阅读说明：本技术 用于评估视觉反应测试中的结果的可靠性的方法和系统 (Method and system for evaluating reliability of results in visual response testing ) 是由 塔皮奥·马里亚迈基 马库·雷诺宁 J·索米宁 I·蒂吉拉尔 P·雅维宁 E·马塔 K·彭 于 2018-12-21 设计创作，主要内容包括：本发明涉及用于对用户进行的视觉反应测试中的结果的可靠性进行评估的方法和系统。具有以下阶段的方法步骤重复多次：-在刺激时间点向用户显示具有可识别值的视觉测试刺激；-当用户识别出刺激值时,在报告时间点报告；-在视觉刺激之间插入可变且相对短的延迟时间；-记录刺激时间点与报告时间点之间的反应时间；-步骤时间计算为每个步骤的延迟时间与反应时间的和。将步骤时间与延迟时间之间的相关性或反应时间与延迟时间之间的相关性与可靠性的预定义的相关性值进行比较,从而确定结果的可靠性水平。(The present invention relates to a method and system for assessing the reliability of results in a visual response test performed by a user. The method steps with the following phases are repeated a number of times: -displaying a visual test stimulus with an identifiable value to a user at a stimulus time point; -reporting at a reporting time point when the user identifies a stimulus value; -inserting a variable and relatively short delay time between visual stimuli; -recording the reaction time between the stimulation time point and the reporting time point; step time is calculated as the sum of the delay time and the reaction time of each step. The correlation between the step time and the delay time or the correlation between the reaction time and the delay time is compared to a predefined correlation value of reliability, thereby determining the reliability level of the result.)

1. A method for assessing the reliability of results in a visual response test performed by a test user, the method comprising at least the steps of:

starting the test;

repeating a plurality of test steps until a stopping criterion is met, a single test step comprising the following phases:

1) selecting a delay time dt (step) for the step, the delay time dt (step) varying between steps;

2) generating no visual test stimulus to be identified within the delay time of the step;

3) generating a visual test stimulus at the stimulus time point stp (step) of said step;

4) receiving an identification of said stimulus at a reporting time point rtp (step) of said step when reported by a user;

5) starting a new testing step by returning to phase 1) until a stopping criterion of the test is met;

the method further comprises the following steps:

and (3) calculating:

-a reaction time of step rt (step), which is the time elapsed between the stimulation point in time of the step and the reporting point in time of the step, i.e. rt (step) -rtp (step) -stp (step), and

-a step time st (step) of a step, which is the sum of the delay time of the step and the reaction time of the step, i.e. st (step) ═ dt (step) + rt (step);

determining a correlation between the step time and the delay time for each test step or a correlation between the reaction time and the delay time for each test step;

the determined correlation is compared to a predefined correlation value of reliability, thereby determining a reliability level of the result and a consistency of the response of the test subject.

2. The method according to claim 1, characterized in that it comprises:

determining a correlation between the step time and the delay time for each test step and a correlation between the reaction time and the delay time for each test step;

the determined correlation is compared to a predefined correlation value of reliability, thereby determining a reliability level of the result and a consistency of the response of the test subject.

3. The method of any preceding claim, further comprising:

providing a pseudo stimulus or another stimulus not intended to be identified in the test during the delay time of the step.

4. Method according to any preceding claim, characterized in that the correlation is calculated from a limited set of steps, for example only from the steps of correctly identifying a stimulus, or only from the steps of the reaction time rt (step) or the step time st (step) or the delay time dt (step) meeting certain criteria.

5. The method of any preceding claim,

the reliability of the response during the test or during a subset of the test steps is also determined by the number of wrong answers associated with the correct answer.

6. Method according to any of the preceding claims, characterized in that for better test-retest repeatability the reported actual reaction time RT of the test object is determined by using the fastest subset piece of test steps, e.g. by dividing the series of steps into sub-blocks and selecting the fastest reliable sub-block.

7. The method of any preceding claim, wherein the method comprises:

determining, for all test steps or for any subset of test steps, a correlation coefficient ST/dt (step) between the step time and the delay time of the step;

comparing the calculated correlation coefficient to a predefined value of reliability, thereby determining a level of reliability of the result and a consistency of the response of the test subject.

