Psychological resilience estimation method, psychological resilience estimation device, and control program

Abstract

To improve estimation accuracy of psychological resilience.SOLUTION: A psychological resilience estimation method includes: a display step (S13) of displaying, side by side, some pair of images related to facial expression image or some pair of images related to emotion image; a measurement step (S14) of measuring movement of a line of sight of a user; an index acquisition step (S16, S18) of acquiring values of line of sight indices from the measured movement of the line of sight; and an estimation step (S17, S19, S20) of estimating psychological resilience of the user by evaluation based on the value of the line of sight index related to the facial expression image and a first evaluation criterion and evaluation based on the value of the line of sight index related to the emotion image and a second evaluation criterion different from the first evaluation criterion.SELECTED DRAWING: Figure 7

Description

[Technical Field]

[0001] The present invention relates to a psychological resilience estimation method, a psychological resilience estimation device, and a control program for estimating a user's psychological resilience. [Background technology]

[0002] The concept of psychological resilience has been attracting attention in recent years in order to deepen understanding of individual differences in stress vulnerability and predisposition. Psychological resilience, also translated as "resilience" or "psychological homeostasis," is defined as the ability and process to maintain a psychologically healthy state even when experiencing stress, and to overcome illness and recover to a healthy state even if illness occurs. In the following, "psychological resilience" may be simply referred to as "resilience."

[0003] Conventionally, resilience has been assessed mainly by self-administered questionnaires, but such assessments are subjective.

[0004] Therefore, the inventor has been conducting research into objectively evaluating the above-mentioned resilience. For example, Non-Patent Document 1 discloses a method for calculating multiple gaze indices for evaluating attention bias based on gaze data and estimating psychological resilience using the calculated multiple gaze indices. Here, attention bias is a bias shown in people with severe depression and anxiety in the information processing process, and refers to selective attention to emotional stimuli. [Prior art documents] [Non-patent literature]

[0005] [Non-Patent Document 1] Kentaro Kotani and three others, "Psychological Resilience Estimation Method Using Eye-Gaze ​​Data from Tablet Devices," March 25, 2024, Occupational Health and Ergonomics Research, Vol. 24, No. 1, pp. 5-16 Summary of the Invention [Problem to be solved by the invention]

[0006] Although the method of Non-Patent Document 1 is capable of predicting resilience trends, the accuracy of resilience estimation is still insufficient.

[0007] One aspect of the present invention aims to improve the accuracy of resilience estimation. [Means for solving the problem]

[0008] In order to solve the above problem, a psychological resilience estimation method according to one embodiment of the present invention includes a display step of displaying a first image and a second image using facial expression images as display images, or displaying a third image and a fourth image using emotional images as display images; a measurement step of measuring the gaze movement of a user viewing the display images; an index acquisition step of obtaining a gaze index value from the measured gaze movement; and an estimation step of evaluating the gaze index value for the facial expression images based on a first evaluation criterion and evaluating the gaze index value for the emotional images based on a second evaluation criterion, thereby estimating the psychological resilience of the user.

[0009] In addition, a psychological resilience estimation method according to another aspect of the present invention includes a display step of displaying a first image and a second image using facial expression images as display images, a measurement step of measuring the gaze movement of a user looking at the display images, an index acquisition step of obtaining a gaze index value from the measured gaze movement, and an estimation step of estimating the psychological resilience of the user based on a first evaluation criterion using the gaze index for the facial expression images.

[0010] In the psychological resilience estimation method according to the above aspect, the first evaluation criterion may use, as the gaze index, a saccade latency, which is the time it takes for the subject to first gaze toward an image with a higher emotional valence. The first evaluation criterion may further use, as the gaze index, a percentage of the total time spent gazing toward an image with a higher emotional valence, a frequency of gazing toward an image with a higher emotional valence, and a duration of initial gaze toward an image with a higher emotional valence. The first evaluation criterion may use a weighted sum of a plurality of the gaze indices for evaluation.

[0011] Furthermore, a psychological resilience estimation method according to yet another aspect of the present invention includes a display step of displaying a third image and a fourth image using an emotional image as the display image, a measurement step of measuring the gaze movement of a user looking at the display image, an index acquisition step of obtaining a gaze index value from the measured gaze movement, and an estimation step of estimating the psychological resilience of the user based on a second evaluation criterion using the gaze index for the emotional image.

[0012] In the psychological resilience estimation method according to the above aspect, the second evaluation criterion may use the frequency of directing gaze to an image with a higher emotional valence as the gaze index. The second evaluation criterion may further use the proportion of initial gaze directed to an image with a higher emotional valence and the duration of initial gaze directed to an image with a higher emotional valence as the gaze index. The second evaluation criterion may also use a weighted sum of the gaze indexes for evaluation.

[0013] The psychological resilience estimation method according to the above aspect may include, before the display step, a fixation step of displaying a fixation point image for fixing a gaze point.

[0014] In the psychological resilience estimation method according to the above aspect, the index acquisition step may obtain the value of the gaze index for the facial expression image from the gaze movement of the user when looking at the first image and the second image, which have different emotional valences, and the gaze movement of the user when looking at the first image and the second image, which both have neutral emotional valences; or may obtain the value of the gaze index for the facial expression image from the gaze movement of the user when looking at the third image and the fourth image, which have different emotional valences, and the gaze movement of the user when looking at the third image and the fourth image, which both have neutral emotional valences.

[0015] The psychological resilience estimation method according to the above aspect may further include an identification step of identifying the target of the user's anxiety by comparing the evaluation results regarding the facial expression image and the evaluation results regarding the emotional image in the estimation step.

[0016] The psychological resilience estimation method according to the above aspect may further include an identification step of identifying the target of the user's anxiety by comparing the evaluation result regarding the facial expression image estimated in the estimation step when one of the first image and the second image includes a target of anxiety with the evaluation result regarding the facial expression image estimated in the estimation step when neither the first image nor the second image includes the target of anxiety, or by comparing the evaluation result regarding the emotional image estimated in the estimation step when one of the third image and the fourth image includes a target of anxiety with the evaluation result regarding the emotional image estimated in the estimation step when neither the third image nor the fourth image includes the target of anxiety.

[0017] Furthermore, a psychological resilience estimation device according to yet another aspect of the present invention includes a display control unit that displays a first image and a second image using facial expression images as display images, or displays a third image and a fourth image using emotional images as display images; a measurement unit that obtains measurement results of the gaze movement of a user looking at the display images; an index acquisition unit that obtains a gaze index value from the measured gaze movement; and an estimation unit that estimates the psychological resilience of the user by evaluating the gaze index value for the facial expression images based on a first evaluation criterion and evaluating the gaze index value for the emotional images based on a second evaluation criterion.

[0018] Furthermore, a psychological resilience estimation device according to yet another aspect of the present invention includes a display control unit that displays a first image and a second image using facial expression images as display images, a measurement unit that obtains measurement results of the gaze movement of a user looking at the display images, an index acquisition unit that obtains a gaze index value from the measured gaze movement, and an estimation unit that estimates the psychological resilience of the user based on a first evaluation criterion using the gaze index for the facial expression images.

