Brain activity evaluation device, brain activity evaluation method, and brain activity evaluation program

Abstract

To highly accurately evaluate a brain activity.SOLUTION: A brain activity evaluation device comprises: an acquisition unit which acquires an f-MRI image to at least one stimulation; a correlation derivation unit which performs statistical processing on the brain activity in the acquired f-MRI image for each voxel and derives a correlation between voxel data and a degree of a value expressed by the voxel data; a region of interest determination unit which determines a region of interest being a brain region being the analysis object in any brain region by selecting a positive correlation voxel having a positive correlation and a negative correlation voxel having a negative correlation; and a brain activity evaluation model generation unit which extracts brain activity data to the stimulation of the determined region of interest and generates a brain activity evaluation model for evaluating the brain activity to the stimulation on the basis of the extracted brain activity data.SELECTED DRAWING: Figure 2

Description

[Technical field]

[0001] The present invention relates to a brain activity evaluation device, a brain activity evaluation method, and a brain activity evaluation program. [Background technology]

[0002] In the above technical field, Patent Document 1 discloses a technique for correlating brain activity with perceptual content in an intermediate representation space (semantic space), projecting brain activity onto the semantic space, learning the correspondence with the semantic space as a quantitative model (projection function), and correlating the representation in the semantic space with brain activity output in order to estimate the perceptual semantic content. Patent Document 2 discloses a technique for constructing a brain semantic space by statistical learning processing from the measurement results of brain activity and annotation vectors, and evaluating new material according to the distance between the position corresponding to the content of new material and the position of the target word in the brain semantic space. Patent Document 3 discloses a technique for predicting brain activity itself by learning the relationship of brain activity data for learning stimuli using a regression model to generate an encoder, and estimating perceptual and cognitive content from predicted brain activity data predicted using a decoder for the predicted brain activity. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] Patent No. 6618702 [Patent Document 2] JP 2017-129924 A [Patent Document 3] JP 2019-32767 A Summary of the Invention [Problem to be solved by the invention]

[0004] However, in the technology described in the above patent document, the acquired brain activity data is subjected to statistical processing, brain activity data is acquired from an arbitrary brain region, and the brain region to be analyzed is not determined, so the brain activity cannot be evaluated with high accuracy. Therefore, the present invention relates to a brain activity evaluation device that can evaluate brain activity with high accuracy. [Means for solving the problem]

[0005] The present invention relates to a brain activity evaluation device including: an acquisition unit that acquires an f-MRI image for at least one stimulus; a correlation derivation unit that statistically processes brain activity in the acquired f-MRI image for each voxel and derives a correlation between voxel data and a degree of value represented by the voxel data; a region of interest determination unit that selects positively correlated voxels and negatively correlated voxels to determine a region of interest, which is a brain region to be analyzed among any brain region; and a brain activity evaluation model generation unit that extracts brain activity data for the stimulus in the determined region of interest, and generates a brain activity evaluation model to evaluate the brain activity for the stimulus based on the extracted brain activity data.

[0006] The present invention also relates to a brain activity evaluation method including: an acquisition step of acquiring an f-MRI image for at least one stimulus; a correlation deriving step of statistically processing brain activity in the acquired f-MRI image for each voxel to derive a correlation between voxel data and a degree of value represented by the voxel data; a region of interest determination step of selecting positively correlated voxels and negatively correlated voxels to determine a region of interest, which is a brain region to be analyzed among any brain region; and a brain activity evaluation model generation step of extracting brain activity data for the stimulus of the determined region of interest, and generating a brain activity evaluation model for evaluating the brain activity for the stimulus based on the extracted brain activity data.

[0007] Furthermore, the present invention relates to a brain activity evaluation program that causes a computer to execute the following steps: an acquisition step of acquiring an f-MRI image for at least one stimulus; a correlation deriving step of statistically processing brain activity in the acquired f-MRI image for each voxel and deriving a correlation between voxel data and a degree of value represented by the voxel data; a region of interest determination step of selecting positively correlated voxels and negatively correlated voxels to determine a region of interest, which is a brain region to be analyzed among any brain region; and a brain activity evaluation model generation step of extracting brain activity data for the stimulus in the determined region of interest, and generating a brain activity evaluation model to evaluate the brain activity for the stimulus based on the extracted brain activity data. Effect of the Invention

[0008] According to the brain activity evaluation device of the present invention, brain activity can be evaluated with high accuracy. [Brief description of the drawings]