8. The method of any preceding claim,

generating the visual test stimulus having a characteristic that includes a visually identifiable value; and

when reported by a user, an identification of a value of a characteristic of the stimulus is received.

9. The method of claim 8,

selecting a value of a characteristic of the visual test stimulus from a group comprising at least two different values.

10. The method of claim 9, wherein the group of at least two different values comprises values "left" and "right".

11. The method of any preceding claim,

the user report is completed by selecting and pressing a report button.

12. The method of claim 11, wherein the user selects a button from a group comprising at least two different reporting buttons, whereby each button is associated with one possible value of the characteristic of the stimulus.

13. A method according to any preceding claim, wherein the visual test stimulus comprises a symbol having a visually identifiable value indicative of direction, such as an arrow or a standard landolt c.

14. The method according to any preceding claim, characterized in that the test user reports the value of the visual test stimulus or the characteristic of the visual test stimulus, such as stimulus direction, by pressing a button, a touch element, a movement of a body or body part, eye movement, oral feedback, magnetic or electrical signals from the brain or nervous system.

15. The method of any preceding claim, wherein incorrect recognition of the visual test stimulus by the test user generates feedback or a signal to the test user indicative of an incorrect response to encourage a correct response.

16. The method of any preceding claim,

generating the visual test stimulus on an electronic screen.

17. The method of claim 16,

if a visual test stimulus is correctly identified, the next visual test stimulus is generated at a different location on the screen.

18. The method of any preceding claim,

defining that the stopping criterion is fulfilled when a number of consecutive steps, e.g. 3-10, 4-8 or 5-7, correctly identifying the stimulus is recorded.

19. The method of any preceding claim,

the delay time of the steps varies randomly, or by any predefined order or by an algorithm, between a lower limit and an upper limit of the delay time.

20. The method of claim 19,

varying the lower and/or upper limit of the delay time between the testing steps according to an observed reaction time of the user or according to other predefined criteria.

21. The method of claim 19 or 20,

the lower limit is selected to be between 50ms and 500ms or between 65ms and 200ms and the upper limit is selected to be between 200ms and 1000ms or between 250ms and 800 ms.

22. The method of any preceding claim,

the reaction times of the steps with shorter delays are summed, averaged or otherwise grouped together as rt (shortdelays), independently of this,

summing, averaging or otherwise grouping together reaction times delayed by steps longer than the step grouped to the RT (ShortDelays) as RT (LongDelays),

further analysis is performed using the absolute values of the rt (shortdelays) and rt (longdelays) or the difference or relationship between the rt (shortdelays) and the rt (longdelays), e.g. using the relationship between the rt (shortdelays) and the rt (longdelays), determining the degree to which the test user's attention is restored from the reporting task of the previous step or determining what the test user's reaction time is without the influence of the previous step, i.e. rt (longdelays).

23. A method according to any preceding claim, wherein the delay time is shortened until the reaction time starts to slow down in order to determine the minimum time required for a person to recover from a previous step to focus on a new stimulus.

24. A system for assessing reliability of results in a visual response test performed by a user, comprising:

-a computer having a memory;

-a computer program executable on said memory, said program comprising program code elements adapted to perform and control said visual response test;

-display means arranged to display visual test stimuli as instructed by the computer program;

-reporting means arranged to receive an identification of the stimulus by a user and arranged to communicate information of the identification to the computer;

characterized in that said program code element is adapted to perform the method steps according to any of the preceding claims 1 to 23.

Technical Field

The present invention relates to a method and system for assessing the reliability of results in a visual response test. More particularly, the present invention relates to a novel method for determining whether a visual response testing user is fully attentive, cheating, or not fully informed of the principles of the testing operation.

Background

Measuring characteristics of human vision requires that cooperating test personnel handle the means for assessing the structure of the eye or visual system or measuring electrical responses from neural pathways. In visual measurements, standardized visual stimuli are typically displayed for detection or identification by a test person. The tester typically gives a response to the visual stimulus using a button or orally. This response may be saved manually or automatically and the next stimulus presented (easier or more difficult) based on the response.

