JP7587794B2 - Estimation device, program, and estimation system - Google Patents

Description

The present invention relates to a device, program, and system for estimating the attractiveness of an object.

Conventionally, there are methods for estimating emotions toward an object, such as a product, from fluctuations in brain activity with long latency (for example, Patent Document 1). The technology in Patent Document 1 estimates emotions toward an object from brain activity with long latency, from tens of seconds to several minutes or more. Such brain activity lasting more than a few seconds is used to analyze information processing, including a final judgment that includes all higher-level cognitive processes, or the process of assigning meaning to that judgment. Such judgments after higher-level cognitive processes include not only the person's instinctive preferences and intuitive interests, but also the experimental environment and metacognition of one's own judgment.

In experimental psychology research, it is said that emotions toward an object are instinctively or intuitively judged unconsciously within one second. If this is the case, emotions toward an object are primarily processed instinctively or intuitively within one second, and thereafter are processed according to a linguistic interpretation, taking into account additional information about the object, such as the object's meaning, efficacy, and price, as well as the situation and environment in which the person finds themselves.

JP 2015-229040 A

It is expected that the degree to which a subject finds an object attractive (hereinafter referred to as attractiveness) can be estimated from fluctuations in the subject's brain activity. However, if the above-mentioned attractiveness is estimated using conventional estimation methods, the fluctuations in brain activity with a long latency from one second onwards obtained from the subject will be used. Therefore, it can be said that the results of estimating attractiveness using conventional methods are easily influenced by the subject's linguistic thoughts about the object, and are easily affected by the relationship between the product's additional information and the subject's situation or environment at the time. Therefore, there is a demand for an objective and reliable method of estimating attractiveness that is not easily influenced by linguistic thoughts, etc.

In view of the above, the object of the present invention is to provide a method for estimating the instinctive or intuitive attractiveness that arises the moment one comes face to face with a product.

In order to solve the above problem, an estimation device according to an embodiment of the present invention comprises:
An estimation device for estimating an attractiveness indicating a degree of interest of a subject in an object, comprising:
an electroencephalogram data acquisition unit that acquires electroencephalogram data of the subject for a period up to 1000 milliseconds after at least one of the five senses of the subject is stimulated by the object;
an event-related potential acquiring unit for acquiring an event-related potential from the electroencephalogram data;
An estimation unit that estimates the attractiveness level based on the event-related potential;
Includes.

A program according to another aspect of the present invention comprises:
A program for estimating an attractiveness indicating a degree of interest of a subject in an object, comprising:
an electroencephalogram data acquisition unit that acquires electroencephalogram data of the subject within 1000 milliseconds after at least one of the five senses of the subject is stimulated by the object;
an ERP acquisition unit for acquiring an event-related potential from the electroencephalogram data;
An estimation unit that estimates the attractiveness level based on the event-related potential;
This is a program that functions as a

A estimation system according to still another aspect of the present invention comprises:
An estimation system for estimating an attractiveness indicating a degree of interest of a subject in an object, comprising:
A five-senses stimulating device for stimulating at least one of the five senses of the subject by the object;
an electroencephalograph for measuring the brain waves of the subject and outputting brain wave data;
a processing device for acquiring electroencephalogram data from the electroencephalograph for 1000 milliseconds after at least one of the five senses of the subject is stimulated by the object, acquiring an event-related potential based on the electroencephalogram data, and estimating the attractiveness level based on the event-related potential;
Equipped with.

In addition, any combination of the above, or mutual substitution of the components or expressions of the present invention among methods, devices, programs, temporary or non-temporary storage media recording programs, systems, etc. are also valid aspects of the present invention.

The present invention makes it possible to provide a method for estimating the instinctive or intuitive attraction that arises the moment a product is encountered.