[0009] [Figure 1A] FIG. 1 is a diagram for explaining an overview of brain activity evaluation by the brain activity evaluation apparatus according to a first embodiment of the present invention. [Figure 1B] FIG. 1 is a diagram for explaining an experimental procedure for brain activity evaluation using the brain activity evaluation apparatus according to the first embodiment of the present invention. [Diagram 2] 1 is a block diagram for explaining the configuration of a brain activity evaluation apparatus according to a first embodiment of the present invention. [Diagram 3] FIG. 2 is a diagram showing an example of a model generation table included in the brain activity evaluation apparatus according to the first embodiment of the present invention. [Figure 4] 1 is a block diagram for explaining a hardware configuration of a brain activity evaluation apparatus according to a first embodiment of the present invention. [Diagram 5] 3 is a flowchart illustrating a processing procedure of the brain activity evaluation apparatus according to the first embodiment of the present invention. [Figure 6A] FIG. 11 is a diagram for explaining an overview of brain activity evaluation by a brain activity evaluation apparatus according to a second embodiment of the present invention. [Figure 6B] FIG. 11 is a diagram for explaining an overview of brain activity evaluation by a brain activity evaluation apparatus according to a second embodiment of the present invention. [Figure 7] FIG. 4 is a block diagram for explaining the configuration of a brain activity evaluation apparatus according to a second embodiment of the present invention. [Figure 8] FIG. 11 is a diagram showing an example of an estimation table included in the brain activity evaluation apparatus according to the second embodiment of the present invention. [Figure 9] FIG. 11 is a block diagram illustrating a hardware configuration of a brain activity evaluation apparatus according to a second embodiment of the present invention. [Figure 10] 10 is a flowchart illustrating a processing procedure of a brain activity evaluation apparatus according to the second embodiment of the present invention. [Figure 11A] FIG. 11 is a diagram for explaining an overview of brain activity evaluation by the brain activity evaluation apparatus according to the third embodiment of the present invention. [Figure 11B] FIG. 11 is a diagram for explaining an overview of brain activity evaluation by the brain activity evaluation apparatus according to the third embodiment of the present invention. [Figure 12] FIG. 11 is a block diagram for explaining the configuration of a brain activity evaluation apparatus according to a third embodiment of the present invention. [Figure 13] FIG. 13 is a diagram showing an example of a comparison table included in the brain activity evaluation apparatus according to the third embodiment of the present invention. [Figure 14] FIG. 11 is a block diagram illustrating a hardware configuration of a brain activity evaluation apparatus according to a third embodiment of the present invention. [Figure 15] 13 is a flowchart illustrating a processing procedure of a brain activity evaluation apparatus according to the third embodiment of the present invention. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0010] Hereinafter, the embodiments of the present invention will be described in detail with reference to the drawings. However, the configurations, numerical values, process flows, functional elements, etc. described in the following embodiments are merely examples, and are free to be modified or changed, and are not intended to limit the technical scope of the present invention to the following description.

[0011] [First embodiment] A brain activity evaluation device 100 according to a first embodiment of the present invention will be described with reference to Figures 1A to 5. The brain activity evaluation device 100 is a device that uses f-MRI images to generate a brain activity evaluation model that evaluates brain activity in response to a stimulus.

[0012] First, referring to FIG. 1A, an outline of the operation of the brain activity evaluation device 100 will be described. The brain activity evaluation device 100 is a device that generates a brain activity evaluation model that evaluates the brain activity of the subject 110 in response to a stimulus from the state of the brain activity of the subject 110 to which a stimulus has been given. As shown in FIG. 1, first, when the brain activity evaluation device 100 shows, for example, a video related to a product as a stimulus to the brain of the subject 110, it measures the brain activity of the subject 110 while the stimulus is being given, that is, while the video is being shown, using an f-MRI (functional Magnetic Resonance Imaging) device 140 to obtain an f-MRI image of the subject 110 in response to the stimulus. Then, the brain activity evaluation device 100 decodes the obtained f-MRI image and generates a brain activity evaluation model 130 that evaluates the brain activity of the subject 110. When the measured brain activity of the subject 110 is input to the brain activity evaluation model 130, it is possible to predict, for example, the degree of value, such as preference or purchase intention, for a product shown in the video to the subject 110. Here, the value includes judgments that continue for a relatively long time, such as evaluations of a person's preferences. Value also includes judgments that continue for a relatively short time and instantaneous judgments that are thought at the time. The imaging conditions for the f-MRI device 140 are a magnetic field of 1.0 to 3.0 T, an imaging time required to capture one f-MRI image of 0.5 to 3 seconds, a slice thickness of 1 to 2 mm, and a slice interval of 0 to 1 mm.

[0013] Here, the stimulus is, for example, a video relating to a product (hereinafter also referred to as a "product video") presented to the subject 110. The product video 120 includes, for example, a commercial film (CF) for advertising the product, a product image video, a video showing the manufacturing process, and the like, but is not limited to these. The subject 110 may be a group or person with some characteristic, or a randomly selected group or person.

[0014] Next, a method for generating a model for evaluating brain activity of the subject 110 using the brain activity evaluation apparatus 100 according to this embodiment will be described with reference to FIG. 1B.

[0015] As shown in Fig. 1B, a brain activity evaluation model for evaluating brain activity of subject 110 was generated by a method in which subject 110 is presented with a product video 120 and asked to respond to the extent to which he or she would like to buy the product introduced in the product video 120. First, a 15-second product video 120 is presented to subject 110, and while viewing product video 120 (viewing), subject 110 presses a button at the timing when he or she decides "how much he or she would like to buy it" (see "(1) Value Decision" in Fig. 1B).

