. Specifically, the stimulus size can be selected as 5.5 degrees of visual angle, where the size of the blurred edge is 0.75×2 = 1.50 degrees of visual angle, and the non-edge area that has not been blurred is 5.5 - 1.5 = 4.0 degrees of visual angle.

[0057] Optionally, the parameters set by the parameter setting module also include the presentation time of the stimulus. To ensure that the measurement results can reflect the function of the central area of the measured individual's visual field and avoid the line of sight of the measured individual from drifting during the test, the presentation time of the stimulus should not be too long, generally 150 - 350 milliseconds. Preferably, the presentation time is 250 milliseconds.

[0058] The training image generation module can generate training images based on the parameters obtained by the parameter acquisition module, the system built-in parameters, and the amount of change in the motion direction (i.e., the additional motion direction) transmitted by the process control module.

[0059] The motion stimulus is a moving grating formed by superimposing two moving gratings with a 90-degree difference in motion direction. The motion direction of the superimposed moving grating follows the principle of vector addition of the motion directions and speeds of the two sub-component gratings, as Figure 2 shown. That is, when the motion speeds of the two sub-component gratings are the same, the motion direction of the superimposed grating stimulus should be exactly in the middle of the motion directions of the two gratings; when the motion speeds of the two sub-gratings are different, the motion direction of the superimposed grating stimulus will be biased towards the sub-component grating with a faster motion speed.

[0060] On this basis, moving gratings can be designed for the two cases where the motion speeds of the two sub-component gratings are the same and different, respectively.

[0061] When the motion speeds of the two sub-component gratings are the same, the specific formula is:

[0062] l(x,y,t)=l(x,y,t) mg1 +l(x,y,t) mg2

[0063]

[0064] θ1=D p +ΔD+45

[0065] θ2=D p +ΔD+135

[0066]

[0067] where l(x,y,t) represents the luminance of the point with coordinates (x,y) on the moving grating stimulus image at time t, l(x,y,t) mg1 and l(x,y,t) mg1 respectively represent the luminance of the point with coordinates (x,y) on the two moving grating stimulus images at time t, L mean represents the average background luminance, C is the contrast of the moving grating, f is the spatial frequency of the moving grating (unit "cycles / degree"), DPD is the visual angle occupied by each point in the stimulus image (unit "degrees / point"), θ1 and θ2 are the orientations of the two moving gratings (unit "degrees") respectively, ω1 and ω2 are the temporal frequencies of the moving gratings (unit "cycles / second"), and are the initial phases of the two moving gratings respectively, S is the motion speed of the motion stimulus formed by integrating the two moving gratings (unit "degrees / second"), D pLet \(D\) be the reference motion direction (unit: degree), and \(\Delta D\) be the angle by which the motion direction of the motion stimulus is deflected relative to the reference motion direction, i.e., the additional motion direction. When \(\Delta D>0\), the motion direction of the motion stimulus is deflected counterclockwise relative to the reference motion direction; when \(\Delta D<0\), the motion direction of the motion stimulus is deflected clockwise relative to the reference motion direction.

[0068] When the motion speeds of the two sub-component gratings are different, the specific formula is:

[0069] \(l(x,y,t)=l(x,y,t)\) mg1 \(+l(x,y,t)\) mg2

[0070]

[0071] \(\theta_1 = D\) p \(+45\)

[0072] \(\theta_2 = D\) p \(+135\)

[0073]

[0074] where \(l(x,y,t)\) represents the luminance of the point with coordinates \((x,y)\) on the moving grating stimulus image at time \(t\), \(l(x,y,t)\) mg1 and \(l(x,y,t)\) mg1 respectively represent the luminance of the points with coordinates \((x,y)\) on the two moving grating stimulus images at time \(t\), \(L\) mean represents the background average luminance, \(C\) is the contrast of the moving grating, \(f\) is the spatial frequency of the moving grating (unit: cycle / degree), \(DPD\) is the visual angle occupied by each point in the stimulus image (unit: degree / point), \(\theta_1\) and \(\theta_2\) are the orientations of the two moving gratings respectively (unit: degree), \(\omega_1\) and \(\omega_2\) are the temporal frequencies of the two moving gratings respectively (unit: cycle / second), and are the initial phases of the two moving gratings respectively, \(S\) is the motion speed of the stimulus integrated by the two moving gratings (unit: degree / second), \(D\) p is the reference motion direction (unit: degree), and \(\Delta D\) is the angle by which the motion direction of the motion stimulus is deflected relative to the reference motion direction, i.e., the additional motion direction. When \(\Delta D>0\), the motion direction of the motion stimulus is deflected counterclockwise relative to the reference motion direction; when \(\Delta D<0\), the motion direction of the motion stimulus is deflected clockwise relative to the reference motion direction.

