JP2022096561A - Behavior estimation system, measurement point selection device, behavior estimation device, method for controlling information processing device, method for controlling behavior estimation device, and program - Google Patents

Abstract

To enable estimation about a behavior with a method having a low processing load by supporting the reduction of a processing load of estimation about the behavior.SOLUTION: Brain wave data of an object person to be measured performing a predetermined kind of behaviors is measured at prescribed N (N is an integer N>2) measurement points of the head of the object person and M (M is a natural number) reference measurement points. At least a pair of points among the N measurement points is selected as a representative point combination on the basis of a result of a functional connection analysis about the brain wave data and a result of behaviors presented by the object person to be measured at the pair of measurement points selected from among the N measurement points, and is used for estimation processing.SELECTED DRAWING: Figure 1

Description

Patent Law Article 30, Paragraph 2 Application Applicable Presented at Keio University Graduate School of Media Design Doctoral Dissertation Hearing on December 18, 1st year

The present invention relates to a behavior estimation system, a measurement point selection device, a behavior estimation device, a control method of an information processing device, a control method of a behavior estimation device, and a program.

In the evaluation of video content, it is an important point of view whether or not the scene being played is memorable to the viewer and leaves an impression on the viewer. For example, conventionally, there has been known a technique for acquiring marketing information useful for improving the quality of program broadcast data on the broadcasting station side by using data broadcasting (Patent Document 1).

Japanese Unexamined Patent Publication No. 2014-217024

However, the above-mentioned conventional technique relies exclusively on subjective evaluation, and it is also difficult to estimate whether or not there is a memorable scene during or immediately after the video is being broadcast.

The present invention has been made in view of the above circumstances, and is a behavior estimation system and measurement point selection that supports reduction of the processing load of estimation regarding behavior and enables estimation of behavior by a method with less processing load. One of the purposes thereof is to provide a device, a behavior estimation device, a control method of an information processing device, a control method of the behavior estimation device, and a program.

One aspect of the present invention that solves the problems of the above-mentioned conventional example is to collect brain wave data of a measurement target person performing a predetermined type of behavior at a predetermined N point (N is N>) of the subject's head. Functions related to the brain wave data at a measurement means for measuring at a measurement point (an integer of 2), a reference measurement point at M point (M is a natural number), and a pair of measurement points selected from the N point measurement points. The representative includes a target coupling analysis and a selection means for selecting at least a pair of measurement points of the N points as a representative point set based on the result of the action presented by the measurement target person. The information that identifies the point set is at n points (n is an integer such that N> n ≧ 2) including the representative point set specified by the information of the measurement target person who is performing the predetermined type of action. It is decided to acquire the brain wave data and use it for the process of estimating the result of the behavior of the measurement target person.

According to the present invention, it is possible to estimate the result of an action based on brain wave data at a relatively small number of n points. For example, even while viewing a video, an objective evaluation based on the brain wave data can be performed. It can be performed by processing with a small processing load, and the processing load of estimation regarding behavior such as video viewing is reduced.

It is a block diagram which shows the example of the behavior estimation system which concerns on embodiment of this invention. It is a functional block diagram which shows the example of the measurement point selection apparatus which concerns on embodiment of this invention. It is a flowchart which shows the operation example of the measurement point selection apparatus which concerns on embodiment of this invention. It is explanatory drawing which shows the example of the information which the measurement point selection apparatus which concerns on embodiment of this invention holds. It is a flowchart which shows the example of the analysis process of the measurement point selection apparatus which concerns on embodiment of this invention. It is a functional block diagram which shows the example of the behavior estimation apparatus which concerns on embodiment of this invention. It is a flowchart which shows the operation example of the behavior estimation apparatus which concerns on embodiment of this invention.

An embodiment of the present invention will be described with reference to the drawings. As illustrated in FIG. 1, the behavior estimation system 1 according to the embodiment of the present invention includes a measurement point selection device 10 and a behavior estimation device 20. Here, both the measurement point selection device 10 and the behavior estimation device 20 are information processing devices such as a computer. Further, in the following description, it is assumed that the measurement point selection device 10 and the behavior estimation device 20 are separate devices, but one computer device may be made to function as each device.

The behavior estimation system 1 of the present embodiment includes a measurement point selection step of selecting a measurement point of an electroencephalogram using the measurement point selection device 10 and an estimation processing step of estimating a behavior result using the behavior estimation device 20. Perform different processing with.

At the measurement point selection stage, the brain wave data is measured at a large number of measurement points while having the measurement target person selected in advance (the person who cooperated with the experiment, hereinafter referred to as the experiment target person) perform actions such as watching a video. In addition, we received information on the results of the behavior from the experiment subject (for example, by asking them to declare the scene that left an impression on the video), and some of the above-mentioned measurement points were measured. Is selected as a representative measurement point.

Further, in the estimation processing stage, the electroencephalogram data at the selected representative measurement point is measured for the measurement target person (hereinafter referred to as the estimation target person) who is performing the same behavior as the experiment target person. Then, the result of the action (information for identifying the memorable scene in the moving image, etc.) is estimated based on the measured EEG data. That is, in the estimation processing stage, processing is performed only for the brain wave data at the measurement points, which is smaller than the brain wave data measured in the measurement point selection stage, so that the processing load is reduced, and as a result, a larger number of estimation targets are used. The target estimation is possible.

[Configuration of each device]
As illustrated in FIG. 1, the measurement point selection device 10 used in the measurement point selection stage is configured to include a control unit 11, a storage unit 12, an operation unit 13, an output unit 14, and an input / output unit 15, and acts. The estimation device 20 includes a control unit 21, a storage unit 22, an operation unit 23, an output unit 24, and an input / output unit 25.

Further, in an example of the present embodiment, the multi-channel electroencephalograph 30 is connected to the measurement point selection device 10, and the number of channels (measurement points) is smaller in the behavior estimation device 20 than the multi-channel electroencephalograph 30. The electroencephalograph 40 is connected.

The control unit 11 of the measurement point selection device 10 is a program control device such as a CPU, and operates according to a program stored in the storage unit 12. In one example of the present embodiment, the control unit 11 receives information on the brain waves of the experimental subject performing a predetermined action from the multi-channel electroencephalograph 30 via the input / output unit 15.

Here, the multi-channel electroencephalograph 30 measures the electroencephalogram at a predetermined N point (N is an integer where N> 2) and a reference measurement point at M point (M is a natural number) on the head of the experiment subject. The control unit 11 receives the electroencephalogram data as the measurement result at these N points of measurement points, and records the received electroencephalogram data (measurement result) in the storage unit 12.

Further, the control unit 11 provides functional coupling analysis related to EEG data measured at a pair of measurement points (a set of two measurement points) selected from the above N measurement points, and the above-mentioned experimental subject provided by the corresponding experiment subject. Based on the information related to the result of the action, a process of selecting at least a pair of measurement points of N points as a representative point set is executed. The details of the processing of the control unit 11 will be described later.