[0016] Next, subject 110 inputs the degree of purchase intention, which is the degree of "how much he or she wants to buy the product" (see "(2) Evaluation of product value" in FIG. 1B). Subject 110 inputs the degree of purchase intention by moving a slider bar displayed on the screen. The degree of purchase intention is judged on a 101-point scale from "0=do not purchase" to "100=purchase", but the input of the degree of purchase intention is not limited to this, and various methods can be used, such as a 5-point scale from 1 to 5 or a rank scale from A to E. Note that the numbers 0 and 100 are not visible to subject 110, and subject 110 can intuitively move the slider bar without being conscious of the numbers.

[0017] The subject 110 watches, for example, a total of 110 such product videos 120. Then, for each of the product videos 120, the state of brain activity of the subject 110 is imaged using the f-MRI device 140, and the captured f-MRI images are analyzed using the brain activity evaluation device 100 to generate a model for predicting the degree of the subject 110's intention to purchase the product, etc.

[0018] The functional configuration of the brain activity evaluation device 100 will be described with reference to Fig. 2. As shown in Fig. 2, the brain activity evaluation device 100 includes an acquisition unit 201, a correlation derivation unit 202, a region of interest determination unit 203, and a brain activity evaluation model generation unit 204. The acquisition unit 201 acquires an f-MRI image 210 for at least one stimulus, but it is preferable that the number of stimuli is about 100 or more in order to improve the generalization of the brain activity evaluation device 100. In this embodiment, for example, 110 product videos 120 are viewed, so that 110 types of f-MRI images 210 are acquired and analyzed for each subject 110.

[0019] The correlation derivation unit 202 statistically processes the brain activity in the acquired f-MRI image 210 for each voxel, and derives the correlation between the voxel data (also called "brain activity data") and the degree of purchase intention (degree of value) represented by the voxel data. Here, a voxel refers to a region obtained by dividing the whole brain of the subject 110 into a cube with sides of 1 to 3 mm, and in this embodiment, the size of a voxel is 2 [mm 3 However, the size of the voxel is not limited to this and can be up to 1 mm 3 ], but 3 [mm 3 ] or more. The correlation is derived by calculating the degree of correlation (β value) between the degree of purchase intention of the subject 110 for the given stimulus and the voxel data based on the amount of brain activity, and performing a T-test on the β value for each voxel to calculate the T-statistic (brain activity data) of the voxel. That is, the larger the absolute value of the T-statistic of the voxel, the further it is from the average value of the distribution of the β values ​​of the voxel, and the higher the degree of involvement in the given stimulus.

[0020] The region of interest determination unit 203 selects positively correlated voxels having a positive correlation and negatively correlated voxels having a negative correlation to determine a region of interest that is a brain region to be analyzed from an arbitrary brain region. Here, the positively correlated voxels are voxels having a positive T statistic, and the negatively correlated voxels are voxels having a negative T statistic.

[0021] The region of interest determination unit 203 determines, as a region of interest, a region having a T statistic in the top 500 out of brain regions (voxels) having a positive correlation, i.e., positive brain activity. Similarly, the region of interest determination unit 203 determines, as a region of interest, a region having a T statistic in the bottom 500 out of brain regions (voxels) having a negative correlation, i.e., negative brain activity. In other words, the region of interest is a region having an absolute value of the T statistic in the top 500 or more out of positive correlation voxels and negative correlation voxels.

[0022] In this way, the region of interest determining unit 203 determines 1,000 brain regions (voxels) having both positive and negative correlations as the region of interest. Note that, although the top 500 and bottom 500 of the T statistics are determined as the region of interest here, the brain region to be determined as the region of interest is not limited to the top 500 and bottom 500, and may be any region above 500.

[0023] The brain activity evaluation model generation unit 204 extracts brain activity data in response to the stimulation of the determined region of interest, and generates a brain activity evaluation model for evaluating brain activity in response to the stimulation based on the extracted brain activity data. Note that the brain activity evaluation model generation unit 204 extracts T statistics or β values ​​as the brain activity data.

[0024] Next, the brain activity evaluation model generation unit 204 performs a principal component analysis (PCA) on the extracted brain activity data, and performs dimensional compression (calculation of principal component scores) of the extracted brain activity data. In this embodiment, the brain activity data of the top 500 voxels and the bottom 500 voxels having positive and negative correlations with the 110 stimuli (a total of 1,000 brain activity data), that is, the 1,000-dimensional brain activity data, is compressed to, for example, about 100 dimensions and used for generating the brain activity evaluation model. The number of dimensions used for generating the brain activity evaluation model may be any number of dimensions as long as it is 10 or more, and is preferably between 10 and 30 dimensions, and more preferably 10 dimensions. It is possible to perform or not to perform dimensional compression, but it is preferable to perform it because it improves the estimation accuracy.

[0025] In addition, the method of dimension reduction is not limited to the principal component analysis, and may be, for example, an autoencoder (AE) or a variational autoencoder (VAE). For example, the principal component analysis and the AE may be combined, or the principal component analysis or the AE may be used alone to reduce the dimension. The method of dimension reduction is not limited to the method shown here, and various methods may be used, or a combination of these methods may be used.