[0075] When generating visual stimuli, the initial value of the additional movement direction also needs to be considered. It has been explained above the necessity of shortening "Phase 1" in the testing process. From this perspective, it seems that the closer the initial difficulty is to the actual functional level of the trained individual, the shorter "Phase 1" will be, which is more conducive to extending "Phase 2" and exerting the training effect of "Phase 2". However, a lower initial difficulty, which is far from the actual functional level of the measured individual, is also helpful for entering and maintaining the training state. That is, it is beneficial for the trained individual to identify the stimuli at the beginning of the training and quickly familiarize with the operation process, avoiding mistakes and mood fluctuations caused by incorrect operations. Therefore, the most appropriate choice is to appropriately adjust the initial difficulty so that it is relatively close rather than absolutely close to the actual functional level of the measured individual. Generally, the setting range of the initial value is 5.0 to 30.0 degrees. Preferably, the initial value is set to 20.0 degrees.

[0076] It should be noted that when actually identifying the movement direction, it is easier to distinguish the movement direction of the moving grid formed by superimposing two sub-component moving gratings with the same speed than that formed by superimposing two sub-component moving gratings with different speeds. This is because, for a moving grating, its movement direction is always perpendicular to the grating orientation. Therefore, when identifying the movement direction of the former, it can be distinguished not only based on the movement direction of the grid, but also with the assistance of the orientation information of the two sub-component moving gratings. However, when identifying the movement direction of the latter, the orientation information of the two sub-component moving gratings cannot be used at all. By using a moving grid formed by superimposing two sub-component moving gratings with different speeds, the movement direction of the overall moving grid can be adjusted by adjusting the speeds of the two sub-component moving gratings while the orientations of the two sub-component moving gratings remain completely unchanged. Therefore, it is recommended that during the long-term training process, one of the two types of moving grid stimuli be fixed for use within a certain period, and the above two different moving grids should not be used in a mixed manner.

[0077] After the training images are generated, they can be presented to the trained individual through the display device in the human-computer interaction module. After the images are presented, the trained individual should distinguish the difference between the movement direction of the stimulus and the established reference movement direction. That is, to judge whether the movement direction of the just-appeared stimulus has a clockwise deflection or a counterclockwise deflection compared with the reference movement direction? After the judgment is completed, the measured individual can feedback their identification of the movement stimulus to the system through the information input devices such as the mouse, keyboard or handle in the human-computer interaction module. After the system obtains the identification of the movement stimulus by the measured individual, it generates and presents the next movement stimulus. This cycle continues until all the trials are completed. For the convenience of the measured individual to identify, in each trial, before the moving grid stimulus with the movement direction to be judged appears, a moving grid stimulus with the reference movement direction can be presented first, and the presentation time can be the same as that of the moving grid stimulus with the movement direction to be judged.

[0078] In addition, the display device in the human-computer interaction module should be able to accurately display motion stimuli. This was basically not a problem when using traditional cathode ray tube displays (i.e., CRT displays) to display motion stimuli in the past. Because in traditional CRT displays, the brightness change of pixel points on the display is caused by the electron beam hitting the fluorescent screen. As long as the electron beam hits the fluorescent screen, the phosphor can immediately emit light, and when the electron beam disappears, it can immediately stop emitting light, and the glow persistence time is extremely short. Therefore, the response time of traditional CRT displays is only 1 - 3 ms, and most motion stimuli can be accurately displayed. However, the displays currently in use are basically liquid crystal displays. The brightness change of pixel points on liquid crystal displays is mainly achieved by applying a voltage to the liquid crystal molecules and twisting the liquid crystal molecules. The process of twisting and restoring the liquid crystal molecules takes time. Therefore, the response time of liquid crystal displays is relatively long, often resulting in a feeling of trailing in dynamic images (especially motion stimuli that require accurate display). Therefore, when using a liquid crystal display to present motion stimuli, each identical image should be maintained for at least 3 frames to avoid blurring and ensure the accurate display of motion stimuli. In addition, to avoid motion illusion, in the multiple images showing the grating motion, the phase change of the grating between adjacent two images should not exceed 90 degrees. This means that at least 4 images are required for the grating to move one cycle (360 degrees). Further, for a moving grating with a temporal frequency of ω (unit: cycles per second), at least 4ω images are required per second. When converted to the refresh rate of the display device (i.e., the number of frames that can be displayed in 1 second), 4ω images correspond to 12ω frames.