The storage unit 12 is a memory device, a disk device, or the like, and holds a program executed by the control unit 11. In the example of the present embodiment, the program may be provided by being stored in a computer-readable and non-temporary recording medium, and may be stored in the storage unit 12. The storage unit 12 also operates as a work memory of the control unit 11 and holds the electroencephalogram data and the like which are the measurement results.

The operation unit 13 is a mouse, a keyboard, or the like, and accepts the content of the operation performed by the user of the measurement point selection device 10, and outputs information representing the content of the operation to the control unit 11. The output unit 14 is a display or the like, and displays a screen or the like according to an instruction input from the control unit 11.

The input / output unit 15 is an interface device or the like, and is connected to the multi-channel electroencephalograph 30. Further, the input / output unit 15 outputs the measurement result (electroencephalogram data) input from the multi-channel electroencephalograph 30 to the control unit 11.

Further, in an example of the present embodiment, the input / output unit 15 is connected to a device (for example, an optical scanner device) that receives information related to the result of the action provided by the experiment subject, and outputs information output by the device. It may be accepted and output to the control unit 11.

Further, the control unit 21 of the behavior estimation device 20 used in the estimation processing stage is a program control device such as a CPU, and operates according to a program stored in the storage unit 22. In an example of the present embodiment, the control unit 21 inputs, from the electroencephalograph 40, the input / output unit 25 to the electroencephalogram data of the estimated subject who is performing the same type of behavior as the behavior performed by the experiment subject. Accept through. Here, the electroencephalograph 40 is estimated to measure and output the electroencephalogram data of the estimation target person at n points (n is an integer such that N> n ≧ 2) including the representative point set selected by the measurement point selection device 10. It shall be attached to the subject.

The control unit 21 then estimates the result of the behavior of the estimation target person based on the acquired EEG data. The operation of the control unit 21 will also be described in detail later.

The storage unit 22 is a memory device, a disk device, or the like, and holds a program executed by the control unit 21. In the example of the present embodiment, the program may be provided by being stored in a computer-readable and non-temporary recording medium, and may be stored in the storage unit 12. The storage unit 22 also operates as a work memory of the control unit 21 and holds information such as the measurement result of the brain wave.

The operation unit 23 is a mouse, a keyboard, or the like, and accepts the content of the operation performed by the user of the behavior estimation device 20, and outputs information representing the content of the operation to the control unit 21. The output unit 24 is a display or the like, and displays information or the like according to an instruction input from the control unit 21.

The input / output unit 25 is an interface device or the like, and is connected to the electroencephalograph 40. Further, the input / output unit 25 outputs the information of the measurement result input from the electroencephalograph 40 to the control unit 21.

[Functional configuration of measurement point selection device]
Next, the operation of the control unit 11 of the measurement point selection device 10 in the measurement point selection stage will be described. By executing the program stored in the storage unit 12, the control unit 11 executes the measurement result receiving unit 31, the measurement result recording unit 32, the information receiving unit 33, and the analysis process, as illustrated in FIG. A configuration including a unit 34, a selection processing unit 35, and an output processing unit 36 is realized.

Further, here, it is assumed that the action performed by the experiment subject is the viewing of a moving image, and the moving image to be viewed by the experiment subject is a moving image having a length of about 10 to 90 minutes. Further, for this moving image, the moving image is divided into "scenes" in units of predetermined time (for example, 1 minute), and a unique scene identifier is assigned to each.

Further, a multi-channel electroencephalograph 30 is connected to the measurement point selection device 10 in the measurement point selection stage, and the multi-channel electroencephalograph 30 has a predetermined N point (N is N>) on the head of the test subject. It is connected to electrodes arranged at a measurement point (an integer of 2) and a reference measurement point at point M (where M is a natural number). A conductive gel is injected between the scalp of the subject of the experiment and each electrode of the multi-channel electroencephalograph 30, and the contact impedance is adjusted to be 10 kΩ or less for measurement.

In the example of this embodiment, 30 measurement points determined by the extended international 10/20 method are used, and the measurement points of N points are 28 points (that is, excluding the measurement points A1 and A2 in the pinna of both ears). N = 28), and the reference measurement points at point M are the measurement points A1 and A2 in the pinna of both ears (that is, M = 2).

The multi-channel electroencephalograph 30 includes an electroencephalogram amplifier and a PC (personal computer) for electroencephalogram measurement, and electrodes attached to each measurement point are connected to the electroencephalogram amplifier via a cable. Then, the multi-channel electroencephalograph 30 uses at least one of the reference measurement points at the M point as reference data, and subtracts the reference data from each measurement data (raw data) measured at the measurement points at the N point to form an electroencephalogram amplifier. And amplify it to obtain brain wave data at each measurement point (differential amplification).

Further, the multi-channel electroencephalograph 30 also records the movements and blinks of the eyes of the experimental subject during the action by the electro-oculography measurement performed separately. The multi-channel electroencephalograph 30 inputs the electroencephalogram data at each measurement point differentially amplified by the electroencephalogram amplifier into the PC for electroencephalogram measurement as the electroencephalogram data of the measurement result. The PC for electroencephalogram measurement monitors the electroencephalogram data as the measurement result, and outputs the electroencephalogram data as the measurement result to the measurement point selection device 10 of the present embodiment.

The multi-channel electroencephalograph 30 sequentially selects measurement points for the electroencephalogram data obtained by measuring the electroencephalograms at each of the above measurement points while the experimental subject is performing a predetermined action. Output to device 10.

The measurement result receiving unit 31 accepts the input of the target person identification information for identifying the experiment target person from the user (operator) of the measurement point selection device 10.

Further, the measurement result receiving unit 31 is the measurement result information output by the multi-channel electroencephalograph 30 while the experiment subject is watching the moving image (electroencephalogram data which is the measurement result at each of the above N points). Hereinafter, a collection of electroencephalogram data at all measurement points is referred to as multi-channel electroencephalogram information), and the data is output to the measurement result recording unit 32 together with the subject identification information. In the following description, it is assumed that the electroencephalogram data is input to the measurement point selection device 10 in a state where the electroencephalogram data at which measurement point can be identified.

The measurement result recording unit 32 receives the target person identification information from the measurement result receiving unit 31, sequentially receives the input of the multi-channel electroencephalogram information, and associates the multi-channel electroencephalogram information with the received target person identification information. It is stored in the storage unit 12. At this time, the measurement result recording unit 32 records information (time information) for specifying the measurement time point in association with the multi-channel electroencephalogram information. In the example here, the time information has 0 as the start time of the action (here, the start time of video reproduction) and T as the elapsed time from the start time of the action to the measurement time.