FIG. 1 is a schematic configuration diagram of an estimation system according to an embodiment. 2 is a block diagram showing an example of the configuration of a computer that functions as the processing device in FIG. 1 . FIG. 2 is a functional block diagram of the processing device of FIG. 1 . FIG. 1 is a diagram showing a flow of an estimation method according to an embodiment. FIG. 5(a) is a diagram showing an example of electroencephalogram data (waveform and ERP) when the degree of intimacy is low, and FIG. 5(b) is a diagram showing an example of electroencephalogram data when the degree of intimacy is high. FIG. 13 is a diagram illustrating an example of the correlation between familiarity and the magnitude (value) of the ERP amplitude. 1A and 1B are diagrams for explaining a method of determining an ERP criterion and a method of estimating an intimacy level using the ERP criterion according to an embodiment. FIG. 13 is a diagram showing a model of the relationship between intimacy and attractiveness.

The present invention will be described below based on a preferred embodiment with reference to the drawings. In the embodiment and the modified example, the same or equivalent components are given the same reference numerals, and duplicated descriptions will be omitted as appropriate.

[Embodiment]
First, an overview of this embodiment will be described. The estimation device of this embodiment estimates the attractiveness of an object based on event-related potentials (ERPs) obtained from electroencephalogram data of a subject when the subject's vision is stimulated by looking at the object. The attractiveness here indicates the degree to which the subject is interested in the object. The estimation device of this embodiment estimates the attractiveness, which indicates whether "attention" is directed more subconsciously and whether the subject wants to pick up a product, in a process within 1000 milliseconds after the subject is stimulated by the object.

More specifically, the estimation device of this embodiment estimates "intimacy," which indicates how well a person knows an object, from ERP. Utilizing the idea that if this intimacy is too high or too low, a product is unlikely to be perceived as attractive, the intimacy estimation result is converted nonlinearly to estimate attractiveness.

Figure 1 shows the configuration of an estimation system for estimating the attractiveness felt by a subject of an embodiment of the present invention for an object. The estimation system 30 includes a display device 2, an electroencephalograph 3, a processing device 4, and an input device 5.

The display device 2 displays an image 9 of an object toward the subject 1. The display device 2 of this embodiment is a head mounted display (HMD), but is not limited thereto, and may be, for example, a tablet terminal having a liquid crystal display screen (including a touch panel) or a display device capable of wireless communication with a processing device 4 such as a smartphone. The display device 2 of this embodiment is an example of a five-senses stimulation device for stimulating at least one of the five senses of the subject 1 with an object.

The electroencephalograph 3 measures the brain waves of the subject 1 in synchronization with the display of an image 9 of the object on the display device 2, and outputs the results as brain wave data to the processing device 4. The electroencephalograph 3 detects the minute potential difference that occurs between the probe electrode 7 attached to the head of the subject 1 and the reference electrode 8 attached to the ear, and amplifies this potential difference with a built-in differential amplifier to obtain an electroencephalogram (EEG).

The processing device 4 processes the electroencephalogram data of the subject 1 from the electroencephalograph 3 to estimate the attractiveness of the object. The processing device 4 of this embodiment is an example of an estimation device. The processing device 4 can be configured, for example, as a computer (PC).

2 is a block diagram showing an example of the configuration of a computer (PC) that functions as the processing device 4 in FIG. 1. The processing device 4 includes a central processing unit (CPU) 40, a storage means 41, and various I/Fs 43, which are connected to each other via a bus 42. The various I/Fs 43 are used as a general term including an input I/F, an output I/F, an external storage I/F, an external communication I/F, etc. Each I/F is connected to a corresponding input/output means 44 such as a keyboard, a mouse, a communication port, a display means 45 such as a CRT or an LCD, and an external storage means 46 such as a semiconductor memory or HDD connected via a USB. The storage means 41 is composed of a RAM, a ROM, a semiconductor memory such as a flash memory, an HDD, etc. The input/output means 44 consisting of a keyboard and a mouse can be used as the input device 5 in FIG. 1, and the display means 45 can be used as the display device 2. If the display device 2 is a touch panel, the panel itself can be used as the input device. In the processing device 4, the brain wave data received from the electroencephalograph 3 is stored in the storage means 41. The storage means 41 stores various programs for executing various controls in the CPU 40. The processing device 4 performs various processes such as averaging the multiple brain wave data stored in the storage means 41 to calculate an event-related potential (ERP). The details of the processes will be described later.