[0079] Therefore, the refresh rate of the display device used in the human-computer interaction module to present motion stimuli should meet the following requirements:

[0080] RefreshRate≥12×max(ω1,ω2)

[0081] where ω1 and ω2 are the temporal frequencies of two moving gratings respectively, and max(·) is the maximum value function.

[0082] In addition, various monitoring devices are also included in the human-computer interaction module to confirm the normality of the behavior of the training individual in each training trial or the stability of the training environment. The monitoring devices include environmental illuminance monitoring devices, training distance monitoring devices, and eye movement monitoring devices.

[0083] Among them, the environmental illuminance monitoring device monitors the illuminance of the training environment in real time through an illuminance sensor; when the relative change amount of the environmental illuminance compared with the environmental illuminance at the start of training exceeds the first specified ratio, it sends a first prompt signal to the training process control module in real time, indicating that the training environment has changed significantly, and sends a voice prompt of "correct the environmental illuminance" to the training individual through a voice prompt device. Preferably, the first specified ratio is 50%.

[0084] The distance evaluation unit monitors the distance between the display device and the training individual in real time through a distance sensor; when the monitored distance exceeds the second specified ratio of the established training distance, it sends a second prompt signal to the training process control module in real time, indicating that the training behavior is not standardized, and sends a voice prompt of "correct the training distance" to the training individual through a voice prompt device. Preferably, the second specified ratio is 10%.

[0085] The eye movement monitoring device is used to monitor the line-of-sight direction of the trained individual when the training image appears; when it is found that the line-of-sight direction of the trained individual deviates from the training image by more than the first specified viewing angle, it sends a third prompt signal to the training process control module in real time, indicating that the training behavior is not standardized, and sends a voice prompt of "correct the line-of-sight direction" to the training individual through a voice prompt device. Preferably, the first specified viewing angle is 6.0 degrees.

[0086] It should be noted that generally the training time is relatively long (about 30 minutes on average), and there are few dedicated personnel to take care of it. Therefore, the setting of using monitoring devices to detect abnormal environmental states and training states is of great significance, which can effectively avoid interference during the training process and effectively improve the training effect.

[0087] Optionally, in order to achieve better training effects, the human-computer interaction module further includes a low-frequency modulated medium-frequency current stimulation device, which can generate a low-frequency modulated medium-frequency current and act on the acupoints around the eyes during the presentation of the training image; optionally, both the medium-frequency carrier current and the low-frequency modulation signal use sinusoidal alternating current, the frequency range of the medium-frequency carrier current is 1.0 kHz - 5.0 kHz, and the frequency range of the low-frequency modulation signal is 10.0 Hz - 120.0 Hz.

[0088] Optionally, when the low-frequency modulated medium-frequency current acts on the acupoints around the eyes, the output power is related to the additional movement direction in the training image, and the formula is:

[0089] P = P max -(P max -P min )×ΔD

[0090] Where P min 、P maxare the minimum and maximum values of the output power respectively, and ΔD is the angle by which the movement direction of the movement stimulus deflects relative to the reference movement direction, that is, the additional movement direction. Preferably, P min and P max are 0.5W and 5.0W respectively.

[0091] It should be noted that the above formula describes the relationship between two kinds of stimulus energies, that is, the relationship between the visual signal energy corresponding to the distinguishing technical features carried by the training image and the current energy. Its internal mechanism is that the perceptions from different sensory organs can be linked to improve visual cognitive function. The biological mechanism of the linkage is very complex and involves multiple aspects such as cognitive integration and neurotransmitter regulation. In practical applications, using medium-frequency current with low-frequency modulation and acting on the acupoints around the eyes during the presentation of the training image can effectively form the linkage of "light stimuli rich in visual meaning" and "body surface electrical stimulation" in the body, effectively promoting visual cognitive optimization and improving the training effect. In addition, when the additional movement direction of the target stimulus is small and the recognition difficulty is high, the intensity of electrical stimulation can be increased to strengthen visual cognitive optimization under high difficulty.