[0026] Then, the brain activity evaluation model generation unit 204 generates the brain activity evaluation model 130 from the dimension-reduced brain activity data (principal component scores) using a general linear model (GLM). Note that the brain activity evaluation model generation unit 204 may generate the brain activity evaluation model 130 using a support vector machine (SVM) instead of using the general linear model.

[0027] Then, by inputting any brain activity data into the obtained brain activity evaluation model 130, it is possible to predict the purchase intention, etc. of the subject 110 without having to ask the subject 110 for answers regarding purchase intention, etc.

[0028] 3 shows an example of a model generation table 301 held by the brain activity evaluation apparatus 100. The model generation table 301 stores an f-MRI image 312, a region of interest 313, and brain activity data 314 in association with a subject ID 311. The subject ID 311 is an identifier for identifying the subject 110. The region of interest 313 is a brain region (voxel) to be analyzed. The brain activity data 314 is brain activity data (T statistics or β value) of the region of interest. The brain activity evaluation apparatus 100 then refers to the model generation table 301 to generate a brain activity evaluation model 130.

[0029] Next, the hardware configuration of the brain activity evaluation device 100 will be described with reference to FIG. 4. The CPU (Central Processing Unit) 410 is a processor for arithmetic control, and executes a program to realize each functional configuration of the brain activity evaluation device 100 in FIG. 2. The CPU 410 may have multiple processors and execute different programs, modules, tasks, threads, etc. in parallel. The ROM (Read Only Memory) 420 stores fixed data such as initial data and programs, and other programs. The network interface 430 communicates with other devices via a network. The CPU 410 is not limited to one, and may be multiple CPUs or may include a GPU (Graphics Processing Unit) for image processing. The network interface 430 preferably has a CPU independent of the CPU 410 and writes or reads transmitted / received data in an area of ​​the RAM (Random Access Memory) 440. It is also preferable to provide a DMAC (Direct Memory Access Controller) for transferring data between the RAM 440 and the storage 450 (not shown). Furthermore, the CPU 410 processes the data upon recognizing that the data has been received or transferred to the RAM 440. The CPU 410 also prepares the processing results in the RAM 440, and leaves the subsequent transmission or transfer up to the network interface 430 or DMAC.

[0030] The RAM 440 is a random access memory used by the CPU 410 as a work area for temporary storage. A storage area for storing data necessary for implementing this embodiment is secured in the RAM 440. The f-MRI image data 441 is data related to the f-MRI image 210 acquired by the acquisition unit 201. The region of interest data 442 is data related to the region of interest determined by the region of interest determination unit 203. The brain activity data 443 is the T-statistic or β value of the voxels included in the region of interest.

[0031] The transmitted / received data 444 is data that is transmitted and received via the network interface 430. The RAM 440 also has an application execution area 445 for executing various application modules.

[0032] The storage 450 stores a database, various parameters, or the following data or programs required to realize this embodiment. The storage 450 stores a model generation table 301. The model generation table 301 is a table that manages the relationship between the subject ID 311 and the f-MRI image 312, etc., shown in FIG. 3.

[0033] The storage 450 further stores an acquisition module 451, a correlation derivation module 452, a region of interest determination module 453, and a brain activity evaluation model generation module 454. The acquisition module 451 is a module that acquires the f-MRI image 210. The correlation derivation module 452 is a module that derives a correlation between voxel data and the degree of purchase intention. The region of interest determination module 453 is a module that determines a region of interest that is a brain region to be analyzed. The brain activity evaluation model generation module 454 is a module that generates a brain activity evaluation model that evaluates brain activity in response to a stimulus based on brain activity data of the region of interest. These modules 451 to 454 are read by the CPU 410 into the application execution area 447 of the RAM 440 and executed. The control program 455 is a program for controlling the entire brain activity evaluation device 100.

[0034] The input / output interface 460 interfaces input / output data with input / output devices. A display unit 461 and an operation unit 462 are connected to the input / output interface 460. A storage medium 464 may be further connected to the input / output interface 460. A speaker 463 serving as an audio output unit, a microphone (not shown) serving as an audio input unit, or a GPS position determination unit may be further connected. Note that the RAM 440 and the storage 450 shown in FIG. 4 do not show programs and data related to the general-purpose functions of the brain activity evaluation device 100 or other feasible functions.