This methodology is known as psychophysics. The most important part of this methodology is the nature of the visual stimulus, which must be designed to fit a theoretical frame of reference. Another key component in order to obtain accurate results from psychophysical measurements is the cooperation of the test personnel: whether s/he can respond consistently based on visual stimuli within a reasonable time. In the case of dementia or cognitive disorders, cognitive performance may be so slow or variable that it affects the accuracy of the measurement. Slow responses to visual stimuli can also impair visual performance, i.e., inability to take rapid action based on visual information, as compared to faster people.

When a person performs a visual response time test, he/she should concentrate on the visual stimulus entirely and respond based on the stimulus only when the visual stimulus appears, rather than think about other things and press a response button in a rhythm that is not based on the appearance of the visual stimulus. This may occur, for example, in the following cases: a person with a head that has suffered a trauma is ill or cheating and attempts to deliberately obtain an abnormally slow result in order to convince the insurance company that he has suffered a brain injury and suffered a traumatic brain injury. The person may experience concentration difficulties which may lead to concentration difficulties in conducting the reaction time test. Problems also arise if one does not fully understand the testing principles. In this case, the reaction time test will not be able to measure the response to the visual stimulus.

Patent document US9095295B2 proposes a method for measuring the visual perception velocity of a user, WO2017/024845a1 proposes a stimulus information editing method, and US2005/0143629a1 and US5957859 propose methods for detecting spoofing.

Disclosure of Invention

It is an object of the present invention to reduce or even eliminate the above-mentioned problems occurring in the prior art.

It is an object of the present invention to obtain reliable results in a simple and easy to use visual response test.

It is an object of the present invention to provide a simple and effective method to determine whether a visual response test user is fully attentive, cheating or not fully aware of the operational principles of the test.

The invention has the object, inter alia, of providing a test of the visual response time, wherein for the purpose of making a judgment on the visual performance of a person, we know the cognitive response time for a very basic and easy visual stimulus, so that the limiting factor of the test is not the visual perception of the test stimulus even in case of a defect of the visual system.

Terms and abbreviations

Ocussweep apparatus	Ocuspecto's unique multiple fixed perimeter, reaction time perimeter
		OccurT test	Ocussweep reaction time test
RT or RealRT	Reaction time of the test step
		DT	Delay time of test step
STP	Stimulation time points of the test procedure
		RTP	Reporting time point of test step
ST or StepTime	Step time of the test step
		ST/DT	Step (a) ofCorrelation coefficient between step time and delay time
RT/DT	Coefficient of correlation between reaction time and delay time of step

Furthermore, to achieve the above objects and in accordance with other objects of the present invention, a method and system for assessing the reliability of results in a visual response test is presented by the appended independent claims.

The embodiments, examples and advantages mentioned herein relate, where applicable, to the method and system according to the invention, even if not specifically mentioned.

An exemplary method for assessing the reliability of results in a visual response test conducted by a test user according to the present invention comprises:

starting the test;

repeating a plurality of test steps until a stopping criterion is met, a single test step comprising the following phases:

1) selecting a delay time DT (step) of the steps, DT (step) varying between steps;

2) no visual test stimulus to be identified is generated within the delay time dt (step) of the step;

3) generating a visual test stimulus at the stimulus time point stp (step) of step;

4) receiving an identification of the stimulus at a reporting time point rtp (step) of the step when reported by the user;

5) a new test step is started by returning to phase 1) until the stopping criterion of the test is fulfilled.

The method further comprises the following steps:

and (3) calculating: step reaction time rt (step), which is the time elapsed between the stimulation time point of the step and the reporting time point of the step, i.e., rt (step) ═ rtp (step) -stp (step), and step time st (step), which is the sum of the step delay time and the step reaction time, i.e., st (step) ═ dt (step) + rt step;

determining a correlation between the step time and the delay time or a correlation between the reaction time and the delay time for each test step;

the correlation is compared to a predefined correlation value for reliability, thereby determining a reliability level of the result and a consistency of the response of the test subject.

One way of describing the present invention is to define the present invention as it relates to a method and system for assessing the reliability of results in a visual response test performed by a user. The method steps with the following phases are repeated a number of times:

-displaying a visual test stimulus with an identifiable value to a user at a stimulus time point;

-at a reporting time point, the user reports when he recognizes the stimulus value;

-inserting a variable and relatively short delay time between visual stimuli;

-recording the reaction time between the stimulation time point and the reporting time point;

step time is calculated as the sum of the delay time and the reaction time of each step.