Due to the operation of the measurement result recording unit 32, the storage unit 12 is in a state in which the electroencephalogram data, which is the measurement result at the measurement point of N points, is stored in time series for each experiment subject.

The information receiving unit 33 receives information related to the result of the action from each experiment subject. In the example here, it is assumed that the experimental subject accepts information that identifies a scene that is at least memorable (for example, "remembered"). Further, the information receiving unit 33 may further accept information specifying a scene that is not left in the impression (for example, "not remembered") from the experiment subject. Hereinafter, the scene that is answered to have left the impression is referred to as a storage scene, and the scene identifier that identifies the storage scene is referred to as storage scene information. Further, a scene that is answered not to leave an impression is called a non-memory scene, and a scene identifier that identifies the non-memory scene is called a non-memory scene information.

Specifically, this answer shall be obtained as follows. That is, here, the operator of the measurement point selection device 10 gives the experiment target a key image (still image may be used) selected in advance for each scene of the viewed video after the experiment target has watched the moving image. Present. Then, the experiment subject is asked to answer at least one (for example, five) key images that he / she remembers best (impressive).

The operator of the measurement point selection device 10 inputs the storage scene information for specifying the storage scene to the measurement point selection device 10 by using the scene related to the key image answered here as a storage scene.

Further, when the information receiving unit 33 accepts the information for identifying the non-memory scene, the operator of the measurement point selection device 10 does not store the key image presented to the experiment target person from the previously presented key image (impressive). Ask them to answer at least one (for example, five) key images. Then, in this case, the operator of the measurement point selection device 10 inputs the non-memory scene information that identifies the non-memory scene to the measurement point selection device 10 by using the scene related to the answered key image as a non-memory scene. do.

The information receiving unit 33 stores the storage scene information input based on the answer in the storage unit 12 as the answer information in association with the target person identification information. When accepting the non-memory scene information, the information receiving unit 33 stores the non-memory scene information as the response information together with the storage scene information in the storage unit 12 in association with the target person specific information.

The analysis processing unit 34 reads out the multi-channel electroencephalogram information and the answer information stored in the storage unit 12 from the storage unit 12, and sequentially selects the multi-channel electroencephalogram information associated with each subject specific information as the processing target. Perform the following processing.

The analysis processing unit 34 performs preprocessing on each electroencephalogram data included in the multi-channel electroencephalogram information selected as the processing target. Here, the preprocessing may be a process widely known as an electroencephalogram data preprocessing, such as a process of removing the influence of blinking from the electroencephalogram data during blinking (a process using independent component analysis may be used).

Then, the analysis processing unit 34 divides the electroencephalogram data at the measurement points of N points among the preprocessed electroencephalogram data into the electroencephalogram data (per unit action) while viewing each scene. Here, it is assumed that the scene is divided every 1 minute of the elapsed time from the start of playback of the moving image, so here, if the brain wave data is divided into each scene based on the correspondingly recorded time information, good.

In the following, in order to make the explanation easier to understand, the experiment target person related to the electroencephalogram data to be processed is u (u = u1, u2, ...), And the measurement point is k (k = 1, 2, ... N). Let i (i = 1, 2, ...) be the scene identifier that identifies the scene. In this way, among the electroencephalogram data at the measurement point k of the experiment subject u, the divided electroencephalogram data when the experiment subject u is viewing the scene of the scene identifier i is E_u (k, i). I will write. Further, the electroencephalogram data E_u (k, i) obtained by this division is referred to as an electroencephalogram data for the sake of distinction.

The analysis processing unit 34 applies a predetermined filtering process, for example, a bandpass filter in the theta band (4 to 7 Hz), to each of the divided electroencephalogram data E_u (k, i) related to the experiment subject u. Perform processing. Then, the electroencephalogram (here, the component of theta band) T_u (k, i) after the processing is extracted. Hereinafter, the electroencephalogram data extracted here (here, the component of theta band) is referred to as an electroencephalogram data to be analyzed.

The analysis processing unit 34 selects a set of EEG data to be analyzed at two different measurement points from the EEG data T_u (k, i) to be analyzed related to the experiment subject u {T_u (p, i), T_u (T_u (p, i), T_u ( q, i)} (i = 1, 2, ..., p, q = 1, 2, ... N, where p ≠ q) is obtained. Here, the analysis processing unit 34 obtains a set of EEG data to be analyzed {T_u (p, i), T_u (q, i)} relating to a set of all possible pairs of measurement points.

The analysis processing unit 34 selects the electroencephalogram data to be analyzed at the pair of measurement points in the set, and for the selected pair of electroencephalogram data to be analyzed T_u (p, i) and T_u (q, i). Functional coupling analysis is performed and the analysis result C_u (p, q, i) is obtained.

The functional coupling analysis evaluates, for example, the phase synchronization degree based on the phase difference between the selected pair of analysis target brain wave data T_u (p, i) and T_u (q, i), and the specific of this phase synchronization degree. For example, PLV (Phase Locking Value), information such as iPLV (imaginary PLV), ciPLV (corrected imaginary PLV) acquired by a method derived from the method of acquiring PLV, or PLI (Phase Lag Index), or Information such as wPLI (weighted PLI) and dwPLI (debiased wPLI) obtained by a method derived from the method of obtaining PLI, or obtained by a method derived from PPC (Pairwise Phase Consistency) or a method of obtaining PPC. Any information such as wPPC (weighted PPC) can be used. Further, the processing of this functional coupling analysis is not limited to the evaluation of the phase synchronization degree, and may be the processing of analyzing the amplitude phase coupling, the Granger causality, the mobile entropy, the PSI (Phase Slope Index), the coherence, and the like. In the following description, the case of evaluating the phase synchronization degree will be taken as an example.

Further, the analysis processing unit 34 retrieves the storage scene information (i = Mu1, Mu2, ...) Related to the experiment target person u from the response information associated with the target person specific information to be processed. Then, the analysis processing unit 34 has the analysis results C_u (p, q, Mu1), ..., C_u (p, q) related to the storage scene information among the analysis results C_u (p, q, i) of the functional coupling analysis. , Mu2), ... are extracted, and the statistical value (for example, average) CM_u (p, q) of the analysis result for each set of common measurement points is obtained. In the following, the analysis processing unit 34 shall calculate the average as a statistical value, and the statistical value obtained here is referred to as a storage phase synchronization degree average CM_u (p, q).

Further, the analysis processing unit 34 is the analysis result C_u (p, q, j) of the experimental subject u in the functional coupling analysis relating to the scene other than the storage scene (however, j is neither M1, M2, ...). ) Is extracted, and the average non-memory phase synchronization degree CN_u (p, q), which is an example of the statistical value, is obtained.

The analysis processing unit 34 performs the above processing for each experiment subject, and obtains a storage time phase synchronization degree average and a non-memory time phase synchronization degree average for each experiment subject.