[0019] Furthermore, a psychological resilience estimation device according to yet another aspect of the present invention includes a display control unit that displays a third image and a fourth image using emotional images as display images, a measurement unit that obtains measurement results of the gaze movement of a user looking at the display images, an index acquisition unit that obtains a gaze index value from the measured gaze movement, and an estimation unit that estimates the psychological resilience of the user based on a second evaluation criterion using the gaze index for the emotional image.

[0020] The psychological resilience estimation method and psychological resilience estimation device according to each aspect of the present invention may be realized by a computer. In this case, the control program for the psychological resilience estimation method and psychological resilience estimation device, which causes the computer to operate as each step and each part (software element) that the psychological resilience estimation method and psychological resilience estimation device respectively comprise, and the computer-readable recording medium on which it is recorded, also fall within the scope of the present invention. [Effects of the Invention]

[0021] According to one aspect of the present invention, it is possible to improve the accuracy of resilience estimation. [Brief explanation of the drawings]

[0022] [Figure 1] 1 is a block diagram showing a schematic configuration of a resilience estimation system according to an embodiment of the present invention. [Figure 2] FIG. 2 is a block diagram showing a schematic configuration of a user terminal in the resilience estimation system. [Figure 3] 10A and 10B are diagrams illustrating an example of a display operation on the user terminal using a facial expression image. [Figure 4] 10A and 10B are diagrams illustrating an example of the display operation using an emotional image. [Figure 5] 10A and 10B are diagrams showing another display example of paired images displayed by the display operation. [Figure 6] FIG. 2 is a block diagram showing a schematic configuration of an estimation server in the resilience estimation system. [Figure 7] 10 is a flowchart showing the flow of resilience estimation processing in the resilience estimation system having the above configuration. [Figure 8] 10 is a flowchart showing the flow of a resilience estimation process in a modified example of the resilience estimation system. [Figure 9]10 is a graph showing the correlation between resilience scores obtained from a self-administered questionnaire and estimated resilience scores calculated using visual field indices and regression equations related to facial expression images. [Figure 10] 1 is a graph showing the correlation between resilience scores from a self-administered questionnaire and estimated resilience scores calculated using visual field indices and regression equations related to emotional images. [Figure 11] 10 is a graph showing the relationship between first saccade latency and resilience score when facial expression images are used. [Figure 12] 10 is a graph showing the relationship between first saccade latency and resilience score when using emotional images. [Figure 13] 10 is a flowchart showing the flow of a resilience estimation process in a resilience estimation system according to another embodiment of the present invention. [Figure 14] FIG. 10 is a block diagram showing a schematic configuration of a user terminal in a resilience estimation system according to yet another embodiment of the present invention. [Figure 15] 10 is a flowchart showing the flow of an anxiety identification process in the resilience estimation system. DETAILED DESCRIPTION OF THE INVENTION

[0023] Hereinafter, embodiments of the present invention will be described in detail. For the sake of convenience, the same reference numerals will be used to designate components having the same functions as those in the embodiments, and the description thereof will be omitted where appropriate.

[0024] [Embodiment 1] An embodiment of the present invention will be described with reference to FIGS.

[0025] Fig. 1 is a block diagram showing a schematic configuration of a resilience estimation system according to this embodiment. As shown in Fig. 1, the resilience estimation system 1 includes an image server 2, a user terminal 3 (psychological resilience estimation device), and an estimation server 4 (psychological resilience estimation device). The user terminal 3 is communicably connected to the image server 2 and the estimation server 4 via a communication network 5 such as the Internet. Note that, hereinafter, the description of the communication network 5 for communication between the user terminal and the image server 2 and the estimation server 4 may be omitted.

[0026] (Image server) The image server 2 stores image data, where the images are classified into at least facial expression images and emotional images.

[0027] The facial expression images are images of people with various facial expressions such as smiling faces, angry faces, etc. In this embodiment, images of happy faces, neutral faces, sad faces, angry faces, etc. are used as the facial expression images.

[0028] The facial expression images are classified into one of positive valence, neutral valence, and negative valence. The valence indicates the degree of emotion. The valences are positive, neutral, and negative, in descending order. For example, the happy facial expression image has a positive valence, the neutral facial expression image has a neutral valence, and the sad facial expression image and the angry facial expression image have a negative valence.

[0029] Emotional images are images of various scenes that evoke emotions such as pleasure and discomfort. In this embodiment, images of positive scenes (Positive), neutral scenes (Neutral), negative scenes (Negative), etc. are used as the emotional images. Note that in this application, emotional images do not include the above-mentioned facial expression images.

[0030] Similar to the facial expression images, the affective images are classified as having a positive valence, a neutral valence, or a negative valence, for example, the positive scene images have a positive valence, the neutral scene images have a neutral valence, and the negative scene images have a negative valence.

[0031] In response to a request from a user terminal 3, the image server 2 transmits data of a plurality of facial expression images and data of a plurality of emotional images to the user terminal 3 via a communication network. The data of the facial expression images includes attribute information, which includes the type of facial expression and valence. The data of the emotional images includes attribute information, which includes the type of scene and valence.

[0032] (user terminal) Fig. 2 is a block diagram showing a schematic configuration of the user terminal 3. Examples of the user terminal 3 include a PC (Personal Computer), a smartphone, and a tablet terminal. As shown in Fig. 2, the user terminal 3 includes a control unit 11, a communication unit 12, a display unit 13, and an imaging unit 14.

[0033] The control unit 11 comprehensively controls the operation of various components of the user terminal 3, and is configured by, for example, a computer including a CPU (Central Processing Unit) and memory. The operation of the various components is controlled by causing the computer to execute a control program. The control unit 11 will be described in detail later.

[0034] The communication unit 12 transmits and receives information to and from the image server 2 and the estimation server 4 via the communication network 5 shown in Fig. 1. The communication unit 12 includes a communication device such as a transmission / reception circuit. The display unit 13 displays information from the control unit 11 on a screen, and includes a display device such as an LCD (Liquid Crystal Display) or an LED (Light Emitting Diode). The image displayed on the display unit 13 is referred to as a display image.

[0035] The image capturing unit 14 captures an image of a subject and generates video data, and is configured to include an imaging element such as a CMOS (Complementary Metal-Oxide Semiconductor) image sensor. In this embodiment, the subject includes at least the user's eyes. The image capturing unit 14 transmits the generated video data to the control unit 11.

[0036] Next, a description will be given of the details of the control unit 11. As shown in Fig. 2, the control unit 11 includes an image acquisition unit 21, a pairing unit 22, a display control unit 23, a line-of-sight measurement unit 24 (measurement unit), an index calculation unit 25, and a resilience acquisition unit 26.

[0037] The image acquisition unit 21 acquires data of a plurality of facial expression images and data of a plurality of emotional images from the image server 2 via the communication network 5 and the communication unit 12. The image acquisition unit 21 sends the acquired data of a plurality of facial expression images and data of a plurality of emotional images to the pairing unit 22.

[0038] The pairing unit 22 performs image pairing using data of multiple facial expression images and / or data of multiple emotional images from the image acquisition unit 21. The two images that make up a pair (hereinafter abbreviated as "paired images") are either both facial expression images or both emotional images. In other words, pairing of facial expression images and emotional images is not performed.