[0092] A large number of scientific studies have shown that giving feedback to the training individual on the correct / incorrect recognition situation during the training process helps the training individual unconsciously adjust the recognition strategy and promote functional optimization, especially in the case of high recognition difficulty. Generally, this kind of feedback is mostly given by sound. For example, when the recognition is correct, a specific prompt sound is given; when the recognition is incorrect, no prompt sound is given, or another prompt sound is given. However, considering the aforementioned sensory linkage effect, giving training feedback through current stimulation can actually achieve better results. Therefore, optionally, the medium-frequency current stimulation device can also give training feedback through current stimulation, specifically: in each trial, after the training individual completes the recognition of the target stimulus in the training graph, if the recognition is correct, a current stimulation with a power of the first specified power and a time length of the first specified duration is given; otherwise, no current stimulation is given. Preferably, the first specified power is 2.5W and the duration is 0.5s.

[0093] Optionally, to avoid the confusion of training data among different training individuals and improve the effectiveness and accuracy of training process evaluation, each time of training, the basic information of the training individual can be input through the information input device of the human-computer interaction module and used as an identifier for training data recording, storage, and extraction.

[0094] Based on the information provided by the human-computer interaction module, the training process control module will analyze the training effectiveness of the previous trial that has ended, and combine the recognition situation of the training individual for the training image to send a control signal to the training image generation module to adjust the additional movement direction of the training image generated in the upcoming next trial. The specific methods include:

[0095] S1-1. If, during the presentation of the training image in the previous trial, the training process control module receives the first prompt signal sent by the ambient illuminance monitoring device, or the second prompt signal sent by the distance evaluation unit, or the third prompt signal sent by the eye movement monitoring device, it determines that the training in the previous trial is "invalid", and sends a control signal to the training image generation module to determine that in the next trial, the additional movement direction of the random dot motion stimulus is not adjusted and continues to use the additional movement direction in the previous trial; otherwise, it is determined as "valid" and proceeds to S2-2;

[0096] S1-2. If, in the previous trial, the training individual correctly identifies the direction of the stimulus movement, then in the next trial, the absolute value of the additional movement direction of the motion grating is reduced by 1 minimum change unit; if the identification result in the previous trial is incorrect, then in the next trial, the absolute value of the additional movement direction is increased by the first specified number of minimum change units; where the first specified number is a positive integer greater than 1; preferably, the first specified number is 3.

[0097] It should be noted that if the current trial is the first trial (i.e., there is no previous trial), then the current trial is defaulted to be valid, and the absolute value of the additional movement direction uses the first value; preferably, the first value is 20.

[0098] Figure 3 Shows some valid training trials of a certain training. The circles in the figure represent each valid trial. The abscissa represents the sequential position of the trial among all valid trials, and the ordinate represents the absolute value of the additional movement direction corresponding to the stimulus presented in the trial. The "√" above and "×" below the circles in the figure indicate the identification situation of the training individual regarding the direction of the stimulus movement in that trial. As can be seen from the figure, except for the first 6 trials at the start of the test, in the remaining trials, when the training individual identifies correctly (i.e., "√"), the difficulty level in the next trial is increased by 1 level, that is, the additional movement direction is reduced by 1 degree; when the measured individual identifies incorrectly (i.e., "×"), the difficulty level in the next trial is decreased by 3 levels, that is, the additional movement direction is increased by 3 degrees.

[0099] The above process control rules will inevitably lead to a result. That is, if the performance of the training individual is stable and not affected by any non-sensory factors (such as fatigue, proficiency, rewards and punishments, etc.), and the training volume is large enough, then the absolute value of the additional movement direction presented to the training individual in the middle and late stages of training will inevitably fluctuate around a certain specific level (reference can be made to Figure 3 the situation after the 10th trial in Figure 3 ). Even if there is an occasional deviation from this level (reference can be made to Figure 3in the 32nd - 41st trials). This specific level is actually the training difficulty level that matches the true functional level of the trained individual. For example, Figure 3 When using the training difficulty level that matches the true functional level of the trained individual, it should be between 4 - 7 degrees in the additional movement direction.

[0100] Therefore, Figure 3 The entire training process in can be divided into two stages. The process from the start of training (the 1st trial) to the first change in the trained individual's recognition of the stimulus (the 13th trial) belongs to Stage 1, and all subsequent trials belong to Stage 2. Within Stage 1, the stimulus recognition difficulty level is gradually adjusted from the initial difficulty level to near the training difficulty level, which is the preparation stage of this training. Within Stage 2, the stimulus recognition difficulty level fluctuates near the training difficulty level, which is the stage where the training truly takes effect.

