Support device, support method, and support program

Abstract

SOLUTION: A support device includes an information acquisition part for acquiring brain wave information of an object person when a living body acts on the object person in a state where it is difficult to communicate, and a determination part for determining the state of the object person based on the brain wave information. The information acquisition part may further acquire biological information of the object person. The determination part may determine the state of the object person based on the brain wave information and the biological information. The information acquisition part may acquire brain wave information of the object person before the action. The determination part may generate state information representing a state of the object person based on a change from the brain wave information before the action to the brain wave information after the action and the biological information, and may determine the state of the object person based on the generated state information.SELECTED DRAWING: Figure 2

Description

【Technical Field】 【0001】 The present invention relates to a support device, a support method, and a support program. 【Background Art】 【0002】 Patent Document 1 describes that "without making the user aware, the target behavior is strengthened" (abstract). Patent Document 2 describes that "realize three-dimensional movement control of the avatar of the operator in the VR space" (abstract). Patent Document 3 describes that "generate a feedback loop in which the state of an individual's brain modulates the parameters of an augmented reality system" (abstract). [Prior Art Documents] [Patent Documents] [Patent Document 1] International Publication No. 2019 / 082687 [Patent Document 2] Japanese Unexamined Patent Application Publication No. 2022-020057 [Patent Document 3] Japanese Patent Application Publication No. 2021-511612 【Summary of the Invention】 【0003】 In a first aspect of the present invention, a support device is provided. The support device includes an information acquisition unit that acquires electroencephalogram information of a target person when a living body acts on the target person in a state where conversation is difficult, and a discrimination unit that discriminates the state of the target person based on the electroencephalogram information. 【0004】 The information acquisition unit may further acquire biometric information of the target person. The discrimination unit may discriminate the state of the target person based on the electroencephalogram information and the biometric information. 【0005】 In any of the above support devices, the information acquisition unit may acquire the electroencephalogram information of the target person before the action. The discrimination unit generates state information indicating the state of the target person based on the change from the electroencephalogram information before the action to the electroencephalogram information after the action and the biometric information, and discriminates the state of the target person based on the generated state information. 【0006】 In any of the above-described support devices, the discrimination unit may generate state information based on the change from the ratio of the amplitude of brain waves in a predetermined frequency band to the total amplitude in the brainwave information before intervention, to the ratio of the amplitude of brain waves in a frequency band to the total amplitude in the brainwave information after intervention, and the ratio of the magnitude of the first power spectrum of the target person's heartbeat to the magnitude of the second power spectrum. The total amplitude is the sum of the amplitudes of alpha waves, beta waves, theta waves, gamma waves, and delta waves. The frequency band of the second power spectrum is a higher frequency band than the frequency band of the first power spectrum. 【0007】 In any of the above-mentioned support devices, the discrimination unit may generate state information based on the change from the ratio of the amplitude of brain waves in a predetermined frequency band to the total amplitude in the brainwave information before intervention to the ratio of the amplitude of brain waves in a predetermined frequency band to the total amplitude in the brainwave information after intervention, and the relationship between the ratio of the magnitude of the first power spectrum to the magnitude of the second power spectrum after intervention and a predetermined threshold for the ratio of the magnitude of the first power spectrum to the magnitude of the second power spectrum. 【0008】 In any of the above support devices, the state information may include information relating to multiple states of the target person. The discrimination unit may generate state information relating to one of the multiple states based on the change from the ratio of the amplitude of brain waves in a predetermined frequency band to the total amplitude in the brainwave information before intervention to the ratio of the amplitude of brain waves in a frequency band to the total amplitude in the brainwave information after intervention, and the ratio of the magnitude of the first power spectrum to the magnitude of the second power spectrum. 【0009】 In any of the above-mentioned support devices, the brainwaves in the frequency band may be at least one of delta waves, theta waves, low alpha waves, and medium alpha waves. 【0010】 In any of the above-mentioned support devices, the brainwaves in the frequency band may be at least one of high alpha waves, low beta waves, high beta waves, and gamma waves. 【0011】 Any of the above-mentioned support devices may further include a learning unit that generates an intervention inference model that infers an intervention to bring a person to a predetermined state based on brainwave information and the person's state, by machine learning the relationship between brainwave information and interventions and interventions that bring the person's state to a predetermined state. 【0012】 Any of the above support devices may further include a state presentation unit that presents the state of the target person identified by the discrimination unit. The information acquisition unit may further acquire attribute information indicating the attributes of the biological organism. The state presentation unit may present the state of the target person based on the attribute information. 【0013】 In any of the above-mentioned support devices, the status display unit may change the way in which the status of the target person is displayed based on attribute information. 