[0039] Furthermore, the two images have different emotional valences. Therefore, the paired images are two images with positive and neutral emotional valences, two images with positive and negative emotional valences, or two images with neutral and negative emotional valences. The pairing unit 22 sequentially sends data of the paired images to the display control unit 23.

[0040] The display control unit 23 controls the display of information on the display unit 13. Specifically, the display control unit 23 controls the display unit 13 to perform the following display operations.

[0041] That is, the display control unit 23 first displays a fixation point image for fixing the viewpoint in the center of the screen for a predetermined period (for example, 1.5 seconds) (fixation step). After that, the display control unit 23 erases the fixation point image from the screen. The fixation point image is, for example, a "+" symbol.

[0042] Next, the display control unit 23 displays one of the paired images from the pairing unit 22 on the left side of the screen and the other on the right side of the screen, side by side, for a predetermined period of time (e.g., 3 seconds). At this time, the display control unit 23 sends the valence of the image displayed on the left side (first image, third image) and the valence of the image displayed on the right side (second image, fourth image) to the index calculation unit 25. Thereafter, the paired images are erased from the screen. The above operation is repeated for all paired images from the pairing unit 22.

[0043] A specific display operation will be described with reference to FIGS.

[0044] FIG. 3 is a diagram showing an example of the display operation using facial expression images. First, as shown on the left side of FIG. 3, a fixation point image FI is displayed in the center of the screen DS for a predetermined period of time. Thereafter, the fixation point image FI is erased from the screen DS. Next, as shown second from the left in FIG. 3, a pair of images PX1 consisting of a happy expression image HX1 (first image) and a neutral expression image NX1 (second image) are displayed side by side for a predetermined period of time, with the happy expression image HX1 on the left side of the screen DS and the neutral expression image NX1 on the right side of the screen DS. Thereafter, the pair of images PX1 is erased from the screen DS.

[0045] Next, as shown in the third image from the left in FIG. 3, the fixation point image FI is again displayed in the center of the screen DS for a predetermined period of time. Thereafter, the fixation point image FI is erased from the screen DS. Next, as shown in the rightmost (fourth image from the left) image in FIG. 3, a pair of images PX2 consisting of a neutral expression image NX2 (first image) and an angry expression image AX1 (second image) is displayed side by side for a predetermined period of time, with the neutral expression image NX2 on the left side of the screen DS and the angry expression image AX1 on the right side of the screen DS. Here, the neutral expression images NX1 and NX2 may be the same or different. Thereafter, the pair of images PX2 is erased from the screen DS.

[0046] FIG. 4 is a diagram showing an example of the above-mentioned display operation using emotional images. First, as shown on the left side of FIG. 4, a fixation point image FI is displayed in the center of the screen DS for a predetermined period of time. Thereafter, the fixation point image FI is erased from the screen DS. Next, as shown second from the left in FIG. 4, a pair of images PS1 consisting of a positive scene image SS1 (third image) and a neutral scene image NS1 (fourth image) are displayed side by side for a predetermined period of time, with the positive scene image SS1 on the left side of the screen DS and the neutral scene image NS1 on the right side of the screen DS. Thereafter, the pair of images PS1 is erased from the screen DS.

[0047] Next, as shown in the third image from the left in FIG. 4, the fixation point image FI is again displayed in the center of the screen DS for a predetermined period of time. Thereafter, the fixation point image FI is erased from the screen DS. Next, as shown in the rightmost (fourth image from the left) image of FIG. 4, a pair of images PS2 consisting of a neutral scene image NS2 (third image) and a negative scene image GS1 (fourth image) is displayed side by side for a predetermined period of time, with the neutral scene image NS2 on the left side of the screen DS and the negative scene image GS1 on the right side of the screen DS. Here, the neutral scene images NS1 and NS2 may be the same or different. Thereafter, the pair of images PS2 is erased from the screen DS.

[0048] Fig. 5 is a diagram showing another display example of paired images displayed in the above-mentioned display operation. In Fig. 3 and Fig. 4, the paired images are displayed side by side in the horizontal direction, but the paired images may also be displayed side by side in the vertical direction, or side by side in the diagonal direction. Also, as shown in Fig. 5, one image PI1 (first image, third image) of the paired images PI may be displayed alone in the center of the screen DS, and the other image PI2 (second image, fourth image) of the paired images PI may be displayed multiple times in the peripheral parts of the screen DS.

[0049] Furthermore, instead of displaying an image of a facial expression, characters indicating the facial expression may be displayed, such as happy, neutral, sad, angry, etc.

[0050] The gaze measurement unit 24 measures the user's gaze movement for each pair of images by analyzing the user's eye movement using the video data from the image capture unit 14. The gaze measurement unit 24 sends the user's gaze movement, which is the measurement result, to the index calculation unit 25. Note that the gaze measurement unit 24 may further use an infrared sensor, a proximity sensor, or the like provided in the user terminal 3. In this case, the user's gaze movement can be measured with high accuracy.

[0051] The index calculation unit 25 calculates a gaze index for evaluating attention bias from the user's gaze movement detected by the gaze measurement unit 24. In this embodiment, the index calculation unit 25 calculates a gaze index based on the display of a pair of facial expression images (first and second images) from among all paired images, and also calculates a gaze index based on the display of a pair of emotional images (third and fourth images). The index calculation unit 25 transmits the gaze index based on the facial expression images and the gaze index based on the emotional images to the estimation server 4 via the communication unit 12 and the communication network 5.

[0052] The gaze indices include, for example, (a) first saccade direction, (b) first saccade latency, (c) first saccade dwell time, (d) frequency, and (e) total time, each of which is described in detail below. Here, saccade refers to a rapid eye movement that occurs when the gaze is shifted quickly. The index calculation unit 25 uses the valence of the paired images from the display control unit 23 to designate the image with a higher valence among the paired images displayed on the display unit 13 as the positive image P, while the image with a lower valence is designated as the control image C. For example, if the paired images consist of a neutral expression image (neutral valence image) and a sad expression image (negative valence image), the neutral expression image is the positive image P, and the sad expression image is the control image C.

[0053] (a) First Fixation The first saccade direction is the ratio of the number of times the user first directs their attention to the left or right image when the pair of images are displayed on the left and right sides of the screen. The first saccade direction is calculated using the following formula (1):

[0054] (First Fixation)=(First P) / ((First P)+(First C)) ···(1).

[0055] Here, First P is the number of times that the user first directs attention to the positive image P when all the paired images are displayed. Furthermore, First C is the number of times that the user first directs attention to the control image C when all the paired images are displayed. If the value of the first saccade direction is greater than 0.5, it can be understood that the user is likely to direct attention to the positive image P among the paired images displayed. Note that the first saccade direction can also be said to be the proportion of times the gaze is first directed to the positive image P, which has a higher emotional valence.

[0056] (b) First saccade latencies The first saccade latency is the ratio of the saccade latency when the first saccade is made after the paired images are displayed. The first saccade latency is calculated using the following formula (2):

[0057] (Latencies)=(Latencies P) / ((Latencies P)+(Latencies C)) ···(2).