[0101] It is worth mentioning that during a single training session, the training effect will gradually improve the true functional level of the trained individual, but the fatigue effect will gradually reduce the true functional level of the trained individual. Therefore, even without considering the impact of long - term training on the true functional level of the trained individual, it is necessary to realize that during a single training session, the true functional level of the trained individual is also constantly changing. Correspondingly, the training difficulty level that matches the true functional level of the trained individual used in training should also be adjusted and changed accordingly. In fact, using the dynamic adjustment method described in S1 - 2 can ensure that the recognition difficulty of the training images presented to the trained individual in Stage 2 changes with the change of the true functional level of the trained individual.

[0102] Theoretically, it can be predicted that within Stage 2, when the absolute value of the additional movement direction of the presented training image reaches the true level of the trained individual, the following relationship should hold:

[0103] p correct ×Step down，correct =p incorrect ×Step up，incorrect

[0104] where p correct is the probability that the trained individual can correctly recognize the stimulus when the absolute value of the additional movement direction reaches the true level of the trained individual, Step down，correct is the degree to which the absolute value of the additional movement direction decreases in the next trial after the trained individual correctly recognizes it, p incorrect is the probability that the trained individual fails to correctly recognize the stimulus when the absolute value of the additional movement direction reaches the true level of the trained individual, Step up，incorrectIt is the degree to which the absolute value of the additional motion direction increases in the next trial when the training individual fails to correctly identify. On this basis, combined with the dynamic adjustment method described in S1-2, "If in the previous trial, the training individual correctly identified the direction of the motion stimulus, then in the next trial, the absolute value of the additional motion direction of the motion grid decreases by 1 minimum change unit; if the identification result in the previous trial was incorrect, then in the next trial, the absolute value of the additional motion direction increases by the first specified number of minimum change units; where the first specified number is a positive integer greater than 1; preferably, the first specified number is 3", the above formula for p can be derived correct = 3 × p incorrect Based on p correct + p incorrect = 100% established fact, p correct = 3 / (3 + 1) = 75.0% can be further derived. Therefore, regardless of how the true functional level of the training individual changes, the recognition difficulty of the training images presented to the training individual in the second stage always remains near the level of "the recognition accuracy rate of the training difficulty corresponding images is 75% based on the current true functional level of the training individual".

[0105] Due to the repeated fluctuations of the current true functional level of the training individual and the existence of various errors, the training difficulty used in actual training generally shows a certain degree of fluctuation. Therefore, statistical methods are usually used to calculate the arithmetic mean of the additional motion directions corresponding to all data points that "change from correct recognition to incorrect recognition" ( Figure 3 represented by the red "×"-marked circles in) and "change from incorrect recognition to correct recognition" ( Figure 3 represented by the red "√"-marked circles in) during the fluctuation process, so as to infer the average level of the training difficulty. From statistical laws, the more such data points there are, the more accurate the average level of the training difficulty calculated based on these data points will be.

[0106] Generally speaking, the above process control method, data analysis rules, and sufficient training volume can ensure the stability and effectiveness of the training difficulty, and thus ensure the training effect.

[0107] The data recording and analysis module will record the key information of all trials during the training process, including the absolute value of the additional motion direction in the training image, the prompt information sent by the monitoring device for abnormal situations, and the recognition situation of the training individual for the training image. After all training trials are completed, the data recording and analysis module will further conduct data analysis. The specific steps include:

[0108] S2-1. Based on the recorded data, list the additional motion directions used for the motion grid stimuli in all "valid" trials, as well as the correct / incorrect situation of the training individual's recognition of the training image;

[0109] S2-2. Based on the list sorted out in S2-1, select the values at the points where the change trend of the absolute value of the additional movement direction changes during the training process, and record them all as C. cri ;

[0110] S2-3. Among all C cri , in the order of appearance time, eliminate the first 3 or the first 4 C cri , so that the number of the remaining C cri is an even number, and then take the arithmetic mean of all the remaining C cri . The result is the threshold of the ability to distinguish the movement direction of the movement grid stimulus obtained in this training.

[0111] The C mentioned in S2-2 cri， i.e. Figure 3 is the additional movement direction at the place where the red circle is drawn. In addition, for the analysis method of taking the arithmetic mean of C cri mentioned in S2-3, its basic principle has been explained in the previous text. However, in actual operation, due to reasons such as the training individual's unfamiliarity with the training method and process and possible operation errors, the first few C cri obtained during the training are all less reliable than the subsequent C cri . Therefore, on the premise of ensuring that there are enough C cri , the first few C cri can be eliminated in S3-3 to improve the accuracy of the analysis result. In addition, the purpose of "eliminating the first 3 or the first 4 C cri , so that the number of the remaining C cri is an even number" is also to ensure that the number of the two types of data points, namely the data points "changing from correct identification to incorrect identification" and the data points "changing from incorrect identification to correct identification", in the remaining C cri is the same, so as to improve the accuracy of the analysis result.