The correlation between the step time and the delay time or the correlation between the reaction time and the delay time is compared with a predefined correlation value for reliability, thereby determining the reliability level of the result.

In an embodiment of the invention, the method comprises:

determining a correlation between the step time and the delay time and a correlation between the reaction time and the delay time for each test step;

the determined correlation is compared to a predefined correlation value for reliability, thereby determining a reliability level of the result and a consistency of the response of the test subject.

It was thus found that to ensure whether a person actually reacts to the presence of a visual stimulus, a variable and relatively short delay time can be inserted before the presence of the visual stimulus.

If the person is fully focused on the occurrence of the visual stimulus his/her reaction time RT is not affected by the length of the delay. In this case, the root of the change in step time ST is due to the change in delay time DT, i.e., there is a statistical relationship between these two variables. Therefore, the corresponding correlation coefficient ST/DT is high.

Table 1. example of simulated response time test, where the test subjects were attentive and therefore consistently answered within 400 milliseconds after the stimulus occurred. This results in a perfect correlation between DT and ST. Both the variance (RT) and the variance (RT)/the variance (DT) are zero. All of which indicate reliable test results. All time units are milliseconds. The total variance and pearson correlation were used in the calculation.

If a person is not fully attentive to the visual stimulus and is thinking about something else (perhaps being ill or cheating, or not fully aware of the principle of the test), he/she tends to press a button at a constant rhythm, ignoring the length of the delay. Thus, the correlation coefficient ST/DT is low.

Table 2. example of simulated reaction time test where the test subjects did not fully attend to and answered consistently except at a single sneeze or other disturbance (2 x 900ms in the RT column) in the middle of the test. This results in a significant increase in the calculated variance to a step where the test can be considered unreliable. The same is true of variance (RT)/variance (DT). However, since there is a fairly linear relationship between DT and ST, the correlation indicates a reliable test case and is therefore a more robust method for isolated errors. All time units are milliseconds. The total variance and pearson correlation were used in the calculation.

Table 3. simulation reaction time example, where test subjects attempt to cheat by answering at regular intervals that are not based on stimulus occurrence. This results in the absence of correlation (DT, ST) and high variance (RT) and variance (RT)/variance (DT). These indicate that the test results are unreliable. All time units are milliseconds. The total variance and pearson correlation were used in the calculation.

Table 4. simulation reaction time example, where the test subjects are learning how to perform the test and are getting faster and faster towards the end of the test. In the latter half of the test, the reaction time of the test subject stabilized between 400-. Since RT became faster during the test, the variance (RT) and variance (RT)/variance (DT) were high, indicating that the test was unreliable. However, since there is a fairly linear relationship between DT and ST, the correlation indicates a reliable test case. All time units are milliseconds. The total variance and pearson correlation were used in the calculation.

The reliability level of the results can also be determined by comparing the variability of the reaction time RT with the variability of the delay time DT, i.e. there is a statistical relationship between the two variables. A corresponding correlation coefficient RT/DT may be determined.

In an embodiment, the method comprises: for these steps in which the stimulus is correctly identified, determining a correlation coefficient ST/DT between the step time and the delay time of the step; the calculated correlation coefficient ST/DT is compared with a predefined value of reliability, thereby determining the reliability level of the result.

Therefore, the reliability of the measurement can be judged by the correlation coefficient ST/DT or RT/DT. If the correlation is high enough, the measurement can be considered reliable.

The correlation may represent any statistical association. In one embodiment, correlation refers to the degree to which variables are close to each other with a linear relationship (i.e., linear correlation). Embodiments of the present invention use pearson correlation coefficients. In one embodiment, the correlation refers to a non-linear correlation. An embodiment uses rank correlations, such as spearman rank correlation coefficients or kender rank correlation coefficients. In one embodiment, correlation refers to any mathematical function derived from DT and RT, or DT and ST, or RT and ST.

In order to find a predefined correlation value for reliability, i.e. a critical value for the reliability measure, the correlation coefficient may be studied in repeated measures within a test room to find the effect of learning. In addition, the correlation coefficient can be studied in a person familiar with the test and the results compared to the results obtained during the increase of cognitive load and deliberate illness or cheating.