The selection processing unit 35 stores the average phase synchronization degree at the time of storage for each experimental subject u (u = u1, u2, ...) For each combination of the two measurement points (p, q), CM_u1 (p, q), CM_u2. Comparing (p, q), ... with the average non-memory phase synchronization degree CN_u1 (p, q), CN_u2 (p, q), ..., These differences are statistically at a predetermined significance level. A combination of measurement points (p, q) judged to be significant is selected as a representative point set.

Further, one of the characteristics in one example of the present embodiment is that the selection processing unit 35 selects the representative point set because the difference between the average phase synchronization degree during storage and the average phase synchronization degree during non-storage time is significant. At that time, the information for specifying the representative point set (specifying a pair of measurement points) is generated by generating the judgment reference information indicating which value of the average of the phase synchronization degree during storage and the average of the phase synchronization degree during non-storage is larger. It is to be output together with information). This criterion information is, for example, "+" when the storage phase synchronization degree average is larger than the non-storage phase synchronization degree average, and "-" when the storage time phase synchronization degree average is smaller than the non-storage phase synchronization degree average. It is assumed that the information is such as.

As described above, in the present embodiment, by performing the calculation using the statistics of the phase synchronization degree for a plurality of experimental subjects, there is a significant difference between the time of storage and the time of non-memory, excluding the influence of individual differences. You have selected a smaller number of measurement points.

The output processing unit 36 outputs information for specifying a combination of measurement points selected by the selection processing unit 35 (information including at least one information (p, q) for specifying a combination of measurement points) as a representative point set. do. As an example, the output processing unit 36 may store information (information representing a combination of measurement points) for specifying the representative point set selected by the selection processing unit 35 in the storage unit 12.

[Other examples of analysis processing / selection processing]
Further, in the description here, the analysis processing unit 34 is based on the electroencephalogram data while the experiment subject is viewing the storage scene and the electroencephalogram data while the experiment subject is viewing a scene other than the storage scene. , The phase synchronization degree of the electroencephalogram data at a pair of measurement points has been obtained, but the present embodiment is not limited to this example.

For example, as already mentioned, when non-memory scene information (information on the scene that the experiment subject answered that it did not leave an impression) is obtained for the experiment subject together with the memory scene information, as follows. Processing may be performed.

In this example, the analysis processing unit 34 first processes the storage scene in the same manner as described above. That is, the analysis processing unit 34 extracts the storage scene information (i = Mu1, Mu2, ...) Related to the experiment target person u from the response information associated with the target person specific information to be processed. Then, the analysis processing unit 34 has the analysis results C_u (p, q, Mu1), ..., C_u (p, q) related to the storage scene information among the analysis results C_u (p, q, i) of the functional coupling analysis. , Mu2), ... Are extracted, and the storage phase synchronization degree average CM_u (p, q), which is a statistical value of the analysis result for each set of common measurement points, is obtained.

Further, the analysis processing unit 34 analyzes the analysis results of the functional coupling analysis related to the non-memory scene information (j = Nu1, Nu2 ...) For each experimental subject u, C_u (p, q, Nu1), ..., C_u (p). , Q, Nu2), ... Are extracted, and the average non-memory phase synchronization degree CN_u (p, q), which is a statistical value of the analysis result for each set of common measurement points, is obtained.

Then, the selection processing unit 35 sets the average storage phase synchronization degree CM_u1 (p, q) for each experimental subject u (u = u1, u2, ...) For each combination of the two measurement points (p, q). Comparing CM_u2 (p, q), ... with the non-memory phase synchronization average CN_u1 (p, q), CN_u2 (p, q), ..., These differences are statistically at a predetermined significance level. The combination of measurement points (p, q) judged to be significant in is selected as the representative point set.

[Operation of measurement point selection device]
Next, the operation of the measurement point selection device 10 according to the above configuration will be described. 30 measurement points (2 points of reference measurement points A1 and A2, as well as Fp1 and Fp2) determined by the extended international 10/20 method on the head surface of each of a plurality of experimental subjects u1, u2 ... , F7, F3, Fz, F4, F8, FC5, FC1, FC2, FC6, T7, T8, C3, Cz, C4, CP5, CP1, CP2, CP6, P7, P3, Pz, P4, P8, O1, Oz , 28 points of O2), an electrode adjusted to have a contact impedance of 10 kΩ or less by applying a conductive gel is arranged and connected to a multi-channel electroencephalograph 30.

Here, as the multi-channel electroencephalograph 30, for example, LiveAmp (registered trademark) manufactured by Brain Products, Germany can be used.

Then, while letting each experiment subject watch a 25-minute video, the electroencephalogram data at each of the above measurement points is acquired by the multi-channel electroencephalograph 30 for 25 minutes.

The multi-channel electroencephalograph 30 uses brain wave data measured at reference electrodes (here, predetermined as electrodes arranged at measurement points A1 and A2 in the ears of both ears) as reference data for each experimental subject. EEG data of the experimental subject during video viewing at 28 measurement points other than the reference electrode is obtained by differential amplification.

Then, the multi-channel electroencephalograph 30 sequentially outputs a set of electroencephalogram data for each experiment subject obtained by this differential amplification as multi-channel electroencephalogram information of each experiment subject.

The operator of the measurement point selection device 10 inputs the target person identification information for identifying the test target person, and the multi-channel electroencephalograph 30 sequentially outputs the target person identification information of the test target person specified by the target person identification information. Have the measurement point selection device 10 accept the multi-channel electroencephalogram information.

The measurement point selection device 10 performs the process illustrated in FIG. 3 according to the instruction of the operator, receives the input of the target person specific information, and sequentially receives the input of the multi-channel electroencephalogram information, and receives the multi-channel electroencephalogram information. , It is stored in the storage unit 12 in association with the received target person identification information (S11). As a result, as illustrated in FIG. 4, the storage unit 12 is in a state in which the electroencephalogram data at the measurement points at the N points are stored in time series for each experiment subject.

On the other hand, the operator of the measurement point selection device 10 asks the experiment target person who has finished watching the 25-minute video for each video scene (here, the video is divided into 1-minute intervals and used as different scenes). Presents the selected key images to and asks them to answer the five most memorized (impressive) key images and the five unmemorized (not memorable) key images.

Further, when the operator of the measurement point selection device 10 receives an answer from the experiment target person u, the information Mu1, Mu2, which identifies the scene related to the key image, which is the most memorable (impressive). Mu3, Mu4, Mu5 and information Nu1, Nu2, Nu3, Nu4, Nu5 for specifying the scene related to the key image that is not memorized (not left in the impression) are stored scene information and non-memory scene, respectively. As the information, it is input to the measurement point selection device 10 together with the target person identification information for specifying the responded experiment target person.