[0035] The processing procedure of the brain activity evaluation device 100 will be described with reference to the flowchart shown in Fig. 5. This flowchart is executed by the CPU 410 in Fig. 4 using the RAM 440, and realizes each functional configuration of the brain activity evaluation device 100 in Fig. 2. In step S501, the brain activity evaluation device 100 presents the product video 120 to the subject 110. In step S503, the acquisition unit 201 acquires an f-MRI image 210 of the subject 110 captured while viewing the product video 120. In step S505, the correlation derivation unit 202 statistically processes the brain activity in the acquired f-MRI image 210 for each voxel, and derives the strength of the correlation between the voxel data and the degree of purchase intention (degree of value) represented by the voxel data. Specifically, the correlation derivation unit 202 calculates a β value from the slope of a line segment that represents the relationship between two variables when the brain activity amount for each voxel is the objective variable and the degree of purchase intention is the explanatory variable, and performs a T-test on the calculated β value to calculate the T-statistics of the voxel, thereby deriving the strength of the correlation. In step S507, the brain activity evaluation device 100 judges whether or not correlations have been derived for all voxels. If it is determined that correlations have not been derived for all voxels (NO in step S507), the brain activity evaluation device 100 returns to step S505. If it is determined that correlations have been derived for all voxels (YES in step S507), the brain activity evaluation device 100 proceeds to step S509.

[0036] In step S509, the region of interest determining unit 203 selects voxels having a positive correlation and voxels having a negative correlation. In step S511, the region of interest determining unit 203 ranks the selected positively correlated voxels and negatively correlated voxels based on the derived T-statistics, and determines voxels within the top 500 and the bottom 500 from the ranking of the voxels as the region of interest. In step S513, the brain activity evaluation model generating unit 204 extracts brain activity data of the voxels in the region of interest. In step S515, the brain activity evaluation model generating unit 204 generates a brain activity evaluation model 130 that evaluates the brain activity of the subject 110 in response to a stimulus based on the extracted brain activity data.

[0037] In the above description, an example has been described in which a video related to a product is presented as a stimulus to the subject 110, but the stimulus to be given to the subject 110 is not limited to the product video 120. For example, even if a product image (still image of the product), a fine art painting, a photograph, a tactile stimulus, an olfactory stimulus, a gustatory stimulus, an auditory stimulus, or the like is used, it is possible to generate a brain activity evaluation model 130 for evaluating the brain activity of the subject 110 from the brain activity data of the subject 110.

[0038] According to this embodiment, voxels having both positive and negative correlations are selected to generate a brain activity evaluation model, so that a highly accurate brain activity evaluation model can be generated. In addition, since a highly accurate brain activity evaluation model is generated, brain activity can be evaluated with high accuracy, and a subject's intention to purchase a product can be predicted with high accuracy.

[0039] [Second embodiment] Next, a brain activity evaluation device 600 according to a second embodiment of the present invention will be described with reference to Figs. 6A to 10. The brain activity evaluation device 600 according to this embodiment differs from the first embodiment in that it includes an estimation unit 701. Other configurations and operations are similar to those of the first embodiment, so the same configurations and operations are denoted by the same reference numerals and detailed description thereof will be omitted.

[0040] As shown in FIG. 6A and FIG. 6B, first, when the brain activity evaluation device 600 shows, for example, a product image 620 as a stimulus to the brain of the subject 110, the brain activity of the subject 110 is measured using the f-MRI device 140 while the stimulus is being given, that is, while the image is being shown, and an f-MRI image of the subject 110 in response to the stimulus is obtained. Then, the brain activity evaluation device 600 analyzes the obtained f-MRI image and generates a brain activity evaluation model 630 for evaluating the brain activity of the subject 110. Then, the brain activity evaluation device 600 estimates what kind of impression the subject 110 has about the new stimulus or the product shown in the product image 620 based on the generated brain activity evaluation model 630. Here, the new stimulus includes, for example, a new product image that is a product other than the product shown in the product image 620, but is not limited thereto.

[0041] For example, when a new stimulus or product image 620 is presented to the subject 110, the brain activity evaluation device 600 estimates, based on the brain activity of the subject 110, from the brain activity evaluation model 630 generated using the product image 620, at least one of the following for the subject 110 regarding the product shown in the new stimulus or product image 620: the degree to which the subject 110 wanted to purchase the product (degree of purchase intention), whether the subject had a good impression of the product (likeability), and whether the subject thought the product was an excellent product (performance evaluation), and displays these degrees on the estimation graph 640. Note that the contents to be estimated may be the degree of interest in the product, the degree of likes and dislikes for the product image 620, in addition to the degree of purchase intention, the likeability, and the performance evaluation.

[0042] The brain activity evaluation device 600 also estimates at what timing the subject 110 had the above-mentioned impressions, and displays them on the timing graph 650. That is, for example, when the new stimulus or product image 620 is a 15-second image, the timing of the subject 110's impressions on the product are displayed as decision-making timings on the timing graph 650 along with the estimated values ​​of the degree of each for each second, regarding the timing of the 15 seconds at which the subject 110 had the above-mentioned impressions. In this way, by displaying the timing graph 650, it is possible to compare the estimated values ​​for each second, and the scene showing the highest estimated value can be confirmed. In addition, since the timing of some decision-making is recorded, it is possible to search for a scene in which the subject 110 showed high interest, even though no decision-making was made. Note that in the timing graph 650, the estimated values ​​are shown every second as the time interval, but the time interval is not limited to this, and it is preferable to show the estimated values ​​at intervals of, for example, 0.5 to 3 seconds. Incidentally, since the time interval is determined depending on the imaging time required to capture one f-MRI image, it is preferable to set the time interval to be equal to or longer than the imaging time.