In an embodiment of the invention, a dummy stimulus or another stimulus not desired to be identified in the test is provided during the delay time of the step. Such stimuli are intended for example to test whether the test user is fully attentive to the test.

In an embodiment, the correlation is computed from a limited set of steps. For example, the correlation may be calculated only from steps that correctly identify the stimulus, or only from steps that meet certain criteria, either from reaction time rt (step) or step time st (step) or delay time dt (step).

In an embodiment, the reliability of the response during the test or during the subset of test steps is further determined by the number of wrong answers related to the correct answer.

In an embodiment, for better test-retest repeatability, the reported actual reaction time RT of the test subject is determined by using the fastest subset segment of the test steps, e.g. by dividing the series of steps into sub-blocks and selecting the fastest reliable sub-block. Fastest, for example, refers to a test or test segment or sub-block of tests that the test user responds acceptably in the shortest amount of time.

In an embodiment, the method comprises:

determining a correlation coefficient ST/dt (step) between step times and delay times of the steps for all or any subset of the test steps;

the calculated correlation coefficient is compared to a predefined value of reliability to determine a reliability level of the result and a consistency of the test subject response.

In an embodiment of the present invention, the generated visual test stimulus comprises a characteristic having a visually recognizable value. The test user must at least concentrate on identifying the value. When the user recognizes the value, s/he reports the value, for example, with a reporting device.

The delay time of the steps may vary randomly or may vary between a lower and an upper limit of the delay time by any predefined order or by an algorithm. The lower and/or upper limits may vary between test steps. This may be done according to the user's reaction time or according to other predefined criteria. The lower and upper limits may be increased if the user is slow to react. The lower and upper limits may be reduced if the user reacts quickly.

In an embodiment, the lower limit is selected to be between 50ms and 500ms or between 65ms and 200 ms. In an embodiment, the upper limit is selected to be between 200ms and 1000ms or between 250ms and 800 ms.

In an embodiment, the user completes the report by pressing a report button. In an embodiment of the invention, the value of the stimulation property is selected from the group comprising at least two different values. The buttons may be selected from a group comprising at least two different reporting buttons, whereby each button is associated with one possible value of the stimulation characteristic.

The visual test stimulus may include a symbol having a visually recognizable value indicating a direction, such as an arrow or a standard landolt c. In an embodiment, the group of at least two different values comprises the values "left" and "right". These are simple and reliable symbols and values.

In an embodiment, the test user reports the value of the visual test stimulus or the visual test stimulus characteristic, such as the stimulus direction, by a button, a touch element, a motion of a body or body part, eye motion, oral feedback, a magnetic or electrical signal from the brain or nervous system.

In an embodiment, incorrect recognition by the test user of the value of the visual test stimulus or visual test stimulus characteristic will generate a false response feedback or signal to the test user indicating a false response, thus encouraging a correct response. The feedback may also be sound, vibration or light or any combination thereof.

In an embodiment of the invention, the visual test stimulus is generated on an electronic screen. In an embodiment, the visual test stimulus may be displayed at a slightly different location on the screen. In particular, if the stimulus is correctly identified, the next visual test stimulus may be generated at a different location on the screen.

In an embodiment of the invention, the stopping criterion is defined to be fulfilled when a number of consecutive steps of correctly identifying the stimulus are recorded. The number may be, for example, 3 to 10, 4 to 8, or 5 to 7.

In an embodiment of the invention, the reaction times of the steps with shorter delays are summed, averaged or otherwise grouped together as rt (shortdelays). Independently of this, the reaction times of steps delayed longer than the step rounded to rt (shortdelays) are summed, averaged or otherwise rounded together as rt (longdelays). The absolute values of RT (ShortDelays) and RT (LongDelays) or the differences or relationships between them were used for further analysis. They can be used to determine the extent to which the test user's attention is restored from the reporting task of the previous step (using the relationship between rt (shortdelays) and rt (longdelays)) or to determine what the reaction time is without the influence of the previous step, i.e. rt (longdelays).

In an embodiment, the invention may be used to determine the minimum time required for a person to recover from a previous step to focus on a new stimulus. This is done by shortening the delay until the reaction time starts to slow down.