[0058] Here, Latencies P[s] is the average time taken for the first saccade to move towards the positive image P when all of the paired images are displayed. Also, Latencies C[s] is the average time taken for the first saccade to move towards the control image C when all of the paired images are displayed. If the value of the first saccade latency is smaller than 0.5, it can be understood that the latency to the positive image P is short, and the user has drawn attention to the positive image P more quickly. Note that the first saccade latency can also be said to be the saccade latency until the gaze is first directed towards the positive image P, which has a higher emotional valence.

[0059] (c) First saccade dwell time The first saccade dwell time is the percentage of time the user's gaze stayed on the image to which the user focused their attention when the pair of images was displayed and the first saccade was made. The first saccade dwell time is calculated using the following equation (3):

[0060] (Dwell Time)=(Dwell Time P) / ((Dwell Time P)+(Dwell Time C)) ···(3).

[0061] Here, Dwell Time P [s] is the average time that the user directs their attention to the positive image P in the first saccade and their gaze stays on the positive image P. C[s] is the average time that the user directs their attention to the control image C in the first saccade and their gaze remains on the control image C. If the value of the first saccade dwell time is 0.5 or more, the user can understand that their gaze remained longer on the positive image P in the first saccade. The first saccade dwell time can also be said to be the time that the user first maintained their gaze on the positive image P, which has a higher emotional valence.

[0062] (d) Frequency The frequency is the ratio of the number of times attention was directed to the positive image P when all of the paired images were displayed. The frequency is calculated using the following formula (4).

[0063] (Frequency)=(Frequency P) / ((Frequency P)+(Frequency C)) ···(4).

[0064] Here, Frequency P is the total number of times that attention was paid to the positive image P while the paired images were displayed, for all of the paired images. Frequency C is the total number of times that attention was paid to the control image C while the paired images were displayed, for all of the paired images. If the frequency value is 0.5 or more, it can be understood that the user paid attention to the positive image P frequently. Note that the frequency can also be said to be the frequency at which the gaze is directed to the positive image P, which has a higher emotional valence.

[0065] (e) Total Viewing Time The total time is the proportion of time that attention was directed to the positive image P when all of the paired images were displayed. The total time is calculated using the following equation (5).

[0066] (Viewing Time)=(Viewing Time P) / ((Viewing Time P)+(Viewing Time C)) ···(5).

[0067] Here, Viewing Time P [s] is the total value of the time spent looking at the positive image P while the paired images are displayed, for all of the paired images. Also, Viewing Time C [s] is the total value of the time spent looking at the control image P while the paired images are displayed, for all of the paired images. If the total time is 0.5 or more, it can be understood that the user spent a long time looking at the positive image P. Note that the total time can also be considered as the proportion of the total time spent looking at the positive image P, which has a higher emotional valence.

[0068] As will be described later, the resilience acquisition unit 26 acquires the resilience (psychological resilience) estimated by the estimation server 4 via the communication network 5 and the communication unit 12. The resilience acquisition unit 26 displays and outputs the acquired resilience via the display control unit 23 and the display unit 13.

[0069] (Estimated server) 6 is a block diagram showing a schematic configuration of the estimation server 4. The estimation server 4 estimates resilience using five gaze indices related to facial expression images and five gaze indices related to emotional images calculated by the user terminal 3.

[0070] As shown in Fig. 6, the estimation server 4 includes a control unit 31, a communication unit 32, and a storage unit 33. The control unit 31 and the communication unit 32 shown in Fig. 6 have the same hardware configuration as the control unit 11 and the communication unit 12 shown in Fig. 2, and therefore a description thereof will be omitted. The storage unit 33 records information and includes a storage device such as a hard disk or flash memory.

[0071] 6, the control unit 31 includes an index acquisition unit 41, a first evaluation unit 42 (estimation unit), a second evaluation unit 43 (estimation unit), and a resilience estimation unit 44 (estimation unit). The storage unit 33 stores a first evaluation criterion 51 and a second evaluation criterion 52.

[0072] The first evaluation criterion 51 is the five gaze indices x (a) to (e) described above regarding facial expression images. 11 ~x 15 The parameter of the first evaluation model, which is trained in advance, is set as an explanatory variable and the resilience score y1 obtained from a self-administered questionnaire is set as a target variable. For example, when the first evaluation model is a linear regression model, the first evaluation criterion 51 is set as a partial regression variable b 1i and the regression constant a1. y1=b 11 x 11 +b 12 x 12 +b 13 x 13 +b14 x 14 +b 15 x 15 +a1···(6).

[0073] The second evaluation criterion 52 is the five gaze indices x (a) to (e) described above regarding emotional images. 21 ~x 25 The parameter of the second evaluation model, which is trained in advance, is set as an explanatory variable and the resilience score y2 obtained from the self-administered questionnaire is set as the objective variable. For example, when the second evaluation model is a linear regression model, the second evaluation criterion 52 is set as a partial regression variable b 2i and the regression constant a2. y2=b 21 x 21 +b 22 x 22 +b 23 x 23 +b 24 x 24 +b 25 x 25 +a2···(7).

[0074] The index acquisition unit 41 acquires five gaze indices related to facial expression images and five gaze indices related to emotional images from the user terminal 3 via the communication network 5 and the communication unit 32. The index acquisition unit 41 sends the five gaze indices related to the acquired facial expression images to the first evaluation unit 42. The index acquisition unit 41 also sends the five gaze indices related to the acquired emotional images to the second evaluation unit 43.

[0075] The first evaluation unit 42 evaluates the above five gaze indices x 11 ~x 15 and the first evaluation criterion 51 stored in the storage unit 33, the first evaluation unit 42 evaluates the first resilience score y1. The first evaluation unit 42 sends the evaluated first resilience y1 to the resilience estimation unit 44.

[0076] The second evaluation unit 43 evaluates the above five gaze indices x relating to the emotional image from the index acquisition unit 41. 21 ~x 25and the second evaluation criterion 52 stored in the storage unit 33, the second evaluation unit 43 evaluates the second resilience score y2. The second evaluation unit 43 sends the evaluated second resilience y2 to the resilience estimation unit 44.

[0077] The resilience estimation unit 44 calculates an estimated value of the resilience score using the first resilience score y1 from the first evaluation unit 42 and the second resilience score y2 from the second evaluation unit 43. The estimated value may be a statistical value (average value, weighted average value, maximum value, minimum value, etc.) of the first resilience score y1 and the second resilience score y2. The resilience estimation unit 44 transmits the calculated estimated value of the resilience score to the user terminal 3 via the communication unit 32 and the communication network 5. The resilience estimation unit 44 may transmit the estimated value to another communication device. The resilience estimation unit 44 may also transmit the first resilience score y1 and the second resilience score y2 to the user terminal 3 or to another communication device.

[0078] (Resilience estimation processing) Fig. 7 is a flowchart showing the flow of the resilience estimation process (psychological resilience estimation method) in the resilience estimation system 1 configured as described above. As shown in Fig. 7, first, in the user terminal 3, the image acquisition unit 21 acquires data of a plurality of facial expression images and data of a plurality of emotional images from the image server 2 (S11). Next, the pairing unit 22 uses the data of the plurality of facial expression images and the data of the plurality of emotional images to create a plurality of paired images related to the facial expression images and a plurality of paired images related to the emotional images (S12).