The input device 5 is a device for input by a user, including the subject 1. In FIG. 1, a mouse 5 operated by a hand 6 is illustrated as an example of the input device 5. The input device 5 is not limited to this, and may be, for example, a touch panel, etc.

Figure 3 is a functional block diagram of the processing device 4 in Figure 1. Each functional block shown in the diagram below is realized in hardware terms by a computer having calculation functions, control functions, memory functions, input functions, and output functions, various electronic elements, mechanical parts, etc., and in software terms by a computer program, etc., but here, functional blocks realized by the cooperation of these are depicted. Therefore, it will be understood by those skilled in the art that these functional blocks can be realized in various forms by combining hardware and software.

The processing device 4 in FIG. 3 includes an EEG data acquisition unit 101, an ERP acquisition unit 102, an estimation unit 103, and a display control unit 104.

The brainwave data acquisition unit 101 acquires brainwave data of the subject 1 when the visual sense is stimulated by an object.

The ERP acquisition unit 102 acquires ERP from the brainwave data.

The estimation unit 103 estimates the attractiveness based on the ERP. The estimation unit 103 includes an intimacy estimation unit 111 and an attractiveness estimation unit 112. The intimacy estimation unit 111 estimates the intimacy indicating the degree to which the subject knows the object based on the ERP. The attractiveness estimation unit 112 estimates the attractiveness based on the estimated intimacy.

The display control unit 104 controls the images to be displayed on the display screen of the display device 2.

Figure 4 is a diagram showing the flow of the estimation method of this embodiment. The estimation process S10 in Figure 4 can be executed, for example, in the estimation system 30 in Figure 1 under the control of the processing device 4. In this embodiment, as an example, the attractiveness of the design of an object is estimated.

In step S11, the brainwave data acquisition unit 101 acquires the brainwaves of the subject 1 using the electroencephalograph 3 before the start of image display in step S12 described below. In this embodiment, the brainwave data acquisition unit 101 acquires brainwave data of the subject 1 from 100 milliseconds to 0 milliseconds before visual stimulation by image display in step S12 described below. A baseline is set using the brainwave data acquired in step S11. A specific method for measuring brainwaves will be described later in step S13.

In step S12, the display control unit 104 displays an image 9 of the object on the display screen of the display device 2. The object here includes products, goods, buildings, scenery (landscape), animals, logos, etc. that can be visualized as an image, and basically everything that one wants to know how attractive people find it. For example, products include automobiles, houses, and machine control panels, and goods may include the designs of the main body and contents of all things that are distributed and sold, such as clothes, tableware, furniture, and food, as well as the designs of packages and cases, or advertising images. The display control unit 104 of this embodiment displays a virtual reality (VR) image as the image 9 of the object on the HMD when the subject 1 wears the HMD as the display device 2. This VR image is controlled so that it is always displayed at the same distance and with the same viewing angle even if the subject 1 moves his or her head. By displaying the VR image in this way, the measurement conditions can be stabilized no matter where the evaluation test is performed. The image 9 may be not only a still image such as a photograph or illustration, but also a moving image that changes over time. Before displaying the image 9 of the object, a gaze point screen may be displayed to stabilize eye movement and make it easier to measure brain waves.

There are two main purposes for displaying the image 9 of the object. The first is (i) when displaying an image of an object whose intimacy with the subject is unknown. The second is (ii) when displaying an image of an object whose intimacy with the subject is known. The latter (ii) is performed prior to or in parallel (successively) with the measurement of (i) in order to obtain an ERP standard required for estimating the intimacy of the former (i). In either case, the image is displayed for about 500 milliseconds, and when multiple images are displayed, it is desirable to leave an interval of about 1 second between the display of each image. This cycle is repeated multiple times (for example, 10 to 20 times) for one image. The image display for about 500 milliseconds was selected as a time of 400 milliseconds or more based on the knowledge obtained by the inventors described above that valid brain wave data can be obtained about 400 milliseconds after the image is displayed. Therefore, the image display time is essentially at least 400 milliseconds, but may be, for example, 450 milliseconds or 550 milliseconds.