The present invention may be implemented as an evaluation system for evaluating reliability of results in a visual response test performed by a user. The system comprises:

-a computer having a memory;

-a computer program executable on a memory, the program comprising program code elements adapted to perform and control a visual response test;

-display means arranged to display the visual test stimulus as instructed by the computer program;

-reporting means arranged to receive the identification of the stimulus by the user and arranged to transmit information of said identification to the computer.

The system, in particular a computer program having program code elements, is adapted to perform the method steps according to embodiments mentioned herein.

Drawings

The invention is described in more detail below with reference to the appended schematic drawings, in which,

FIG. 1 shows the correlation of step times for a first round of OccurT measurement of OccurT 1;

FIG. 2 shows the correlation of step times for a second round of OccurT measurements OccurT 2;

FIG. 3 shows the correlation of step times for a third round of OcURT measurement OcURT 3;

FIG. 4 shows the correlation of the OccurT test during increasing cognitive load;

FIG. 5 illustrates the correlation of the OccurT test during intentional cheating;

FIG. 6 illustrates an example of softening a standard Landolt C-target;

fig. 7 shows an example of an error response indicator image.

Detailed Description

Example 1: ocussweep reaction time test (OccurT test)

To find the critical value of reliable measurement, we studied the correlation coefficient of repeated measurement in a test subject to understand the learning effect. Furthermore, we compared the correlation coefficient of the test skilled person with the results obtained during increased cognitive load and deliberate loading or cheating.

The test subjects sit in front of the ocusweet device and the ocusweet tablet at close distances and then give them the ocusweet remote control. In the test, the tablet displayed a series of black rings with openings on the left or right side. The test subject is asked to press the left or right button of the remote control depending on which side the opening is located and tell them to do so as soon as possible. They are also instructed to answer only when they see an opening, not to guess blindly. As a result of the false answer, a red circle appears on the tablet computer, which disappears upon a correct button press. The next symbol is displayed only after a random delay period of 100ms to 400 and 560ms, depending on the individual reaction time of the subject. After three rounds (called OcuRT1, OcuRT2, OcuRT3), the test was terminated and the shortest median reaction time of six consecutive correct answers was calculated and displayed on the tablet screen.

In a variation of this test, referred to as "cognitive load," test subjects were told to count down from 200 to 7 seconds as they performed the reaction time test. Indicating that they are not to count between two rounds and start counting again when a new round is started. The test subjects are also told to start counting again at 200 if they forget which number to count from in the middle of the test, or if they count to zero.

In another variation, called "cheating," the test object is told the following mind: imagine that you hit the head and you will see the doctor to test your reaction time. However, you want to get some money from your insurance company and answer slowly on purpose, as if head trauma had affected your reaction time. In summary, you should answer slowly but in a plausible way to cheat the insurer's money.

The reliability of the OcuRT measurement is classified by means of the correlation coefficient ST/DT. If one is fully focused on the appearance of visual stimuli, the reaction time is not affected by the variation of the random delay, but the variation can be detected in the total step time, resulting in a high correlation value.

Fig. 1 shows the correlation coefficient of the delay of the first round OcuRT measurement OcuRT1 for each test object (n ═ 44) with respect to the total step time ST (here labeled StepTime, x-axis) and the individual reaction times RT (RealRT, y-axis). Test subjects with legends showed less than 70% correlation with StepTime, meaning that these people did not fully focus on the appearance of visual stimuli, perhaps because of other thoughts or not fully mastering the OcuRT test. Subject No. 13 was unable to perform the test due to discomfort glare caused by the tablet.

Fig. 2 shows the correlation coefficient of the second round of OcuRT measurements OcuRT2 for each test object (n ═ 44) with respect to the total step time (x-axis) and the individual reaction times (RealRT, y-axis). For further explanation, please refer to the description of fig. 1.

Fig. 3 shows the correlation coefficient of the third OcuRT measurement OcuRT3 for each test subject (n ═ 44) with respect to the total step time (x-axis) and the individual reaction times (RealRT, y-axis). For further explanation, please refer to the description of fig. 1.