The measurement point selection device 10 receives the response information including the storage scene information and the non-storage scene information, and the target person identification information, and stores them in the storage unit 12 in association with each other (S12).

The measurement point selection device 10 repeatedly executed steps S11 and S12 until the multi-channel electroencephalogram information and the response information for all the experiment subjects were input, and the information input for all the experiment subjects was completed. Occasionally, the following processing is started according to the instruction of the operator.

That is, the measurement point selection device 10 reads out the multi-channel electroencephalogram information and the answer information stored in the storage unit 12 from the storage unit 12, and sequentially selects each experimental target person specified by the target person identification information as the target person. (S13), the following processing is executed.

The measurement point selection device 10 reads out the multi-channel electroencephalogram information associated with the subject-specific information of the selected subject u, and removes the influence of blinking on each electroencephalogram data included in the multi-channel electroencephalogram information. Preprocessing such as (S14).

The measurement point selection device 10 divides the brain wave data at the measurement points of N points among the preprocessed brain wave data into the brain wave data every minute from the start of viewing the moving image, and divides the brain wave data E_u ( k, i) is obtained (S15). In the example here, since the moving image is divided into scenes every minute, this process corresponds to the process of dividing into brain wave data while viewing each scene. Here, k (k = 1, 2, ... N) represents a measurement point, and i corresponds to a scene identifier (i = 1, 2, ...).

The measurement point selection device 10 applies a bandpass filter in the theta band (4 to 7 Hz) to each of the divided electroencephalogram data E_u (k, i) obtained in step S15, and the electroencephalogram to be analyzed, which is a component of the theta band. The data T_u (k, i) is extracted (S16).

Next, the measurement point selection device 10 executes the analysis process (S17). In this analysis process, the measurement point selection device 10 lists a pair of measurement points (p, q) that are different from each other (S31), as shown in FIG. In the example here, the measurement points (excluding the reference measurement points) are Fp1, Fp2, F7, F3, Fz, F4, F8, FC5, FC1, FC2, FC6, T7, T8, C3, Cz, C4, CP5, Since there are 28 points of CP1, CP2, CP6, P7, P3, Pz, P4, P8, O1, Oz, O2,
(Fp1, Fp2) set,
(Fp1, F7) set,
…
(Fp1, O2) set,
(Fp2, F7) set,
…
There are 378 combinations, such as the set of (Oz, O2), which is the number of combinations for extracting 2 points from 28 points. After enumerating these combinations, the measurement point selection device 10 enumerates each combination for each of the scene identifiers Mu1, Mu2, Mu3, Mu4, Mu5, Nu1, Nu2, Nu3, Nu4, Nu5 included in the response information of the target person u. {T_u (Fp1, i), T_u (Fp2, i)},
{T_u (Fp1, i), T_u (F7, i)}, ...
378 pairs of. Then, the dwPLI value as the analysis result of the functional coupling analysis for each set of EEG data to be analyzed:
C_u (Fp1, Fp2, i),
C_u (Fp1, F7, i), ...
Is obtained (S32). However, i = Mu1, Mu2, Mu3, Mu4, Mu5, Nu1, Nu2, Nu3, Nu4, Nu5. As a result, the measurement point selection device 10 functionally binds the analysis result of the analysis target brain wave data of the analysis target brain wave data when viewing the storage scene of the subject u and the analysis target brain wave data of the analysis target brain wave data when viewing the non-memory scene. Obtain the analysis result of the analysis.

The measurement point selection device 10 averages the dwPLI values (average of phase synchronization during storage), which is the analysis result of the functional coupling analysis of the electroencephalogram data to be analyzed at the time of viewing the storage scene, for each set of the electroencephalogram data to be analyzed.
CM_u (Fp1, Fp2),
CM_u (Fp1, F7), ...
Is obtained (S33).

Further, the measurement point selection device 10 averages the dwPLI values (phase synchronization degree at the time of non-memory) which is the analysis result of the functional coupling analysis of the electroencephalogram data to be analyzed at the time of viewing the non-memory scene for each set of the electroencephalogram data to be analyzed. average)
CN_u (Fp1, Fp2),
CN_u (Fp1, F7), ...
(S34).

The measurement point selection device 10 returns to step S18 in FIG. 3, returns to step S13 until all the experimental subjects are processed, newly selects the previously unselected subject u, and steps S14 to S17. Repeat the process.

When the measurement point selection device 10 obtains the storage phase synchronization degree average and the non-storage phase synchronization degree average for all the experimental subjects, the same pair of measurement points as those listed in step S31 (p). , Q) = (Fp1, Fp2), (Fp1, F7), ... For each experimental subject u (u = u1, u2, ...), Average phase synchronization degree at storage CM_u1 (p, q), CM_u2 (P, q), ... And the non-memory phase synchronization average CN_u1 (p, q), CN_u2 (p, q), ... were compared (S19), and the difference between them was statistically determined in advance. Determine if it is significant at the significance level.

Then, the measurement point selection device 10 combines a pair of measurement points that are determined to have a significant difference between the storage phase synchronization degree average and the non-storage phase synchronization degree average at a statistically predetermined significance level. , Select as a representative point set (S20), and output the selected representative point set (S21).

When the measurement point selection device 10 selects the representative point set in step S19, which value is larger, the storage phase synchronization degree average or the non-storage phase synchronization degree average, in which the difference is significant. Is generated, and in step S20, it is output together with the information for specifying the representative point set (information for specifying the pair of measurement points). This criterion information is, for example, "+" when the storage phase synchronization degree average is larger than the non-storage phase synchronization degree average, and "-" when the storage time phase synchronization degree average is smaller than the non-storage phase synchronization degree average. It is information such as.

Specifically, in an example of the present embodiment, the measurement point selection device 10 stores the memory of each experimental subject u (u = u1, u2, ...) For each combination of a pair of measurement points (p, q). Distribution of time-phase synchronization average CM_u1 (p, q), CM_u2 (p, q), ... And distribution of non-memory phase synchronization average CN_u1 (p, q), CN_u2 (p, q), ... The null hypothesis that there is no statistically significant difference between them is tested by t-test, and it is determined whether or not it can be rejected at a predetermined significance level (for example, 5% or 1%). Then, the measurement point selection device 10 selects the combination of measurement points (p, q) that can reject the null hypothesis at a predetermined significance level as a representative point set.

In one example of this embodiment, when the significance level is 5%, the measurement point selection device 10 is
・ (FC5, Fp2) (+)
・ (O2, Fp2) (-)
・ (P7, T7) (+)
・ (T7, CP5) (+)
・ (CP5, FC6) (+)
・ (FC2, FC6) (-)
・ (FC6, F4) (-)
・ (FC5, FC6) (+)
・ (Pz, CP6) (+)
・ (CP2, CP6) (+)
・ (CP1, FC5) (-)
・ (CP1, FC2) (-)
・ (C4, Fp2) (+)
As there was a significant difference among the 13 sets of, these were selected as the representative point set. Here, (+) or (-) is the judgment reference information, and becomes (+) when the storage phase synchronization degree average is larger than the non-storage phase synchronization degree average, and the non-storage phase synchronization degree average is stored. If it is larger than the average time-phase synchronization degree, it is assumed to be (-).