[0079] Next, the display control unit 23 selects any of the plurality of paired images related to facial expression images and the plurality of paired images related to emotional images, and displays the selected paired images side by side on the display unit 13 (S13, display step). At this time, the gaze measurement unit 24 measures the movement of the user's gaze using video data from the imaging unit 14 (S14, measurement step). Then, steps S13 and S14 are repeated until all of the plurality of paired images related to facial expression images and the plurality of paired images related to emotional images have been selected (S15).

[0080] Next, the following operation is performed on the facial expression image. That is, in the user terminal 3, the index calculation unit 25 calculates five gaze indices related to the facial expression image from the gaze movements of the user measured for all pair images related to the facial expression image (S16, index acquisition step). Next, in the estimation server 4, the first evaluation unit 42 calculates the five gaze indices x related to the facial expression image. 11 ~x 15 and the first evaluation criterion 51 stored in the storage unit 33, the first resilience score y1 is evaluated (S17, estimation step).

[0081] On the other hand, the following operation is performed for the emotional image. That is, in the user terminal 3, the index calculation unit 25 calculates five gaze indices for the emotional image from the gaze movements of the user measured for all paired images related to the emotional image (S18, index acquisition step). Next, the second evaluation unit 43 calculates the five gaze indices x for the emotional image. 21 ~x 25 and the second evaluation criterion 52 stored in the storage unit 33, the second resilience score y2 is evaluated (S19, estimation step). Note that steps S16 and S17 and steps S18 and S19 may be performed simultaneously, or one of them may be performed first.

[0082] Then, the resilience estimation unit 44 calculates an estimated value of the resilience score using the first resilience score y1 evaluated by the first evaluation unit 42 and the second resilience score y2 evaluated by the second evaluation unit 43 (S20, estimation step), and then ends the resilience estimation process.

[0083] (Variation 1) In this embodiment, the resilience estimation system 1 uses both facial expression images and emotional images. However, it is also possible to use only facial expression images without using emotional images. In this case, the estimation server 4 can omit the second evaluation unit 43 and the second evaluation criterion 52. That is, the five gaze indices related to emotional images are not used in evaluating resilience. Furthermore, the resilience estimation unit 44 can simply use the first resilience score y1 evaluated by the first evaluation unit 42 as the estimated value of the resilience score. In this case, in FIG. 7, steps S18 and S19 can be omitted, and only the first resilience score y1 can be used in step S20. Furthermore, in FIG. 7, all parts related to emotional images can be omitted, and resilience can be estimated using only facial expression images.

[0084] (Variation 2) Furthermore, the resilience estimation system 1 may use only emotional images without using facial expression images. In this case, the estimation server 4 may omit the first evaluation unit 42 and the first evaluation criterion 51. That is, the five gaze indices related to facial expression images are not used in the evaluation of resilience. Furthermore, the resilience estimation unit 44 may use the second resilience score y2 evaluated by the second evaluation unit 43 as the estimated value of the resilience score. In this case, in FIG. 7, steps S16 and S17 may be omitted, and only the second resilience score y2 may be used in step S20. Furthermore, in FIG. 7, all parts related to facial expression images may be omitted, and resilience may be estimated using only emotional images.

[0085] (Variation 3) Furthermore, the resilience estimation system 1 may allow the user to select an image to be used for evaluating resilience from among facial expression images and emotional images.

[0086] 8 is a flowchart showing the flow of the resilience estimation process (estimation method) in this modified example. As shown in FIG. 8, first, steps S11 to S15 in FIG. 7 are executed.

[0087] Next, in step S15, if all of the plurality of paired images are selected (YES in step S15), the index calculation unit 25 determines which of the facial expression images and emotional images to use for the resilience evaluation based on the user's selection (S31). If both the facial expression images and emotional images are to be used for the evaluation in step S31, steps S16 to S20 in Fig. 7 are executed, and then the resilience estimation process is terminated.

[0088] On the other hand, if only the facial expression image is used for the evaluation in step S31, steps S16 and S17 in Fig. 7 are executed. Next, in the estimation server 4, the resilience estimation unit 44 sets the first resilience score y1 evaluated by the first evaluation unit 42 as an estimated value of the resilience score (S32). Thereafter, the resilience estimation process is terminated.

[0089] On the other hand, if only the emotional image is used for the evaluation in step S31, steps S18 and S19 in Fig. 7 are executed. Next, in the estimation server 4, the resilience estimation unit 44 sets the second resilience score y2 evaluated by the second evaluation unit 43 as the estimated value of the resilience score (S33). Thereafter, the resilience estimation process is terminated.

[0090] (Example of evaluation criteria) A detailed description will be given of specific examples of the first evaluation criterion 51 and the second evaluation criterion 52 of the estimation server 4 in the resilience estimation system 1 configured as above.

[0091] First, in the user terminal 3, the display control unit 23 displays paired images related to facial expression images via the display unit 13, and the gaze measurement unit 24 measures the movement of the user's gaze. Then, this process is repeated for all paired images related to facial expression images. Next, the index calculation unit 25 calculates five gaze indices x related to facial expression images from the movement of the user's gaze measured for all paired images related to facial expression images. 11 ~x 15 was calculated and used as an explanatory variable for the facial expression image.

[0092] Next, in the user terminal 3, the display control unit 23 displays paired images related to emotional images via the display unit 13, and the gaze measurement unit 24 measures the movement of the user's gaze. Then, this process is repeated for all paired images related to emotional images. Next, the index calculation unit 25 calculates five gaze indices x related to emotional images from the movement of the user's gaze measured for all paired images related to emotional images. 21 ~x 25 was calculated and used as an explanatory variable for emotional images.

[0093] Meanwhile, the user filled out a self-administered questionnaire to calculate a resilience score, which was used as the objective variables y1 and y2 for the facial expression images and emotional images. The self-administered questionnaire utilized the Adolescent Resilience Scale (ARS). The ARS is a scale that measures mental resilience, a psychological characteristic that promotes recovery from mental depression.

[0094] By performing the above process for 21 users, the explanatory variable x 11 ~x 15 and 21 sets including the objective variable y1 and the explanatory variable x 21 ~x 25 and 21 sets including the response variable y2 were obtained.

[0095] Explanatory variable x for facial expression images 11 ~x 15When multiple regression analysis was performed using the above set of response variables y1 and y2, the following regression equation (8) corresponding to the above equation (6) was obtained. y1=-107x 12 +14x 13 -27x 14 -31x 15 +151 ···(8). That is, the partial regression variable b 1i But, b 11 =0, b 12 =-107, b 13 =+14, b 14 = -27, and b 15 =-31, and the regression constant a1 was 151.

[0096] Therefore, when the above formula (8) is used, the first evaluation unit 42 calculates the line-of-sight index x 12 The first saccade latency corresponding to the gaze index x 13 The first saccade dwell time, corresponding to the gaze index x 14 The frequency corresponding to and gaze index x 15 The first evaluation unit 42 may use the total time corresponding to the line-of-sight index x 12 , x 13 , x 14 , x 15 The weighted sum of these can be used for evaluation.

[0097] Furthermore, referring to the above equation (8), the gaze index x 12 Partial regression coefficient b 12 The magnitude of the other partial regression coefficients b 13 , b 14 , and b 15 Therefore, the first evaluation unit 42 determines the magnitude of the line-of-sight index x 12 Only the first saccade latency corresponding to may be used as the first evaluation criterion.