In step S13, the brainwave data acquisition unit 101 acquires the brainwaves of the subject 1 after the image display of step S12 is started using the electroencephalograph 3. Specifically, the brainwave data acquisition unit 101 detects a small potential difference generated between the probe electrode 7 attached to the head of the subject 1 and the reference electrode 8 attached to the ear, which is obtained for each image display (one cycle) using the electroencephalograph 3, and amplifies the potential difference with a built-in differential amplifier to obtain an electroencephalogram (EEG). The brainwave data acquisition unit 101 acquires brainwave data from the obtained electroencephalogram (EEG) and stores it in the storage means 41. Specifically, the brainwave data acquisition unit 101 filters the electroencephalogram (EEG) obtained by the electroencephalograph 3 to remove noise, then converts it into a digital signal (brainwave data) using an A/D converter built into or external to the electroencephalograph 3 and sends it to the processing device 4. The processing device 4 stores the brainwave data received via the brainwave data acquisition unit in the storage means 41.

In step S14, the ERP acquisition unit 102 acquires an ERP by averaging the multiple pieces of EEG data stored in the storage means 41, with the baseline set based on the EEG data acquired in S11 as a reference. Figure 5 shows an example of EEG data (waveform and ERP). The time (milliseconds) on the horizontal axis indicates the elapsed time from the start time of image display (stimulus). Figures 5(a) and 5(b) show data when images of objects with different attractiveness to subject 1 are displayed. Here, Figure 5(a) shows a case where the intimacy level is relatively low, and Figure 5(b) shows a case where the intimacy level is relatively high.

The multiple overlapping waveforms at the top of Fig. 5(a) and Fig. 5(b) are EEGs obtained when multiple trials (image displays) were performed on the same image. The lower waveforms in Fig. 5(a) and Fig. 5(b) each show an example of an ERP waveform obtained by averaging multiple brainwave data. Comparing the two ERP waveforms, it can be seen that the ERP waveform in Fig. 5(a) when the intimacy level is low has an amplitude fluctuation (decrease) at about 200 milliseconds, whereas the ERP waveform in Fig. 5(b) when the intimacy level is high has an amplitude increase at about 100 milliseconds and further has a relatively large amplitude decrease at about 300 to 400 milliseconds. The ERP acquisition unit 102 supplies the acquired ERP to the intimacy level estimation unit 111.

In step S15, the intimacy estimation unit 111 estimates the intimacy level based on the ERP supplied from the ERP acquisition unit 102. The intimacy estimation unit 111 supplies the intimacy level estimation result to the attractiveness estimation unit 112. The intimacy level estimation method is described below.

The intimacy estimation unit 111 compares the ERP calculated in step S14 with the ERP standard. The ERP standard represents the magnitude (value) of the ERP amplitude corresponding to the level of intimacy, that is, the correlation between the intimacy and the magnitude (value) of the ERP amplitude, which is determined in advance for each subject or prior to the comparison, and the correspondence between the two is uniquely determined (there is a correlation), for example, as shown in the graph in FIG. 6. The example in FIG. 6 shows a relationship in which the ERP increases linearly as the intimacy increases. The correspondence as shown in FIG. 6 can be stored in the storage means 41 as a single table of data. Such an ERP standard can be determined, for example, as follows:

In calculating the ERP standard, the flow up to finding the ERP is basically the same as steps S11 to S14 in FIG. 4. The difference is that in step S12, the image of the object shown to the subject is an image of an object known to the subject and the degree of intimacy is known as described above (ii). Specifically, for example, n images {R1(x1), R2(x2), R3(x3), , , , Rn(xn)} (n ≧ 2) with known intimacy (x1, x2, , , , xn) and different intimacy levels are displayed on the display device 2 in random order. Here, the larger the value of intimacy (x), the better the familiarity (higher the recognition). Steps S13 to S14 are executed for each image displayed, and the corresponding ERP is calculated.