[0043] Next, the functional configuration of the brain activity evaluation device 600 will be described with reference to FIG. 7. The brain activity evaluation device 600 has an estimation unit 701. The estimation unit 701 estimates at least one of the degree of purchase intention, favorability rating, and performance evaluation for a product using the generated brain activity evaluation model 630. Here, the purchase intention for a product is an index relating to whether or not a person feels motivated to purchase the product after viewing the product video 620. Furthermore, the favorability rating is an index relating to whether or not a person has a good impression or image of the product after viewing the product video 620. Furthermore, the performance evaluation is an index relating to whether or not the performance of the product is considered to be excellent based on knowledge obtained from the product video 620.

[0044] The brain activity evaluation device 600 may have a display control unit that visualizes and displays the estimated values ​​of the degree of purchase intention, the favorability rating, and the performance evaluation. For example, the display control unit of the brain activity evaluation device 600 visualizes and displays each estimated value in the form of an estimated graph 640 or a timing graph 650. Note that the visualization form is not limited to the form shown here, and visualization can be performed in various forms.

[0045] FIG. 8 shows an example of an estimation table 801 held by the brain activity evaluation device 600. The estimation table 801 stores an evaluation timing 811 and an evaluation score 812 in association with the subject ID 311. The evaluation timing 811 is the timing at which the subject 110 evaluates the product shown in the product image 620, and is expressed, for example, as the number of seconds from the start of presentation of the product image 620, for example, as "xx seconds after the start of presentation of the product image 620". The evaluation score 812 is the evaluation given by the subject 110 on the product itself, and is expressed, for example, as a score. The brain activity evaluation device 600 then refers to the estimation table 801 to estimate at least one of the degree of the subject 110's purchase intention, favorability, and performance evaluation of the product.

[0046] Next, the hardware configuration of the brain activity evaluation device 600 will be described with reference to FIG. 9. The RAM 940 is a random access memory used by the CPU 410 as a work area for temporary storage. A storage area for storing data required for implementing this embodiment is secured in the RAM 940. The timing data 941 is data related to the timing at which the subject 110 evaluates the product. The evaluation data 942 is data related to the evaluation itself of the product by the subject 110. The estimated data 943 is data of the estimation result such as the degree of the subject 110's intention to purchase the product, favorability rating, and performance evaluation.

[0047] The storage 950 stores a database, various parameters, or the following data or programs required to realize this embodiment. The storage 950 stores an estimation table 801. The estimation table 801 is a table that manages the relationship between the subject ID 311 and the evaluation timing 811 and the evaluation score 812 shown in FIG.

[0048] The storage 950 further stores an estimation module 951. The estimation module 951 is a module that estimates at least one of the degree of purchase intention, the likeability, and the performance evaluation of the subject 110 with respect to a product, using the generated brain activity evaluation model 630.

[0049] Fig. 10 is a flowchart explaining the processing procedure of the brain activity evaluation device 600. This flowchart is executed by the CPU 410 in Fig. 9 using the RAM 940, and realizes each functional configuration of the brain activity evaluation device 600 in Fig. 7. In step S1001, the estimation unit 701 estimates at least one of the degree of purchase intention, favorability rating, and performance evaluation of the subject 110 for the product, using the generated brain activity evaluation model. In step S1003, the brain activity evaluation device 600 displays the estimation result.

[0050] According to this embodiment, the generated brain activity evaluation model can be used to estimate how the subject feels about the product in the product video that the subject has viewed.

[0051] [Third embodiment] Next, a brain activity evaluation device 1100 according to a third embodiment of the present invention will be described with reference to Figs. 11A to 15. The brain activity evaluation device 1100 according to this embodiment differs from the first and second embodiments in that it includes a comparison unit 1201. Other configurations and operations are similar to those of the first and second embodiments, so the same configurations and operations are denoted by the same reference numerals and detailed descriptions thereof will be omitted.

[0052] As shown in FIG. 11A and FIG. 11B, the brain activity evaluation device 1100 generates a brain activity evaluation model 1130 that evaluates brain activity for each of the product images 1120 when multiple product images 1120 are presented to the subject 110 as stimuli, and compares the subject 110's impressions of multiple new stimuli or each product shown in the product images 1120 using the generated brain activity evaluation model 1130. For example, the brain activity evaluation device 1100 displays and compares estimated values ​​of the degree of purchase intention for each product using an estimation graph 1140, or displays and compares the timing of decision-making using a timing graph 1150. Note that the subject 110 is not limited to a single subject, and may be multiple. When there are multiple subjects, the estimated values ​​for each subject may be compared between the subjects.

[0053] 12, the brain activity evaluation device 1100 has a comparison unit 1201. The acquisition unit 201 acquires f-MRI images 1210 for multiple product videos 1120. That is, there is a corresponding f-MRI image 1210 for each product video, and the acquisition unit 201 acquires a number of f-MRI images 1210 corresponding to the number of product videos. Also, for example, when there are multiple subjects 110, the acquisition unit 201 acquires a number of f-MRI images 1210 corresponding to the number of subjects 110.