[0098] Figure 9 shows the relationship between the resilience score from the self-administered questionnaire, which is the objective variable y1, and the explanatory variable x 11 ~x 159 is a graph showing the correlation between the estimated value of the resilience score calculated using the above formula (8) and the correlation coefficient r. In FIG. 9, the correlation coefficient r is 0.80, and the adjusted coefficient of determination R *2 The coefficient of determination R of the regression equation used in fields such as medical statistics was 0.55. *2 is considered to be accurate if it is 0.5 or more. Therefore, by using the regression equation of the above equation (8), it is possible to accurately estimate the resilience score from the facial expression image.

[0099] On the other hand, when multiple regression analysis was performed using the above set of emotional images, the following regression equation (9) corresponding to the above equation (7) was obtained. y2=-29x 21 +12x 23 -95x 24 +129 ···(9). That is, the partial regression variable b 2i But, b 21 =-29, b 22 =0, b 23 =12, b 24 =-95, and b 25 =0, and the regression constant a2 was 129.

[0100] Therefore, when the above formula (9) is used, the second evaluation unit 43 calculates the line-of-sight index x 21 The initial saccade direction, corresponding to the gaze index x 23 the first saccade dwell time corresponding to x, and the gaze index x 24 The second evaluation unit 43 may use the frequency corresponding to the line-of-sight index x 21 , x 23 , x 24 The weighted sum of these can be used for evaluation.

[0101] Furthermore, referring to the above equation (9), the gaze index x 24 Partial regression coefficient b 24 The magnitude of the other partial regression coefficients b 21 , b 23 , and b 24 Therefore, the second evaluation unit 43 determines the magnitude of the line-of-sight index x24 Alternatively, only the frequency corresponding to the above may be used as the second evaluation criterion.

[0102] Figure 10 shows the relationship between the resilience score from the self-administered questionnaire, which is the objective variable y2, and the explanatory variable x 21 ~x 25 10 is a graph showing the correlation between the estimated resilience score calculated using the above formula (9) and the correlation coefficient r. In FIG. 10, the correlation coefficient r is 0.86, and the adjusted coefficient of determination R *2 was 0.70. Therefore, by using the regression equation (9) above, it is possible to accurately estimate resilience scores from emotional images.

[0103] It can also be seen that the above formulas (8) and (9) are completely different. Therefore, it can be seen that the resilience score can be accurately estimated by classifying images into facial expression images and emotional images, and using the above formula (8) for the facial expression images and the above formula (9) for the emotional images.

[0104] FIG. 11 is a graph showing the relationship between the first saccade latency and the resilience score when facial expression images are used. FIG. 12 is a graph showing the relationship between the first saccade latency and the resilience score when emotional images are used. Referring to FIG. 11, it can be seen that there is a negative correlation between the first saccade latency and the resilience score when facial expression images are used. On the other hand, referring to FIG. 12, it can be seen that there is not much correlation between the first saccade latency and the resilience score when emotional images are used.

[0105] As mentioned above, when facial expression images are used, the gaze index x 12 The first saccade latency corresponding to the first resilience score y1 significantly affected the first resilience score, while the gaze index x 22This is because the initial saccade latency corresponding to the first saccade has almost no effect on the second resilience score y2. Also, referring to Figure 11, when saccade latency is measured using facial expression images, it can be seen that the lower a person's resilience is, the larger the saccade latency value is, and in other words, it can be seen that people who are more susceptible to stress spend a longer time paying attention to positive images.

[0106] Therefore, unlike conventional self-administered questionnaires, which are heavily influenced by subjectivity, it is possible to evaluate resilience without subjectivity.

[0107] [Embodiment 2] Another embodiment of the present invention will be described with reference to Fig. 2 and Fig. 13. The resilience estimation system 1 according to this embodiment is similar to the resilience estimation system 1 shown in Fig. 1 except that a user terminal 3a is provided instead of the user terminal 3.

[0108] Incidentally, the gaze movement of a user may contain the user's habits. Therefore, in this embodiment, the user terminal 3a further displays a pair of images both of which have neutral valence, and calculates a corrected gaze index based on the gaze movement of the user caused by the display of the pair of images. This makes it possible to exclude the user's habits from each gaze index, thereby improving the accuracy of the gaze index and, as a result, improving the accuracy of resilience estimation.

[0109] For example, the user terminal 3a may take the difference between the user's gaze movement when paired images with different valences are displayed and the user's gaze movement when paired images with both neutral valences are displayed, and use this difference to calculate the corrected gaze index. Alternatively, the user terminal 3a may calculate a correction value for the gaze index from the user's gaze movement when paired images with both neutral valences are displayed, and correct the gaze index calculated from the user's gaze movement when paired images with different valences are displayed with this correction value.

[0110] As shown in FIG. 2, the user terminal 3a of this embodiment differs from the above-described user terminal 3 in that a pairing unit 22a is provided instead of the pairing unit 22, and an index calculation unit 25a is provided instead of the index calculation unit 25, but the other configurations are the same.

[0111] 2 in that pairing unit 22a further pairs two images both of which have neutral emotional valence, but is otherwise similar. As a result, a pair of images both of which have neutral facial expressions is further displayed, and a pair of images both of which are images of neutral scenes is further displayed.

[0112] The index calculation unit 25a calculates five gaze indices from the user's gaze movements measured for all paired images with different valences for the facial expression images and the user's gaze movements measured for all paired images with both neutral valences for the emotional images. The index calculation unit 25a also calculates five gaze indices from the user's gaze movements measured for all paired images with different valences for the emotional images and the user's gaze movements measured for all paired images with both neutral valences for the emotional images. The index calculation unit 25a transmits the gaze indices for the facial expression images and the gaze indices for the emotional images to the estimation server 4 via the communication unit 12 and the communication network 5.

[0113] Fig. 13 is a flowchart showing the flow of the resilience estimation process (psychological resilience estimation method) in the resilience estimation system 1 of this embodiment. The resilience estimation process shown in Fig. 13 differs from the resilience estimation process shown in Fig. 7 in that steps S41 and S42 are provided instead of steps S16 and S18, but the other processes are similar.

[0114] In step S41, the index calculation unit 25a calculates five gaze indices from the user's gaze movements measured for all paired images with different valences and all paired images with neutral valences for the facial expression images. In step S42, the index calculation unit 25a calculates five gaze indices from the user's gaze movements measured for all paired images with different valences and all paired images with neutral valences for the emotional images.

[0115] [Embodiment 3] Another embodiment of the present invention will be described with reference to Figures 14 and 15. The resilience estimation system 1 according to this embodiment is similar to the resilience estimation system 1 shown in Figure 1 except that a user terminal 3b is provided instead of the user terminal 3.

[0116] It is known that the target of anxiety and the target of attention vary depending on the type of anxiety disorder (see, for example, Pergamin-Hight, L., et al. (2015) Content specificity of attention bias to threat in anxiety disorders: A meta-analysis, Clinical Psychology Review, 35, 10-18). Therefore, it is thought that the way attention is directed to displayed images containing the above target differs from the way attention is directed to other displayed images, which in turn affects resilience.