With reference to Figure 7, a method of determining an ERP criterion in one embodiment of the present invention and a method of estimating intimacy using this will be described. Figure 7(a) shows the ERP obtained when an image 9 of an object to be evaluated is shown on a display screen 2 to each subject 1. At this time, in order to determine the ERP criterion, an image of an object known to the subject 1 and of which the degree of intimacy is known is displayed. Here, the ERP can be obtained by averaging the electroencephalogram data for each image of a known object to be evaluated. Note that while the ERP in Figure 7(a) shows ERP waveforms for two types of evaluation objects, in reality there are many evaluation objects (assuming there are N objects), and there are as many ERP waveforms as there are evaluation objects (N).

Of the ERP waveforms of the two evaluation objects in Fig. 7(a), the amplitudes of the ERP waveforms at time t are P1(t) and P2(t) in this figure. In addition, for each evaluation object, a subjective rating value of intimacy is set at the same time, and the subjective rating values of intimacy for the above two evaluation objects are assumed to be a and b as shown in the figure. The ERP amplitude Pi(t) and the subjective rating value of intimacy at time t for N evaluation objects measured in this way are taken on the vertical and horizontal axes, respectively, and three scatter diagrams are drawn for three times t = t _i , t _M , and t _j in Fig. 7(b), Fig. 7(c), and Fig. 7(d). The correlation coefficient r(t) shown in each figure is obtained as a result of performing correlation analysis for each time t = t _i , t _M , and t _j . From the relationship in the figure, it can be seen that the correspondence between the ERP waveform amplitudes P1(t), P2(t) and the intimacy levels (subjective ratings of intimacy) a, b can be uniquely determined.

From the graph shown in Fig. 7(e) obtained as a result of performing correlation analysis for times t = t _i , t _M , and t _j , the time (t = t _M ) at which the correlation coefficient r(t) is maximum, or the time period before or after that time, is determined. The scatter plot of Fig. 7(c) showing the subjective rating of intimacy and the ERP amplitude at this time (or time period) becomes a model for estimating the subjective rating of intimacy from the ERP amplitude Pi(t), i.e., the ERP criterion (the correspondence between ERP amplitude and intimacy).

The intimacy estimation unit 111 estimates the intimacy of an evaluation object whose intimacy is unknown to the subject, for example, by using the ERP criterion shown in Fig. 7(c) obtained as described above. Specifically, when an image of a product or the like to be evaluated whose intimacy is unknown is shown to the subject in step S12 described above, the ERP amplitude P( _tM ) at time t = _tM calculated in steps S13 to S14 is applied (compared) to the ERP criterion (relationship between ERP amplitude P( _tM ) and intimacy) shown in Fig. 7(c) to determine the intimacy corresponding to that amplitude P( _tM ).

Note that Figures 7(b) to (d) are based on the assumption that there is a linear relationship between the ERP amplitude P(t) and the subjective rating of intimacy, but it is also possible to use a nonlinear function approximation (e.g., a logarithmic function) as a model describing the relationship between the ERP amplitude P(t) and the subjective rating of intimacy. In addition, in the embodiment described above, the ERP for the image display of each evaluation target is used as an EEG index for estimating the subjective rating of intimacy, but it is also possible to adopt a method in which the changes in the intensity of alpha waves, beta waves, theta waves, and gamma waves that occur with the image display of the evaluation target are used as EEG indices.

In step S16, the attractiveness estimation unit 112 estimates the attractiveness based on the intimacy estimation result. The method for estimating the attractiveness is described below.