Fig. 1 to 3 show that practicing the OcuRT test helps to obtain shorter reaction times. The mean reaction time in the study group with a correlation above 70% was about 8ms better than the previous measurements in the replicate measurements (OccurT 1: 460.6 ms; OccurT 2: 451.7 ms; OccurT 3: 443.4 ms).

If a person is not fully attentive to the presence of a test stimulus and identifies and reports his direction by pressing a button, he may be slow to react. To make it more difficult for them to focus on performing the OcuRT test, the cognitive load of the test subjects was increased by telling them to count down from 200 to 7 seconds when they performed the reaction time test. In fig. 4, we can see that all test subjects have a low StepTime correlation, indicating that they cannot concentrate on counting and testing for the presence of stimuli at the same time. None of the testers achieved high StepTime correlation with a maximum of 59% and an average of 23.7%.

Fig. 5 shows the correlation of the OcuRT test during intentional cheating. The test subjects responded slowly but attempted to cheat insurance companies in a plausible manner by being ill with brain damage resulting from head trauma. In the results shown in fig. 5, we can see that only one test subject can achieve a high StepTime correlation score (80%), while the other test subjects focus on cheating rather than visual reaction time testing.

The results of fig. 1 to 5 show that the correlation coefficient ST/DT clearly distinguishes the following OcuRT measurements: these measurements are made with the attention resource fully aimed at testing for the presence of a stimulus and identifying the stimulus and reporting its direction by pressing the button as soon as possible. Unfamiliar with testing, while taking cheating into account or attempting to perform difficult mathematical calculations, would significantly reduce the StepTime correlation coefficient. Based on visual inspection of the above figures, the cut-off value was 70%, which seems to be a sign of a reliable OcuRT test.

Example 2: example of OccurT test Algorithm

Some possible examples of OcuRT are as follows.

Left and right directional images or optotypes (such as standard landolt c) may be used as visual test stimuli. Fig. 6 shows an example of a softening standard landolt c pointing to the right.

After testing each correct response of the user, the visual test stimulus may be moved slightly on the screen, making it more difficult for the user to conclude from the "motion" of the test image direction.

If the test subject responds incorrectly, for example by pressing the wrong direction button, an incorrect response indicator image may be displayed in place of the test image. The error response indicator image may be displayed continuously for, for example, 1-3 seconds or 2.5 seconds. Fig. 7 shows an example of an error response indicator image, i.e. a ring. The ring may be, for example, red.

The response wait timeout time may be set to, for example, 2 to 5 seconds, for example, 4 seconds. If the test user does not respond within this timeout, the algorithm may interpret that the test user has not seen the optotype. In an embodiment, if the test subject does not respond during the timeout period, it is not interpreted as an error, but a red error ring may still be displayed.

Before the test starts, there may be an inverse phonetic number (3-2-1). During this countdown, an image, such as a black ring, may be displayed. The time between reciprocal numbers may be, for example, about 1-3 seconds or 2 seconds. After the countdown, the test subject begins to recognize the visual test stimulus and report, i.e., answer, e.g., by pressing the left or right button for each image.

The test may comprise a plurality of rounds, for example 3 rounds. The end criteria for a round may vary. For example, the end criterion may be 6 correct answers in a sequence of 15 steps. If possible, the round will continue until 4 correct answers or a maximum number of steps occur in succession, for example up to 30 steps in a round of testing. After the end of one round, the countdown of the next round will start immediately. After, for example, three rounds, the test will stop.

The final reaction time may be calculated, for example, as the fastest median of the 6 or 4 correct responses that are the fastest in succession. The end result is considered reliable when the percentage of errors in the report is below a certain limit (e.g., 15%). After the test is finished, the final result is displayed to the user. The results may show, for example, the reaction time, the duration of the test, and the percentage of error calculated in one round of calculating the reaction time.

The background light level may be measured with an ambient light sensor and the test stopped if the light level reaches a predetermined stop value, such as 200cd/m 2.

The drawings and examples only show some preferred embodiments according to the invention. Minor facts about the main idea of the invention, facts known or obvious to a person skilled in the art need not be discussed. It is obvious to the person skilled in the art that the invention is not limited solely to the examples described above, but that it may be varied within the scope of the claims presented below. The dependent claims present some possible embodiments of the invention and they are not to be considered as limiting the scope of protection of the invention.