[0054] Then, the brain activity evaluation model generation unit 204 generates a brain activity evaluation model 1130 for each of the multiple product videos 1120. That is, the number of brain activity evaluation models 1130 generated corresponds to the number of acquired f-MRI images 1210.

[0055] The comparison unit 1201 compares at least one of the degree of purchase intention, favorability rating, and performance evaluation of a plurality of products estimated by the estimation unit 701. For example, the comparison unit 1201 derives an estimate of the degree of purchase intention of the subject 110 for each product, and compares which product has the highest or lowest degree of purchase intention. That is, the comparison result of the comparison unit 1201 makes it possible to know various tendencies and preferences of the subject 110, such as the product image 1120 with the highest degree of purchase intention and, conversely, the product image 1120 with the lowest degree of purchase intention. Similarly, various tendencies and preferences of the subject 110 can be known with respect to favorability rating and performance evaluation.

[0056] Furthermore, the comparison unit 1201 may compare which product the subject 110 decided to purchase the earliest or the latest for each product. This makes it possible to know which product image 1120 is appealing to the subject 110, or which image encourages the subject 110 to examine and convince the subject 110 before making a decision. The brain activity evaluation device 1100 may include an output control unit that outputs the comparison result to an external display or the like.

[0057] 13 shows an example of a comparison table 1301 held by the brain activity evaluation device 1100. The comparison table 1301 stores a stimulus ID 1311 and the like in association with a subject ID 311. The stimulus ID 1311 is an identifier for identifying each of a plurality of product images 1120. The brain activity evaluation device 1100 refers to the comparison table 1301 to generate a brain activity evaluation model 1130 and compares at least any of the estimated results of the degree of purchase intention, the likeability rating, and the performance evaluation of a plurality of products.

[0058] The hardware configuration of the brain activity evaluation device 1100 will be described with reference to Fig. 14. The RAM 1440 is a random access memory used by the CPU 410 as a work area for temporary storage. A storage area for storing data necessary for implementing this embodiment is secured in the RAM 1440. The comparison result data 1441 is data indicating the result of the comparison of opinions, etc., for each product by the comparison unit 1201.

[0059] The storage 1450 stores a database, various parameters, or the following data or programs required to realize this embodiment. The storage 1450 stores a comparison table 1301. The comparison table 1301 is a table that manages the relationship between the stimulus ID 1311 and the subject ID 311, etc., shown in FIG. 13.

[0060] The storage 1450 further stores a comparison module 1451. The comparison module 1451 is a module that compares at least one of the degree of purchase intention, the likeability rating, and the performance evaluation of a plurality of estimated products.

[0061] FIG. 15 is a flowchart for explaining the processing procedure of the brain activity evaluation device 1100. This flowchart is executed by the CPU 410 in FIG. 14 using the RAM 1440, and each functional configuration of the brain activity evaluation device 1100 in FIG. 12 is realized. In step S1501, the brain activity evaluation device 1100 presents a plurality of product images 1120 to the subject 110. In step S1503, the acquisition unit 201 acquires an f-MRI image 1210 of the subject 110 viewing the plurality of product images 1120. In step S1505, the comparison unit 1201 compares at least one of the degree of purchase intention, favorability rating, and performance evaluation estimated for the products in the plurality of product images 1120. In step S1507, the display control unit of the brain activity evaluation device 1100 outputs the comparison result to a display device such as a display and displays it.

[0062] According to this embodiment, since the estimation results are compared, it is possible to know which of a plurality of product videos is most appealing to the subject. This makes it possible to create more appealing product videos, and ultimately to develop more appealing products.

[0063] [Other embodiments] Although the present invention has been described above with reference to the embodiments, the present invention is not limited to the above-mentioned embodiments. Various modifications that can be understood by those skilled in the art can be made to the configuration and details of the present invention within the scope of the present invention. In addition, systems or devices that combine the separate features included in each embodiment in any way are also included in the scope of the present invention.

[0064] The present invention may be applied to a system consisting of multiple devices, or to a single device. Furthermore, the present invention is also applicable to a case where an information processing program for realizing the functions of the embodiments is supplied directly or remotely to a system or device. Therefore, the scope of the present invention also includes a program installed on a computer to realize the functions of the present invention by a computer, a medium storing the program, and a WWW (World Wide Web) server for downloading the program. In particular, the scope of the present invention includes at least a non-transitory computer readable medium storing a program for causing a computer to execute the processing steps included in the above-mentioned embodiments. [Explanation of symbols]

[0065] 100 Brain activity evaluation device 110 subjects 120 Product Video 130 Brain Activity Evaluation Model 201 Acquisition Department 202 Correlation Derivation Part 203 Region of interest determination unit 204 Brain Activity Evaluation Model Generation Unit 210 f-MRI images 600 Brain activity evaluation device 620 Product Video 630 Brain Activity Evaluation Model 701 Estimation section 1100 Brain activity evaluation device 1120 Product video 1130 Brain activity evaluation model 1201 Comparison Section