Refer to FIG. 8. If the intimacy level is too high or too low, the product is unlikely to be perceived as attractive, and the level of intimacy cannot be directly related to the level of attractiveness. Therefore, the inventors have modeled the relationship between intimacy and attractiveness based on knowledge of design and marketing (FIG. 8). According to the model shown in FIG. 8, in descending order of intimacy with respect to the object, the subjects will feel, for example, nostalgic déjà vu, standard (standard model), mimetic déjà vu, evolution, innovation, novelty, eccentricity, and too strange to understand. The extent to which the subjects feel that the target product is evolution, innovation, and novelty is likely to be within the acceptable range for the product to be accepted by the market. As shown in FIG. 8, regardless of the type of product or the historical background, it is believed that both designs with very high intimacy and designs with very low intimacy have little appeal as product designs.

In this embodiment, in order to appropriately convert intimacy into attractiveness taking into account the relationship between intimacy and attractiveness as described above, the Gaussian function h(x) shown in the following formula (1) is used.

Here, the Gaussian function h(x) is a bell-shaped function that converts the intimacy degree x into the attractiveness degree. c is a width determination coefficient for determining the width (degree of smoothness) of the bell-shaped graph shape of the Gaussian function h(x). The width determination coefficient c is determined according to the category of the object. For example, the width determination coefficient c can be the deviation σ when the intimacy degree estimated for a plurality of other reference products in the same category as the target product to be estimated is normalized, or an arbitrary value that can be set by the user according to the category of the target product. d is a median correction coefficient for determining the median of h(x). The Gaussian function h(x) is at its maximum value when x=d, and the larger the width determination coefficient c in the formula, the wider the tail of the curve becomes. The median correction coefficient d (median of the function) is determined according to the category of the object.

The median correction coefficient d is expressed by the following formula (2): In formula (2), T is an attenuation coefficient, and α and β are linear conversion coefficients that can be set to any value by the user.
d = αT + β Equation (2)

The attenuation coefficient T indicates the product value for a target product category, and is set based on the marketing data of that product category. The longer the standard product value is maintained, the smaller the attenuation coefficient is set, and in a product category where standards change in the short term, the larger the attenuation coefficient is set. For example, for N products in a certain category, the attenuation coefficient T is set using the following formula (3) where F ₀ (n) is the production volume at the time of release of product n and F _t (n) is the production volume t hours after release.

The attenuation coefficient T may be set arbitrarily based on user input based on the marketing data of the product category. By adjusting the attenuation coefficient T, the median correction coefficient d varies. Therefore, it is possible to adjust the median of h(x) to match the characteristics of the market value of the product category.

The attractiveness estimation unit 112 fits the width and median of the graph shape of the Gaussian function h(x) to correspond to the width determination coefficient c and median correction coefficient d set according to the category of the object as described above. The attractiveness estimation unit 112 estimates the attractiveness by inputting the intimacy estimated in step S15 to the fitted Gaussian function h(x). The attractiveness estimation unit 112 supplies the estimated attractiveness to the display control unit 104.

In step S17, the display control unit 104 outputs the attractiveness. In this embodiment, the display control unit 104 displays the attractiveness on the display screen of the display device 2. This makes it possible to grasp the attractiveness that the subject 1 feels about the object.

After step S17, estimation process S10 ends.

The action and effect of this embodiment will be explained.

According to this embodiment, it is possible to estimate "attractiveness," which is the automatic and latent interest that arises the moment a product is viewed, from a small amount of short-term electroencephalogram data. As a result, it is possible to obtain quantitative and objective data on the attractiveness felt by consumers for various products. In addition, because estimation is possible from short-term data within 1000 milliseconds after visual stimulation, it is possible to conduct tests for estimating attractiveness in a shorter time.

In this embodiment, the electroencephalogram data used to estimate attractiveness is electroencephalogram data from 400 milliseconds to 1000 milliseconds after visual stimulation. This configuration provides electroencephalogram data that is effective for estimating attractiveness, improving the accuracy of the attractiveness estimation.