Further, when the significance level was set to 1%, the measurement point selection device 10 selected only the combination of (FC5, Fp2) and (+) as the representative point set, assuming that there was a significant difference.

That is, the present embodiment functions between the channel pairs in which the phase synchronization degree of the brain wave is significantly different between the case where the moving image scene is memorized and the case where the moving image scene is not memorized as in the example here. The pair of channels judged to have these functional connections, which are considered to exist, is used as a representative point set.

[Other examples of representative point set]
Further, in the description so far, since the representative point set is selected based on the phase synchronization degree of the electroencephalogram data at the pair of measurement points, it is selected as the pair of measurement points. The form of is not limited to this.

For example, the phase synchronization degree between the brain wave data for each combination of three or more measurement points is evaluated, and the phase synchronization degree with respect to the phase synchronization degree between the brain wave data related to the combination of other measurement points while viewing the storage scene. A combination of measurement points having a significant difference may be found, and the measurement points related to the found combination may be selected as a representative point set.

[Functional configuration of behavior estimation device]
Next, the operation of the control unit 21 of the behavior estimation device 20 will be described. By executing the program stored in the storage unit 22, the control unit 21 functionally includes the brain wave information acquisition unit 41, the estimation processing unit 42, and the output processing unit 43, as illustrated in FIG. Realize the configuration.

Further, the electroencephalograph 40 connected to the behavior estimation device 20 focuses on the network between at least a pair of measurement points included in the representative point set selected by the measurement point selection device 10, and obtains the electroencephalogram data at the measurement points. It is to measure. Therefore, the electroencephalograph 40 is simpler than the multi-channel electroencephalograph 30, and the equipment cost and the burden required for measurement are relatively light.

In one example of the present embodiment, the electroencephalograph 40 sequentially acquires the electroencephalogram data of the estimated target person in action, and among the measurement points on the head surface of the estimated target person, FC5 and Fp2. EEG data at two points (measurement points selected by the measurement point selection device 10 at a significance level of 1%) and measurement points A1 and A2 in the ears of both ears as reference measurement points. measure. Then, the electroencephalograph 40 subtracts and amplifies the electroencephalogram data measured at the measurement points A1 or A2 as reference data from each electroencephalogram data (raw data) measured at the two points FC5 and Fp2, and amplifies the electroencephalogram data of the FC5 and Fp2. It is output as brain wave data at each point. Here, FC5 and Fp2 are all measurement points defined by the Extended International 10/20 Law.

Further, in the following, in order to facilitate the explanation, the behavior is assumed to be the viewing of a moving image, and in the behavior estimation device 20, was the scene included in the moving image memorable to the estimation target person (whether it became a memory scene)? It is assumed that it is to estimate whether or not it is.

The electroencephalogram information acquisition unit 41 sequentially accepts the input of the electroencephalogram data at each point of FC5 and Fp2 of the estimated target person in action from the electroencephalograph 40. In the following description, it is assumed that the electroencephalogram data is input to the behavior estimation device 20 in a state where the electroencephalogram data at which measurement point can be identified.

The estimation processing unit 42 accepts and stores the input of the information for specifying the representative point set selected by the measurement point selection device 10 for a predetermined action and the determination reference information in the representative point set. Then, when the electroencephalogram information acquisition unit 41 sequentially receives each electroencephalogram data, the estimation processing unit 42 divides each of the electroencephalogram data at predetermined timings and divides each of the electroencephalogram data into action units (viewing in the case of video viewing). Divided EEG data E (k, i) (k represents a measurement point, i is an identifier that identifies a unit of action, and in this example, it is a scene identifier and i = 1, 2, ...). For example, the estimation processing unit 42 takes out the electroencephalogram data at each point received in the past one minute every minute as divided electroencephalogram data at each measurement point.

The estimation processing unit 42 performs processing such as predetermined filtering (processing equivalent to the processing performed by the measurement point selection device 10, here, for example, a band of theta band (4 to 7 Hz)) for each of the extracted divided electroencephalogram data. The process of applying the path filter) is performed, and the EEG data T (k, i) to be analyzed is obtained from each divided EEG data E (k, i) (here, the component of the theta band is extracted).

The estimation processing unit 42 then analyzes at least a pair of measurement points including a representative point set selected by the measurement point selection device 10 at a pair of measurement points (p, q) related to the analysis target. Functional coupling analysis is performed on the target EEG data T (p, i) and T (q, i), and the analysis result C (p, q, i) is obtained.

The functional coupling analysis performed by the estimation processing unit 42 is the same as that in the measurement point selection device 10, and for example, the phase synchronization degree based on the phase difference between the pair of analysis target electroencephalogram data as the analysis target is evaluated. It is a thing. Specifically, this functional coupling analysis includes PLV (Phase Locking Value) and derived iPLV (imaginary PLV), ciPLV (corrected imaginary PLV), etc., or PLI (Phase Lag Index) and wPLI (weighted) derived from it. It is performed by adopting a widely known method for requesting information of any one of PLI), dwPLI (debiased wPLI), etc., or PPC (Pairwise Phase Consistency), and wPPC (weighted PPC) derived from the information. Further, the processing of this functional coupling analysis is not limited to the evaluation of the phase synchronization degree, and may be the processing of analyzing the amplitude phase coupling, the Granger causality, the mobile entropy, the PSI (Phase Slope Index), the coherence, and the like.

The estimation processing unit 42 refers to the result C (p, q, i) (for example, the value of dwPLI) of the functional connection analysis corresponding to each scene identifier and the judgment criterion information corresponding to each representative point set, and refers to the scene. Of the analysis results C (p, q, i) of the representative point set (p, q) corresponding to the identifiers i = 1, 2, ..., The determination indicated by the judgment criterion information corresponding to the representative point set (p, q). Search for a scene identifier that meets the criteria.

Specifically, when the determination criterion information is "+", the estimation processing unit 42 sets the average CM (p, q) of the analysis result C (p, q, i) related to the representative point set (p, q), for example. On the other hand, a scene identifier j in which C (p, q, j)> CM (p, q) and C (p, q, j) -CM (p, q) is larger than a predetermined threshold value is searched. do. Then, when such a scene identifier j is found, the estimation processing unit 42 determines that the scene identified by the found scene identifier is a memorable scene.