[0112] In addition, it should be noted that the "threshold of the ability to distinguish the movement direction of the movement grid stimulus obtained in this training" mentioned in S2-3 is the training difficulty level used in this training and matching the true functional level of the training individual. Since this level matches the true functional level of the training individual, it can be gradually improved with the progress of long-term training.

[0113] Based on the general laws of functional training, it can be known that in long-term training, the entire training process can generally be divided into three stages. At the beginning of the training, the functional level of the training individual improves rapidly, and this rapid improvement stage can be regarded as the first stage. Subsequently, the improvement of the functional level enters a slow growth stage, and this stage can be regarded as the second stage. Finally, the functional level enters a plateau period, and even if training continues, no significant increase in the functional level can be observed, and this stage can be regarded as the third stage.

[0114] Therefore, the data recording and analysis module can also further evaluate the overall training process of the training individual in long-term training based on the "threshold of the ability to distinguish the movement direction of the movement grid stimulus obtained in this training" obtained in S2-3. The specific steps include:

[0115] S2-4. List all the thresholds of the ability to distinguish the movement direction in this and previous records. If the quantity is not less than the second specified quantity, then evaluate the overall training process; during the evaluation, first use the following formula for data fitting:

[0116] lg(threshold n )=c×lg(n)+d (Formula 1)

[0117] lg(threshold n )=a×0.5 lg(n) +b (Formula 2)

[0118] Among them, threshold n is the threshold of the ability to distinguish the movement direction obtained during the nth training in this and previous multiple trainings; n is a positive integer greater than or equal to 1; a, b, c, and d are all parameters to be fitted; preferably, the second specified quantity is 8;

[0119] S2-5. Evaluate the results of data fitting using Formula 1 and Formula 2 in S2-4; if the goodness of fit of the data obtained when fitting with Formula 1 is better than the goodness of fit of the data obtained when fitting with Formula 2, and the data fitting using Formula 1 is statistically significant, then determine that "the training process is in the first stage" and end the evaluation; if the goodness of fit of the data obtained when fitting with Formula 1 is better than the goodness of fit of the data obtained when fitting with Formula 2, but the data fitting using Formula 1 is not statistically significant, then determine that "the training process is in the third stage" and end the evaluation; if the goodness of fit of the data obtained when fitting with Formula 2 is better than the goodness of fit of the data obtained when fitting with Formula 1, and the data fitting using Formula 2 is statistically significant, then preliminarily determine that "the training process enters the second stage" and enter S2-6;

[0120] S2-6. Based on the latest obtained third specified quantity of the discrimination ability threshold of the movement direction, re-use the formula in S2-4 for data fitting. If the goodness of fit of the data obtained when fitting with Formula 1 is better than that obtained when fitting with Formula 2 at this time, but the data fitting performed using Formula 1 is not statistically significant, then determine that "the training process enters the third stage"; otherwise, officially determine that "the training process enters the second stage"; preferably, the third specified quantity is 6.

[0121] It should be noted that Formula 1 indicates a linear relationship between the number of training times and the functional level, and Formula 2 indicates an exponential function relationship between the number of training times and the functional level. Correspondingly, Formula 1 can better match the situation in "Stage 1" of the training process, while Formula 2 can better match the situation of "Stage 1 + Stage 2". Therefore, if the result of data fitting shows that the goodness of fit of Formula 1 is better than that of Formula 2, it can be determined that the current training process is still in the first stage. If the result of data fitting shows that the goodness of fit of Formula 2 is better, the training process must have exceeded "Stage 1". However, whether it is in "Stage 2" or "Stage 3" at this time still requires an analysis of the latest obtained part of the training data. If the latest obtained training data shows the characteristics of linear level change (that is, "the goodness of fit of the data obtained when fitting with Formula 1 is better than that obtained when fitting with Formula 2, but the data fitting performed using Formula 1 is not statistically significant" mentioned in S2-6), it indicates that the functional level no longer improves with the training process, and the training enters "Stage 3"; otherwise, it can be determined that the current training process is still in "Stage 2".