In this embodiment, the estimation unit 103 estimates the attractiveness level based on the estimated intimacy level. With this configuration, the attractiveness level is estimated based on the intimacy level obtained with a relatively high degree of accuracy, improving the accuracy of the estimation of the attractiveness level.

In this embodiment, the baseline is set using the brainwave data of subject 1 from 100 milliseconds to 0 milliseconds before the visual stimulation in step S11. With this configuration, the brainwave data immediately before the visual stimulation of subject 1 is stimulated by the image display is set as the baseline, so that the brainwave data can be effectively analyzed and the accuracy of the estimation of attractiveness is improved.

(Modification)
Modifications of the embodiment will be described below.

The processing device 4 may further include a judgment unit that judges whether the subject 1 feels attraction to the object based on the estimated attractiveness. An example of the judgment method in this case is described below. For example, the judgment unit calculates the attractiveness of other reference objects belonging to the object category, and calculates what percentage of the attractiveness of the object is included in the top of the attractiveness of all objects belonging to the object category. Here, an evaluation standard value that is a standard for whether the subject 1 feels attraction to the object is stored in the storage means. The evaluation standard value indicates what percentage of the attractiveness of all objects that the subject 1 feels is the top attractiveness of the object. This evaluation standard value is set in advance by the evaluator. The judgment unit judges that the subject 1 feels attraction to the object when the percentage indicated by the evaluation standard value (e.g., 20%) includes the percentage indicated by the attractiveness evaluation value (e.g., 10%). The judgment unit supplies the judgment result to the display control unit 104. The display control unit 104 causes the judgment result to be displayed on the display screen of the display device 2. The determination unit may also determine whether or not the subject 1 feels attraction to the object based on a comparison between the estimated attractiveness and a predetermined attractiveness threshold. In this case, the determination unit may determine that the subject 1 feels attraction to the object when the estimated attractiveness is greater than the attractiveness threshold.

In the embodiment, the electroencephalogram data of subject 1 when his/her vision is stimulated by looking at an object is used, but this is not limited to vision, and electroencephalogram data when at least one of subject 1's five senses (sight, taste, smell, touch, hearing) is stimulated by an object may also be used.

In the embodiment, an image of an object is displayed to stimulate the sense of sight, but this is not limited thereto, and for example, an actual object may be presented to subject 1.

The method of estimating intimacy is not limited to the method described in step S15 above. Intimacy can be estimated by various other methods.

In the embodiment, the attractiveness is estimated based on the degree of intimacy, but this is not limited to this. Any parameter that can be estimated by a known method may be used, as long as it is a parameter that is unlikely to be perceived as an attractive product if it is too high or too low, and is likely to be perceived as an attractive product near its median. For example, the attractiveness may be estimated using novelty, which indicates the degree of novelty or rarity. The attractiveness may be estimated by machine learning a learning model that inputs electroencephalogram data or EPR and outputs attractiveness, and inputting electroencephalogram data into this learning model to obtain the attractiveness from the learning model.

Furthermore, attractiveness may be estimated based on the degree of intimacy and the degree of tension/relaxation of subject 1. In this case, the degree of tension/relaxation of subject 1 may be estimated based on alpha suppression by time-frequency analysis of ERP from the time the subject is stimulated by the object until 500 milliseconds.

In the embodiment, the attractiveness is estimated from the intimacy using the Gaussian function h(x) shown in formula (1), but this is not limited to this. The function y = f(x) used to estimate the attractiveness y may be a function that is upwardly convex in the range between a relatively high value and a relatively low value of the intimacy. The function f(x) may be a function that makes the attractiveness value relatively low when the intimacy is a relatively high value or a relatively low value, and makes the attractiveness value relatively high when the intimacy is a value between a relatively high value and a relatively low value. For example, the function f(x) may be a fuzzy inference membership function μ(x) (= f(x)) exemplified by the following formula (4), or may be a function having a curved shape with an upwardly convex portion in a sine wave, a triangular wave, or the like.

Multiple pairs of probe electrodes 7 and reference electrodes 8 may be provided, and the attractiveness level may be estimated by combining the brainwave data obtained from these multiple pairs of electrodes.