Further, in this case, the estimation processing unit 42 has the analysis result C (p, q, i) related to the representative point set (p, q) for the representative point set (p, q) whose determination reference information is “−”. ) With respect to the average CM (p, q), C (p, q, j) <CM (p, q), and CM (p, q) -C (p, q, j) is predetermined. Search for the scene identifier j that is larger than the threshold value. Then, when such a scene identifier j is found, the estimation processing unit 42 determines that the scene identified by the found scene identifier is a memorable scene.

The estimation processing unit 42 finds a scene identifier of a scene that is presumed to be memorable by such processing. Then, the output processing unit 43 outputs the scene identifier found by the estimation processing unit 42.

[Operation of behavior estimation device]
Next, the operation of the behavior estimation device 20 will be described. In the example of this embodiment, a relatively small number of measurement points for measuring EEG data are selected in advance by the measurement point selection device 10. As described above, the measurement point selection device 10 selects at least a pair of measurement points as a representative point set by using the electroencephalogram data at the N measurement points of the experiment subject who performs a predetermined action.

The user of the behavior estimation device 20 (the worker who estimates the behavior of the estimation target person) is a measurement point including at least one pair of the above representative point sets on the head of the measurement target person (estimation target person) and a reference measurement point. The electrodes of the electroencephalograph 40 are arranged at and. This presumed subject shall perform the same behavior as the experiment subject performed when selecting this representative point set.

Specifically, in the following example, the action of watching the video is assumed as in the example already described, and the scene in the video (here, the scene in which the video to be watched is divided into 1 minute is referred to as each scene). Of these, it is estimated whether or not there was an action of memorizing the scene (viewing a memorable scene), and which was the memorable scene is detected.

Prior to the use of the behavior estimation device 20, when the representative point set is selected by the measurement point selection device 10, the moving image watched by the experiment subject who acquired the brain wave data and the brain wave input to the behavior estimation device 20. It may be different from the video viewed by the estimated target person who acquires the data.

The behavior estimation device 20 receives the electroencephalogram data measured by the electroencephalograph 40. Here, the electroencephalograph 40 focuses on the network between at least a pair of measurement points included in the representative point set selected by the measurement point selection device 10, and the electroencephalogram at the measurement point of the estimation target person who is watching the moving image. It is assumed that the data is measured and output. In the following, the electroencephalograph 40 has two measurement points of FC5 and Fp2 as representative point sets (measurement points selected by the measurement point selection device 10 at a significance level of 1%) among the measurement points on the head surface of the estimation target person. ), And the electroencephalogram data at the measurement points A1 and A2 in the pinna of both ears, which are the reference measurement points, shall be measured. Further, it is assumed that the judgment criterion information corresponding to the representative point set (FC5, Fp2) is "+".

The behavior estimation device 20 sequentially receives the electroencephalogram data from the electroencephalograph 40 and performs the process illustrated in FIG. 7. The behavior estimation device 20 receives and stores in advance the input of the information for specifying the representative point set selected by the measurement point selection device 10 regarding the behavior of watching the moving image and the determination reference information in the representative point set.

The behavior estimation device 20 performs preprocessing (same processing as step S14 of the measurement point selection device 10) for each of the electroencephalogram data at each measurement point sequentially received from the electroencephalograph 40, and then the behavior estimation device 20 performs the same processing. It is divided into brain wave data for each scene (here, every minute). The behavior estimation device 20 is divided brain wave data E (FC5, i), E (Fp2, i) for each scene (i is an identifier that identifies a unit of behavior, and in this example, it is a scene identifier and i. = 1, 2, ...) (S51).

The behavior estimation device 20 applies a bandpass filter in the theta band (4 to 7 Hz) to each of the divided electroencephalogram data extracted in step S51, and each divided electroencephalogram data E (FC5, i), E (Fp2,). i) The components of the theta band are extracted to obtain the electroencephalogram data T (FC5, i) and T (Fp2, i) to be analyzed (S52).

Then, the behavior estimation device 20 repeats the following processing for each analysis target and for each scene identifier, with the measurement points included in the representative point set selected by the measurement point selection device 10 as analysis targets.

In the process of this repetitive processing, the behavior estimation device 20 receives the analysis target electroencephalogram data T (FC5, i) and T (Fp2, i) at the pair of measurement points (FC5, Fp2) related to the analysis target. As a functional coupling analysis, for example, a process for obtaining dwPLI is performed, and the analysis result C (FC5, Fp2, i) is obtained (S53).

Further, after the analysis results C for all the scenes are obtained, the behavior estimation device 20 sets the dwPLI value C (FC5, Fp2, i) corresponding to each scene identifier and the representative point set for each representative point set. Of the analysis results C (FC5, Fp2, i) of the representative point set (FC5, Fp2) corresponding to the scene identifiers i = 1, 2, ... With reference to the corresponding determination criterion information (here, "+"). , Search for a scene identifier that satisfies the determination criteria indicated by the determination criteria information "+" corresponding to the representative point set (FC5, Fp2) (S54).

That is, here, the behavior estimation device 20 has C (FC5, Fp2, j)> with respect to the average CM (FC5, Fp2) of the analysis result C (FC5, Fp2, i) relating to the representative point set (FC5, Fp2). The scene identifier j which is CM (FC5, Fp2) and whose C (FC5, Fp2, j) -CM (FC5, Fp2) is larger than a predetermined threshold value is searched. Then, when such a scene identifier j is found, the behavior estimation device 20 outputs the found scene identifier j (S55).

As a result, the user of the behavior estimation device 20 obtains a scene identifier of a scene that is estimated to leave an impression on the estimation target person.

Then, the above processing is repeated for a plurality of different estimation target persons, and the identifier of the scene estimated to be memorable for each estimation target person is obtained, and which part of the video is the most memorable scene. It is used for processing such as analyzing whether it is.

According to this embodiment, it is possible to perform an objective analysis without conducting a questionnaire survey on behavior. In addition, since a pair of measurement points that can detect characteristic electroencephalograph data at a predetermined action are selected in advance without using a multi-channel electroencephalograph, it is possible to estimate an action using a simple electroencephalograph with a small number of channels. It has become.

In addition, since the number of channels to be processed is small, it is possible to estimate the memory of the scene by processing with a relatively light load, and it is possible to collect estimation results from a larger number of estimation targets than before.

Furthermore, by using the results estimated in this way, it is possible to provide an objective evaluation of how memorable the content such as the presented video is, and to provide a more impressive video and a more promising feeling. It is also possible to contribute to the purpose such as providing videos that can give.

In addition, it can be used for processing such as recommendation of contents for each individual by measuring by an individual using an electroencephalograph and analyzing a scene that is likely to remain in his / her own memory by the above processing.