[0117] Therefore, in this embodiment, the user terminal 3b identifies the target of the user's anxiety using the first resilience score (evaluation result) regarding the facial expression image and / or the second resilience score (evaluation result) regarding the emotional image evaluated by the estimation server 4.

[0118] Fig. 14 is a block diagram showing a schematic configuration of a user terminal 3b. The user terminal 3b shown in Fig. 14 differs from the user terminal 3 shown in Fig. 2 in that an operation unit 15 is added, a resilience acquisition unit 26b is provided in the control unit 11 instead of the resilience acquisition unit 26, and an anxiety identification unit 27 is added to the control unit 11, but the other configurations are the same.

[0119] The operation unit 15 is made up of input devices such as a touch panel, a keyboard, etc., and receives input from the user. The operation unit 15 creates input data corresponding to the received input and transmits it to the control unit 11.

[0120] 1, the resilience acquisition unit 26b is similar to the resilience acquisition unit 26 shown in FIG. 1 except that the resilience acquisition unit 26b further acquires a first resilience score related to the facial expression image and a second resilience score related to the emotional image from the estimation server 4 via the communication network 5 and the communication unit 12. The resilience acquisition unit 26b sends the first resilience score and / or the second resilience score to the anxiety identification unit 27.

[0121] The anxiety identifying unit 27 identifies the target of the user's anxiety based on the first resilience score and / or the second resilience score from the resilience acquiring unit 26 b. The anxiety identifying unit 27 displays and outputs the identification result via the display control unit 23 and the display unit 13.

[0122] Specifically, the anxiety identifying unit 27 first acquires the target of anxiety to be identified from the user via the operation unit 15. Note that the index calculating unit 25 may estimate the target of anxiety of the user from the display image displayed by the display control unit 23 and the movement of the user's gaze measured by the gaze measuring unit 24.

[0123] For example, people with social anxiety disorder tend to pay attention to people's facial expressions. Therefore, when the target of the anxiety is a person, the anxiety identification unit 27 sets the first resilience score and the second resilience score from the resilience acquisition unit 26b as an anxiety score and a non-anxiety score, respectively. Then, when the anxiety score is lower than the non-anxiety score as a result of comparing the anxiety score with the non-anxiety score and the difference between the anxiety score and the non-anxiety score exceeds a threshold, the anxiety identification unit 27 identifies the person as the target of the user's anxiety.

[0124] Additionally, people with post-traumatic stress disorder tend to shift their attention away from emotional images that include traumatic events such as war or disasters, and people with specific phobias tend to take longer to shift their attention to emotional images that include objects of fear, such as snakes or spiders.

[0125] Therefore, when the object of anxiety is included in the emotional image, the anxiety identification unit 27 indicates the object of anxiety to the pairing unit 22. As a result, the pairing unit 22 uses data of a plurality of emotional images to create a first pair of images in which one of the images includes the object of anxiety, and a second pair of images in which neither of the images includes the object of anxiety.

[0126] The display control unit 23, the gaze measurement unit 24, and the index calculation unit 25 operate on the plurality of first paired images and the plurality of second paired images created by the pairing unit 22. As a result, the resilience acquisition unit 26b acquires the second resilience score for the first paired images and the second resilience score for the second paired images.

[0127] Next, the anxiety identifying unit 27 sets the second resilience score for the first pair of images from the resilience acquiring unit 26b as an anxiety score, and sets the second resilience score for the second pair of images from the resilience acquiring unit 26b as a non-anxiety score. Then, as a result of comparing the anxiety score with the non-anxiety score, if the anxiety score is lower than the non-anxiety score and the difference between the anxiety score and the non-anxiety score exceeds a threshold, the anxiety identifying unit 27 identifies the object of anxiety as the object of the user's anxiety.

[0128] If the object of anxiety is included in the facial expression image, the above-mentioned "emotional image" is changed to "facial expression image" and the above-mentioned "second resilience score" is changed to "first resilience score", and then the above-mentioned operation is executed. As described above, by comparing the anxiety score with the non-anxiety score, the object of the user's anxiety can be identified.

[0129] (Anxiety specific processing) Next, a description will be given of the anxiety identification process in the resilience estimation system 1 of this embodiment. When the target of anxiety is a person, after step S20 shown in Fig. 7, a step can be added in which the anxiety identifying unit 27 sets the first resilience score related to the facial expression image as an anxiety score, sets the second resilience score related to the emotional image as a non-anxiety score, and identifies the person as the target of the user's anxiety if the difference between the anxiety score and the non-anxiety score exceeds a threshold.

[0130] In the following, we will explain the case where the object of anxiety is included in the emotional image. When the object of anxiety is included in the facial expression image, as described above, the "emotional image" can be changed to the "facial expression image" and the "second resilience score" can be changed to the "first resilience score."

[0131] Fig. 15 is a flowchart showing the flow of the anxiety identification process in the resilience estimation system 1 of this embodiment. As shown in Fig. 15, first, in the user terminal 3, the image acquisition unit 21 acquires data of a plurality of emotional images from the image server 2 (S51). Next, the pairing unit 22 uses the data of the plurality of emotional images to create a plurality of first paired images, one of which includes the object of anxiety instructed by the anxiety identification unit 27, and a plurality of second paired images, both of which do not include the object of anxiety (S52).

[0132] Next, the display control unit 23 selects any one of the plurality of first paired images and the plurality of second paired images, and displays the selected paired images side by side on the display unit 13 (S53). At this time, the gaze measurement unit 24 measures the movement of the user's gaze using the video data from the imaging unit 14 (S54). Then, steps S53 and S54 are repeated until all of the plurality of first paired images and the plurality of second paired images have been selected (S55).

[0133] Next, the following operation is performed for the first pair of images. That is, the index calculation unit 25 calculates five gaze indices for the first pair of images from the gaze movements of the user measured for all of the first pair of images, and transmits them to the estimation server 4 (S56). Next, the resilience acquisition unit 26b acquires the second resilience score for the first pair of images from the estimation server 4 (S57).

[0134] On the other hand, the following operation is performed for the second pair of images. That is, the index calculation unit 25 calculates five gaze indices for the second pair of images from the gaze movements of the user measured for all of the second pair of images, and transmits them to the estimation server 4 (S58). Next, the resilience acquisition unit 26 acquires the second resilience score for the second pair of images from the estimation server 4 (S59). Note that steps S56 and S57 and steps S58 and S59 may be executed simultaneously, or one may be executed first.

[0135] Then, the anxiety identification unit 27 sets the second resilience score for the first pair of images as an anxiety score, and the second resilience score for the second pair of images as a non-anxiety score, and if the difference between the anxiety score and the non-anxiety score exceeds a threshold, identifies the object of anxiety as the object of the user's anxiety and displays it (S60, identification step). Thereafter, the anxiety identification process is terminated.

[0136] (Additional notes) The image server 2 may be separated into a server that stores facial expression image data and a server that stores emotional image data. The user terminal 3 may be integrated with either or both of the image server 2 and the estimation server 4. The user terminal 3 may be separated into a notebook PC equipped with a control unit 11 and a communication unit 12, and a tablet terminal equipped with a display unit 13 and a photographing unit 14. In this case, the notebook PC and the tablet terminal may be capable of communicating with each other via wire or wirelessly.