Claims

1. an acquisition unit that acquires an f-MRI image for at least one stimulus; A correlation deriving unit that statistically processes brain activity in the acquired f-MRI image for each voxel and derives a correlation between voxel data and a degree of value represented by the voxel data; a region of interest determination unit that selects positive correlation voxels having a positive correlation and negative correlation voxels having a negative correlation to determine a region of interest that is a brain region to be analyzed from among any brain region; a brain activity evaluation model generation unit that extracts brain activity data in response to the stimulation of the determined region of interest, and generates a brain activity evaluation model for evaluating the brain activity in response to the stimulation based on the extracted brain activity data; Equipped with the stimuli include images relating to the product; The brain activity includes a degree of intention to purchase the product, the acquiring unit acquires, while viewing a video relating to a plurality of the products, an f-MRI image at the time when the purchase of the products is decided and an f-MRI image during an answer to a degree of the purchase intention for the products using a multi-level evaluation; the brain activity evaluation model generation unit generates the brain activity evaluation model for a video related to the plurality of products; an estimation unit that estimates a degree of purchase intention of the product or a new product by using the generated brain activity evaluation model; a comparison unit that compares the degree of purchase intention of the plurality of products estimated by the estimation unit; The brain activity evaluation device further comprises:

2. the correlation deriving unit derives a T-statistic by performing a T-test on the β value of the voxel as the statistical processing; 2. The brain activity evaluation device according to claim 1, wherein the region of interest determination unit determines the region of interest by selecting voxels having a positive T statistic as the positive correlation voxels and selecting voxels having a negative T statistic as the negative correlation voxels.

3. 3. The brain activity evaluation device according to claim 1, wherein the brain activity evaluation model generation unit performs a principal component analysis on the extracted brain activity data to generate the brain activity evaluation model.

4. 3. The brain activity evaluation device according to claim 2, wherein the brain activity evaluation model generation unit extracts the T statistics or the β values ​​as the brain activity data, performs principal component analysis on the extracted T statistics or the β values, and generates the brain activity evaluation model.

5. 5. The brain activity evaluation device according to claim 1, wherein the brain activity evaluation model generation unit generates the brain activity evaluation model using at least one of regression analysis and machine learning.

6. The brain activity evaluation device according to claim 2 , wherein the region of interest determination unit selects voxels that are ranked in the top 500 or more positions in absolute values ​​of the T statistics as the positive correlation voxels and the negative correlation voxels.

7. acquiring an f-MRI image for at least one stimulus; A correlation deriving step of statistically processing brain activity in the acquired f-MRI image for each voxel and deriving a correlation between voxel data and a degree of value represented by the voxel data; a region of interest determination step of selecting positive correlation voxels having a positive correlation and negative correlation voxels having a negative correlation to determine a region of interest that is a brain region to be analyzed from among any brain region; a brain activity evaluation model generation step of extracting brain activity data for the stimulation of the determined region of interest, and generating a brain activity evaluation model for evaluating the brain activity for the stimulation based on the extracted brain activity data; Including, the stimuli include images relating to the product; The brain activity includes a degree of intention to purchase the product, In the acquiring step, while watching a video relating to a plurality of the products, an f-MRI image at the time when the purchase of the product was decided and an f-MRI image during the answering of the degree of the purchase intention for the product using a multi-level evaluation are acquired; In the brain activity evaluation model generation step, the brain activity evaluation model is generated for images related to the plurality of products; an estimation step of estimating a degree of purchase intention of the product or a new product using the generated brain activity evaluation model; a comparison step of comparing the degree of purchase intention of the plurality of products estimated in the estimation step; The method for evaluating brain activity further comprises:

8. acquiring an f-MRI image for at least one stimulus; A correlation deriving step of statistically processing brain activity in the acquired f-MRI image for each voxel and deriving a correlation between voxel data and a degree of value represented by the voxel data; a region of interest determination step of selecting positive correlation voxels having a positive correlation and negative correlation voxels having a negative correlation to determine a region of interest that is a brain region to be analyzed from among any brain region; a brain activity evaluation model generation step of extracting brain activity data for the stimulation of the determined region of interest, and generating a brain activity evaluation model for evaluating the brain activity for the stimulation based on the extracted brain activity data; Run the following on your computer: the stimuli include images relating to the product; The brain activity includes a degree of intention to purchase the product, In the acquiring step, while watching a video relating to a plurality of the products, an f-MRI image at the time when the purchase of the product was decided and an f-MRI image during the answering of the degree of the purchase intention for the product using a multi-level evaluation are acquired; In the brain activity evaluation model generation step, the brain activity evaluation model is generated for images related to the plurality of products; an estimation step of estimating at least one of a degree of purchase intention, a favorability rating, and a performance evaluation of the product or a new product using the generated brain activity evaluation model; a comparison step of comparing the degree of purchase intention of the plurality of products estimated in the estimation step; A brain activity evaluation program that further has a computer run this.

Images