1 Behavior estimation system, 10 Measurement point selection device, 11 Control unit, 12 Storage unit, 13 Operation unit, 14 Output unit, 15 Input / output unit, 20 Behavior estimation device, 21 Control unit, 22 Storage unit, 23 Operation unit, 24 Output unit, 25 input / output unit, 30 multi-channel electrosurgical unit, 31 measurement result receiving unit, 32 measurement result recording unit, 33 information receiving unit, 34 analysis processing unit, 35 selection processing unit, 36 output processing unit, 40 electrosurgical unit, 41 Brain wave information acquisition unit, 42 estimation processing unit, 43 output processing unit.

Claims (10)

It is a behavior estimation system that estimates the result of a predetermined type of behavior performed by a measurement target person by measuring brain waves, and includes a measurement point selection device and a behavior estimation device.
The measurement point selection device
The electroencephalogram data of the measurement target person performing the predetermined type of behavior is the measurement point of the predetermined N point (N is an integer where N> 2) of the subject's head and the M point (M is a natural number). ) Reference measurement point and measurement means to measure,
The measurement points at the N points are based on the functional connection analysis of the EEG data at the pair of measurement points selected from the measurement points at the N points and the result of the action presented by the measurement target person. A selection means for selecting at least a pair of points as a representative point set, and
Including
The behavior estimation device
Acquisition of acquisition of brain wave data at n points (n is an integer such that N> n ≧ 2) including the representative point set selected by the measurement point selection device of the measurement target person performing the predetermined type of behavior. Means and
An estimation processing means that estimates the result of the behavior of the measurement target person based on the acquired EEG data, and
Behavior estimation system including. The EEG data of the measurement target person performing a predetermined type of behavior is the measurement point of the predetermined N point (N is an integer of N> 2) and the M point (M is a natural number) of the subject's head. Measurement means to measure with the reference measurement point of
The measurement points at the N points are based on the functional connection analysis of the EEG data at the pair of measurement points selected from the measurement points at the N points and the result of the action presented by the measurement target person. A selection means for selecting at least a pair of points as a representative point set, and
Including
The information that identifies the representative point set is n points (n is an integer such that N> n ≧ 2) including the representative point set specified by the information of the measurement target person who is performing the predetermined type of action. A measurement point selection device that is used for processing to acquire the brain wave data in the device and estimate the result of the behavior of the measurement target person. The measurement point selection device according to claim 2.
The selection means is derived from a method of acquiring a PLV (Phase Locking Value) or PLV as a process of the functional coupling analysis for a combination of all two measurement points that can be selected from the N points of measurement points. Information obtained by the method obtained, or information obtained by a method derived from the method of obtaining PLI (Phase Lag Index) or PLI, or a method derived from PPC (Pairwise Phase Consistency) or a method of obtaining PPC. Perform a phase-locked analysis based on any of the information acquired in, and based on the result of the phase-locked analysis and the result of the action presented by the measurement target person, the measurement point of the N point is A measurement point selection device that selects at least a pair of points as a representative point set. The measurement point selection device according to claim 2 or 3.
The selection means is a process of the functional binding analysis, in which the result of the functional binding analysis when the result of the action satisfies the predetermined condition and the result of the result of the action not satisfying the predetermined condition. A measurement point selection device that selects at least a pair of measurement points among the N points as a representative point set by comparison with the result of the functional coupling analysis in the case of. The measurement point selection device according to any one of claims 2 to 4.
The above-mentioned action is an action of memorizing the contents of each scene included in the video.
The selection means is based on a functional connection analysis of brain wave data at two measurement points selected from the N measurement points for each scene, and the result of the action presented by the measurement target person. Then, a measurement point selection device that selects two of the N points of measurement points as representative point set candidates and selects two of the N points of measurement points as representative points based on the selection result for each scene. .. It is a behavior estimation device that estimates the results of predetermined types of behavior performed by the measurement target by measuring brain waves.
A measurement point at a predetermined N point (N is an integer where N> 2) of the head selected based on the measurement result for the measurement target person who has performed the predetermined type of action, and an M point (M is a natural number). ) Refers to the information that identifies two points that are the representative point set selected from the N points of the measurement points, and n points (n is n) including the representative point set specified by the information. An acquisition means for acquiring EEG data with N> n ≧ 2) and
An estimation processing means that estimates the result of the behavior of the measurement target person based on the acquired EEG data, and
Behavior estimation device including. The EEG data of the measurement target person performing a predetermined type of behavior is the measurement point of the predetermined N point (N is an integer of N> 2) and the M point (M is a natural number) of the subject's head. The measurement process measured at the reference measurement point of
Based on the functional coupling analysis of the electroencephalogram data at the two measurement points selected from the N point measurement points and the result of the action presented by the measurement target person, the N point measurement point A selection process that selects at least a pair of points as a representative point set,
Including
The information that identifies the representative point set is n points (n is an integer such that N> n ≧ 2) including the representative point set specified by the information of the measurement target person who is performing the predetermined type of action. A control method for an information processing device that is used for processing to acquire the brain wave data in the above and estimate the result of the behavior of the measurement target person. It is a control method of a behavior estimation device that causes a behavior estimation device to estimate the result of a predetermined type of behavior performed by a measurement target person by measuring brain waves.
A measurement point at a predetermined N point (N is an integer where N> 2) of the head selected based on the measurement result for the measurement target person who has performed the predetermined type of action, and an M point (M is a natural number). ) Refers to the information that identifies two points that are the representative point set selected from the N points of the measurement points, and n points (n is) including the representative point set specified by the information. The process of acquiring EEG data with N> n ≧ 2) and
A control method of a behavior estimation device including a step of estimating a behavior result of the measurement target person based on the acquired electroencephalogram data. Computer,
The EEG data of the measurement target person performing a predetermined type of behavior is the measurement point of the predetermined N point (N is an integer of N> 2) and the M point (M is a natural number) of the subject's head. Measurement means to measure with the reference measurement point of
Based on the functional coupling analysis of the electroencephalogram data at the two measurement points selected from the N point measurement points and the result of the action presented by the measurement target person, the N point measurement point A selection method for selecting at least a pair of points as a representative point set,
To function as
The information that identifies the representative point set is n points (n is an integer such that N> n ≧ 2) including the representative point set specified by the information of the measurement target person who is performing the predetermined type of action. A program that acquires brain wave data in the above and uses it for processing to estimate the result of the behavior of the person to be measured. A program that causes a computer to function by measuring brain waves to estimate the results of predetermined types of actions performed by a person to be measured.
A measurement point at a predetermined N point (N is an integer where N> 2) of the head selected based on the measurement result for the measurement target person who has performed the predetermined type of action, and an M point (M is a natural number). ) Refers to the information that identifies two points that are the representative point set selected from the N points of the measurement points, and n points (n is n) including the representative point set specified by the information. An acquisition means for acquiring EEG data with N> n ≧ 2) and
An estimation processing means that estimates the result of the behavior of the measurement target person based on the acquired EEG data, and
A program that functions as.

Images