[0122] When performing data fitting, the calculation method of the goodness of fit is as follows:

[0123]

[0124] where r 2 is the goodness of fit of the data, data theory is the data predicted by the model (here, the linear model corresponding to Formula 1 or the exponential function model corresponding to Formula 2) (in this embodiment, that is, the discrimination ability threshold of the movement direction that should be obtained in different training sessions), data measured is the discrimination ability threshold of the movement direction obtained during actual training, and mean(·) is the function of taking the average value.

[0125] The above is only a preferred specific embodiment of the present invention, but the protection scope of the present invention is not limited thereto. Any person skilled in the art within the technical scope disclosed by the present invention, according to the technical solution of the present invention and its inventive concept, makes equivalent substitutions or changes, and should be covered by the protection scope of the present invention.

Claims

1. A training system for the ability to distinguish the direction of movement, characterized in that, Including: A training parameter setting module, which is used to set parameters related to training, including the reference motion direction, motion speed, minimum change unit of the motion direction change amount, and the training amount of a single training; A training image generation module, which generates training images based on the parameters obtained by the parameter acquisition module, the system built-in parameters, and the motion direction change amount transmitted by the process control module; A human-computer interaction module, including a display device, an information input device, a sound prompt device, and a monitoring device; the display device is used to present the generated training images to the training individual, and the information input device is used to obtain the recognition situation of the training individual for the training images; the sound prompt device is used to feedback the correct / incorrect situation of the training image recognition to the training individual; the monitoring device is used to monitor the training environment and the training status of the training individual, and can remind the training individual to correct abnormal situations through the sound prompt device; A training process control module, which is used for the sequential control between different trials in a single training process. Specifically, it includes: analyzing the training effectiveness of the previous trial that has ended based on the environment and training status information obtained by the monitoring device, and combining the recognition situation of the training individual for the training images, sending a control signal to the training image generation module to adjust the additional motion direction of the training images generated in the upcoming next trial; and so on in a cycle until the training ends; A data recording and analysis module, which is used for the recording and analysis of the training data of this time, and can evaluate the overall training process based on all the training data recorded this time and before; 2. The training system for the ability to distinguish the direction of movement according to claim 1, characterized in that, The training images generated by the training image generation module are motion grids formed by superimposing two motion gratings with a 90-degree difference in motion direction; 3. The training system for the ability to distinguish movement directions according to claim 2, characterized in that After receiving the additional motion direction transmitted by the process control module, the visual stimulus generation module will randomly select one of the two situations: "the motion direction of the newly generated motion stimulus deflects clockwise by the additional motion direction relative to the reference motion direction" and "the motion direction of the newly generated motion stimulus deflects counterclockwise by the additional motion direction relative to the reference motion direction" to generate a motion stimulus; 4. The training system for the ability to distinguish the direction of movement according to claim 2, characterized in that, The monitoring device includes an environmental illuminance monitoring device, a training distance monitoring device, and an eye movement monitoring device; The environmental illuminance monitoring device monitors the illuminance of the training environment in real time through an illuminance sensor; when the relative change amount of the environmental illuminance compared with the environmental illuminance at the start of training exceeds the first specified ratio, it sends a first prompt signal to the training process control module in real time, and sends a sound prompt of "correct the environmental illuminance" to the training individual through the sound prompt device; The distance evaluation unit monitors the distance between the display device and the training individual in real time through a distance sensor; when the monitored distance exceeds the second specified ratio of the established training distance, it sends a second prompt signal to the training process control module in real time, and sends a sound prompt of "correct the training distance" to the training individual through the sound prompt device; An eye movement monitoring device is used to monitor the line-of-sight direction of the individual being trained when a training image appears; when it is found that the angle of the line-of-sight direction of the individual being trained deviates from the training image by more than the first specified viewing angle, a third prompt signal is sent to the training process control module in real time, and a voice prompt of "correct the line-of-sight direction" is sent to the individual being trained through a voice prompt device.

5. The training system for the ability to distinguish movement directions according to claim 4, characterized in that, The training process control module analyzes the training effectiveness of the previous trial that has ended, and combines the identification situation of the individual being trained with respect to the training image, and sends a control signal to the training image generation module to adjust the specific method of the additional movement direction of the training image generated in the upcoming next trial, including: S1-1. If, during the presentation of the training image in the previous trial, the training process control module receives the first prompt signal sent by the ambient illuminance monitoring device, or the second prompt signal sent by the distance evaluation unit, or the third prompt signal sent by the eye movement monitoring device, it is determined that the training in the previous trial is "invalid", and a control signal is sent to the training image generation module to determine that in the next trial, the absolute value of the additional movement direction of the training image is not adjusted and continues to use the absolute value of the additional movement direction in the previous trial; otherwise, it is determined to be "valid", and proceed to S2-2; S1-2. If, in the previous trial, the individual being trained correctly identified the direction of the stimulus movement, then in the next trial, the absolute value of the additional movement direction of the movement grid is reduced by 1 minimum change unit; if the identification result in the previous trial is incorrect, then in the next trial, the absolute value of the additional movement direction is increased by the first specified number of minimum change units; where the first specified number is a positive integer greater than 1.