[0137] [Software implementation example] The functions of the user terminal 3 and the estimation server 4 (hereinafter referred to as "devices") can be realized by a program that causes a computer to function as the device, and a program that causes a computer to function as each control block of the device (particularly each part included in the control unit 11 and the control unit 31).

[0138] In this case, the device includes a computer having at least one control device (e.g., a processor) and at least one storage device (e.g., a memory) as hardware for executing the program. The control device and storage device execute the program, thereby realizing the functions described in each of the above embodiments.

[0139] The program may be non-transitory and may be recorded on one or more computer-readable recording media. The recording media may or may not be included in the device. In the latter case, the program may be supplied to the device via any wired or wireless transmission medium.

[0140] Furthermore, some or all of the functions of the control blocks can be realized by logic circuits. For example, an integrated circuit in which a logic circuit that functions as each of the control blocks is formed is also included in the scope of the present invention. In addition, the functions of the control blocks can also be realized by, for example, a quantum computer.

[0141] Furthermore, each process described in each of the above embodiments may be executed by AI (Artificial Intelligence). In this case, the AI ​​may run on the control device or on another device (for example, an edge computer or a cloud server).

[0142] The present invention is not limited to the above-described embodiments, and various modifications are possible within the scope of the claims. Embodiments obtained by appropriately combining the technical means disclosed in different embodiments are also included in the technical scope of the present invention. [Explanation of symbols]

[0143] 1 Resilience Estimation System 2 Image Server 3, 3a, 3b User terminal (estimation device) 4. Estimation Server (Estimation Device) 5. Communication Network 11 Control section 12 Communications Department 13 Display section 14 Photography Department 15 Control section 21 Image acquisition unit 22, 22a Pairing section 23 Display control unit 24 Line of sight measurement section (measurement section) 25, 25a Indicator calculation section 26, 26b Resilience Acquisition Section 27 Anxiety Identification Department 31 Control Unit 32 Communications Department 33 Storage section 41 Indicator acquisition part 42 First evaluation section (estimation section) 43 Second Evaluation Section (Estimation Section) 44 Resilience Estimation Unit (Estimation Unit) 51 First Evaluation Criterion 52 Second Evaluation Criterion

Claims

1. a display step of displaying the first image and the second image using facial expression images as display images, or displaying the third image and the fourth image using emotional images as display images; a measuring step of measuring a movement of a user's line of sight when viewing the displayed image; an index obtaining step of obtaining a gaze index value from the measured gaze movement; and an estimation step of estimating the psychological resilience of the user by evaluating the value of the gaze index for the facial expression image based on a first evaluation criterion and evaluating the value of the gaze index for the emotional image based on a second evaluation criterion.

2. a display step of displaying the first image and the second image using facial expression images as display images; a measuring step of measuring a movement of a user's line of sight when viewing the displayed image; an index obtaining step of obtaining a gaze index value from the measured gaze movement; and estimating the psychological resilience of the user based on a first evaluation criterion using the gaze index for the facial expression image.

3. a display step of displaying the third image and the fourth image using emotional images as display images; a measuring step of measuring a movement of a user's line of sight when viewing the displayed image; an index obtaining step of obtaining a gaze index value from the measured gaze movement; and an estimation step of estimating the psychological resilience of the user based on a second evaluation criterion using the gaze index for the emotional image.

4. The psychological resilience estimation method according to claim 1 or 2, wherein the first evaluation criterion uses, as the gaze index, a saccade latency until the gaze is first directed to an image with a higher emotional valence.

5. 5. The psychological resilience estimation method according to claim 4, wherein the first evaluation criterion further uses as the gaze indexes the percentage of the total time spent gazing at the image with higher emotional valence, the frequency of gazing at the image with higher emotional valence, and the duration of initial continuous gaze at the image with higher emotional valence.

6. The psychological resilience estimation method according to claim 5 , wherein the first evaluation criterion uses a weighted sum of the plurality of gaze indices for evaluation.

7. The psychological resilience estimation method according to claim 1 or 3, wherein the second evaluation criterion uses, as the gaze index, a frequency of directing gaze to an image having a higher emotional valence.

8. The psychological resilience estimation method according to claim 7, wherein the second evaluation criterion further uses, as the gaze index, a proportion of initial gaze directed to an image with a higher emotional valence and a duration of initial gaze directed to an image with a higher emotional valence.

9. The psychological resilience estimation method according to claim 8 , wherein the second evaluation criterion uses a weighted sum of the plurality of gaze indices for evaluation.

10. before the display step, The psychological resilience estimation method according to claim 1 , further comprising a fixation step of displaying a fixation point image for fixing a gaze point.

11. The index acquisition step includes: Obtaining the gaze index value for the facial expression image from a gaze movement of the user looking at the first image and the second image having different emotional valences and a gaze movement of the user looking at the first image and the second image having neutral emotional valences, or 2. The psychological resilience estimation method according to claim 1, wherein the value of the gaze index for the facial expression image is obtained from the gaze movement of the user when looking at the third image and the fourth image, which have different emotional valences, and the gaze movement of the user when looking at the third image and the fourth image, which both have neutral emotional valences.

12. The psychological resilience estimation method according to claim 1 , further comprising an identification step of identifying a target of the user's anxiety by comparing the evaluation results regarding the facial expression image and the evaluation results regarding the emotional image in the estimation step.

13. Identifying the target of the user's anxiety by comparing an evaluation result regarding the facial expression image estimated in the estimation step when one of the first image and the second image includes a target of anxiety with an evaluation result regarding the facial expression image estimated in the estimation step when neither the first image nor the second image includes the target of anxiety, or 2. The psychological resilience estimation method of claim 1, further comprising an identification step of identifying the target of the user's anxiety by comparing an evaluation result regarding the emotional image estimated in the estimation step when one of the third image and the fourth image includes a target of anxiety with an evaluation result regarding the emotional image estimated in the estimation step when neither the third image nor the fourth image includes the target of anxiety.

14. a display control unit that displays the first image and the second image using facial expression images as display images, or that displays the third image and the fourth image using emotional images as display images; a measurement unit that obtains a measurement result of measuring the movement of the user's line of sight as they view the displayed image; an index acquisition unit that acquires a gaze index value from the measured gaze movement; an estimation unit that estimates the psychological resilience of the user by evaluating the value of the gaze index for the facial expression image based on a first evaluation criterion and evaluating the value of the gaze index for the emotional image based on a second evaluation criterion.

15. a display control unit that displays the first image and the second image using facial expression images as display images; a measurement unit that obtains a measurement result of measuring the movement of the user's line of sight as they view the displayed image; an index acquisition unit that acquires a gaze index value from the measured gaze movement; and an estimation unit that estimates the psychological resilience of the user based on a first evaluation criterion using the gaze index for the facial expression image.

16. a display control unit that displays the third image and the fourth image using emotional images as display images; a measurement unit that obtains a measurement result of measuring the movement of the user's line of sight as they view the displayed image; an index acquisition unit that acquires a gaze index value from the measured gaze movement; and an estimation unit that estimates the psychological resilience of the user based on a second evaluation criterion using the gaze index for the emotional image.

17. A control program for causing a computer to execute the display step, the measurement step, the index acquisition step, and the estimation step in the psychological resilience estimation method according to any one of claims 1 to 3.

Images