In the embodiment, to set the baseline, the brainwave data of subject 1 between 100 milliseconds and 0 milliseconds before the visual stimulation is applied in step S11 is acquired, but this is not limited to this. The baseline may be set using brainwave data from more than 100 milliseconds before the visual stimulation is applied, or the brainwave data of subject 1 may be measured in advance and the baseline may be set in advance based on the measured data.

1 Subject, 2 Display device, 3 Electroencephalograph, 4 Processing device, 5 Input device, 30 Estimation system, 101 EEG data acquisition unit, 102 ERP acquisition unit, 103 Estimation unit, 104 Display control unit, 111 Intimacy estimation unit, 112 Attractiveness estimation unit.

Claims (9)

An estimation device for estimating an attractiveness indicating a degree of interest of a subject in an object, comprising:
an electroencephalogram data acquisition unit that acquires electroencephalogram data of the subject for a period up to 1000 milliseconds after at least one of the five senses of the subject is stimulated by the object;
an event-related potential acquiring unit for acquiring an event-related potential from the electroencephalogram data;
An estimation unit that estimates the attractiveness level based on the event-related potential;
Including,
The estimation device includes an intimacy estimation unit that estimates an intimacy indicating the degree to which the subject knows the object based on the event-related potential, and an attractiveness estimation unit that estimates the attractiveness based on the estimated intimacy . The estimation device according to claim 1, wherein the electroencephalogram data is electroencephalogram data from 400 milliseconds after at least one of the subject's five senses is stimulated to 1000 milliseconds after the stimulation. 3. The estimation device according to claim 1 or 2, wherein the attractiveness estimation unit estimates the attractiveness using a function y = f(x) that is upwardly convex in a range between a relatively low value and a relatively high value of the intimacy x, where x is the intimacy and y is the attractiveness. The estimation apparatus according to claim 3 , wherein the median value of the function is set based on a category of the object. The estimation device according to claim 1 , further comprising a determination unit that determines whether or not the subject feels attractiveness to the object based on the estimated attractiveness. The estimation device according to claim 1 , wherein the electroencephalogram data is electroencephalogram data obtained when the subject's sense of vision is stimulated by the object. the electroencephalogram data acquisition unit acquires electroencephalogram data of the subject from 100 milliseconds to 0 milliseconds before the stimulation;
7. The estimation device according to claim 1, wherein the event-related potential acquisition unit acquires the event-related potential from the electroencephalogram data obtained after the stimulation, using a baseline set based on the electroencephalogram data obtained before the stimulation as a reference. A program for estimating an attractiveness indicating a degree of interest of a subject in an object, comprising:
an electroencephalogram data acquisition unit that acquires electroencephalogram data of the subject within 1000 milliseconds after at least one of the five senses of the subject is stimulated by the object;
an ERP acquisition unit for acquiring an event-related potential from the electroencephalogram data;
An estimation unit that estimates the attractiveness level based on the event-related potential;
Function as a
The estimation unit includes an intimacy estimation unit that estimates an intimacy indicating the degree to which the subject knows the object based on the event-related potential, and an attractiveness estimation unit that estimates the attractiveness based on the estimated intimacy, the program. An estimation system for estimating an attractiveness indicating a degree of interest of a subject in an object, comprising:
A five-senses stimulating device for stimulating at least one of the five senses of the subject by the object;
an electroencephalograph for measuring the brain waves of the subject and outputting brain wave data;
a processing device for acquiring electroencephalogram data from the electroencephalograph for 1000 milliseconds after at least one of the five senses of the subject is stimulated by the object, acquiring an event-related potential based on the electroencephalogram data, and estimating the attractiveness level based on the event-related potential;
Equipped with
The processing device is an estimation system including: an intimacy estimation unit that estimates an intimacy indicating the degree to which the subject knows the object based on the event-related potential; and an attractiveness estimation unit that estimates the attractiveness based on the estimated intimacy .

Images