6. The training system for the ability to distinguish the direction of movement according to claim 5, characterized in that, The initial value of the minimum change unit of the additional movement direction is 5.0 degrees. After completing the measurement of the first 3 trials, it is adjusted to 1 / 5 of the initial value, that is, 1.0 degree. After completing the measurement of another 3 trials after the adjustment, it is further adjusted to 1 / 10 of the initial value, that is, 0.5 degree.

7. The training system for the ability to distinguish the direction of movement according to claim 5, characterized in that, The steps for the data recording and analysis module to analyze the training data of this time and evaluate the overall training process include: S2-1. Based on the recorded data, list the absolute values of the additional movement directions used in all "valid" trials, as well as the correct / incorrect situation of the individual being trained's identification of the training image; S2-2. Based on the list sorted out in S2-1, select the values at the points where the absolute value change trend of the additional movement direction changes during the training process, and record them all as C cri ; S2-3. Among all Cs cri in chronological order of appearance, eliminate the first 3 or the first 4 Cs cri so that the number of remaining Cs cri is even, and then take the arithmetic mean of all the remaining Cs cri The result is the threshold of the motion direction discrimination ability in the current reference motion direction obtained from this training; S2-4. List all the movement direction discrimination ability thresholds in this and previous records. If the number is not less than the second specified number, then evaluate the overall training process; during the evaluation, first use the following formula for data fitting: lg(threshold n ) = c×lg(n) + d (Formula 1) lg(threshold n ) = a × 0.5 lg(n) + b (Formula 2) where threshold n is the threshold of the motion direction discrimination ability obtained in the nth training in this and previous multiple trainings; n is a positive integer greater than or equal to 1; a, b, c, and d are all parameters to be fitted; S2-5. Evaluate the results of data fitting using Formula 1 and Formula 2 in S2-4; if the goodness of fit of the data obtained when using Formula 1 for fitting is better than that obtained when using Formula 2 for fitting, and the data fitting using Formula 1 is statistically significant, then determine that "the training process is in the first stage" and end the evaluation; if the goodness of fit of the data obtained when using Formula 1 for fitting is better than that obtained when using Formula 2 for fitting, but the data fitting using Formula 1 is not statistically significant, then determine that "the training process is in the third stage" and end the evaluation; if the goodness of fit of the data obtained when using Formula 2 for fitting is better than that obtained when using Formula 1 for fitting, and the data fitting using Formula 2 is statistically significant, then preliminarily determine that "the training process enters the second stage" and proceed to S2-6; S2-6. Based on the latest obtained third specified number of motion direction discrimination ability thresholds, re-use the formula in S2-4 for data fitting. If the goodness of fit of the data obtained when using Formula 1 for fitting is better than that obtained when using Formula 2 for fitting at this time, but the data fitting using Formula 1 is not statistically significant, then determine that "the training process enters the third stage"; otherwise, officially determine that "the training process enters the second stage".

8. The training system for the ability to distinguish the direction of movement according to claim 2, characterized in that, The human-computer interaction module further includes a low-frequency modulated medium-frequency current stimulation device, which can generate a low-frequency modulated medium-frequency current and act on the acupoints around the eyes during the presentation of the training image.

9. The training system for the ability to distinguish the direction of movement according to claim 8, characterized in that, When the low-frequency modulated medium-frequency current acts on the acupoints around the eyes, the output power is related to the additional motion direction in the training image, and the formula is: P = P max -(P max -P min )×ΔD where P min and P max are the minimum and maximum values of the output power respectively, and ΔD is the angle by which the movement direction of the movement stimulus is deflected relative to the reference movement direction, i.e., the additional movement direction.

10. The training system for the ability to distinguish the direction of movement according to claim 9, wherein The medium-frequency current stimulation device can also give training feedback to the training individual through current stimulation, specifically: in each trial, when the training individual completes the identification of the training image, if the identification is correct, then give the training individual a current stimulation with a power of the first specified power and a time length of the first specified duration; otherwise, no current stimulation is given.

Images