【0014】 A second aspect of the present invention provides a support method. The support method comprises an information acquisition step in which an information acquisition unit acquires brainwave information of a target person when a biological device interacts with the target person who is in a state where dialogue is difficult, and a discrimination step in which a discrimination unit determines the state of the target person based on the brainwave information acquired in the information acquisition step. 【0015】 A third aspect of the present invention provides a support program. The support program includes an information acquisition step in which a computer acquires brainwave information of a target person when a biological device interacts with that person, who is in a state where dialogue is difficult, and a determination step in which the computer determines the state of the target person based on the brainwave information acquired in the information acquisition step. 【0016】 It should be noted that the above summary of the invention does not enumerate all of its features. Furthermore, subcombinations of these features may also constitute an invention. [Brief explanation of the drawing] 【0017】 [Figure 1] The figure shows an example of the situation before the living body 120 acts on the target person 110 and an example of the situation where the act is being performed. [Figure 2] The block diagram shows an example of the support device 100 according to one embodiment of the present invention. [Figure 3] The figure shows an example of the information acquisition unit 10. [Figure 4] The figure shows an example of the state information Is. [Figure 5A] The figure shows an example of the learning by the learning unit 40. [Figure 5B] The figure shows an example of the inference by the act inference model 42. [Figure 6] The figure shows another example of the situation before the living body 120 acts En on the target person 110 and the situation where the act En is being performed. [Figure 7] The flowchart shows an example of the support method according to one embodiment of the present invention. [Figure 8] The figure shows an example of the computer 2200 in which the support device 100 according to one embodiment of the present invention may be wholly or partially embodied. 【Embodiments for Carrying Out the Invention】 【0018】 Hereinafter, the present invention will be described through embodiments of the invention. However, the following embodiments do not limit the invention according to the claims. Also, not all combinations of features described in the embodiments are essential for the solution means of the invention. 【0019】 FIG. 1 is a diagram showing an example of a situation before a living body 120 acts on a target person 110 and a situation where the living body 120 is acting. The target person 110 is in a state where it is difficult to communicate, for example, due to illness, disability, or other reasons. The state where it is difficult to communicate may include cases where it is difficult to express intentions through physical movements due to difficulties in physical movements. The living body 120 is a living being that can affect the potential state of the target person 110 by acting on the target person 110. The living body 120 may be a human, or may be an animal such as a dog or a cat. In this example, the living body 120 is a human. 【0020】 Let the action on the target person 110 be the action En. The action En may refer to an action that affects the potential state of the target person 110, such as talking to the target person 110, showing a gesture, making contact, or an action that changes the environment around the target person 110, such as playing music or changing the air conditioning. In the example of FIG. 1, the human living body 120 is talking to the hospitalized target person 110, saying "I came to visit you." 【0021】 Let the state of the target person 110 be the state S. The state S may be the potential state of the target person 110. The potential state of the target person 110 is the psychological state of the target person 110 that the target person 110 is not aware of by himself / herself. The state presentation unit 30 presents the state S. The state presentation unit 30 may be a display, a monitor, or the like. 【0022】 State S1 is the state S of the target person 110 before the organism 120 performs the intervention En. State S2 is the state S of the target person 110 after the organism 120 performs the intervention En. State S2 is the state S of the target person 110 after the organism 120 performs the intervention En. The dashed lines in Figure 1 show the states S1 and S2 of the target person 110. In the example in Figure 1, before the organism 120 performs the intervention En, the target person 110 feels, "I'm always alone, it's boring." In the example in Figure 1, after the organism 120 performs the intervention En, the target person 110 feels, "I'm happy that everyone came to see me today." In the example in Figure 1, the intervention En is the organism 120, being a human, showing itself to the target person 110 and saying, "I came to visit you." 【0023】 Figure 2 is a block diagram showing an example of a support device 100 according to one embodiment of the present invention. The support device 100 comprises an information acquisition unit 10 and a discrimination unit 20. The support device 100 may also comprise a state presentation unit 30, a state learning unit 40, a storage unit 50, and a control unit 90. The information acquisition unit 10 may include a recognition unit 12. 【0024】 Part or all of the support device 100 may be implemented by a computer. The control unit 90 may be the CPU (Central Processing Unit) of the computer. If the support device 100 is implemented by a computer, the computer may have a program installed to make the computer function as the support device 100, and may also have a program installed to execute the support method described later. 【0025】 The electroencephalogram (EEG) information of the subject person 110 is denoted as EEG information Ib. The information acquisition unit 10 acquires EEG information Ib of the subject person 110. EEG information Ib may be information that reproduces at least a portion of the temporal waveform of the subject person 110's brainwaves. EEG information Ib may include data sampled from the temporal waveform of the brainwaves, may include data indicating the magnitude of frequency components of the brainwaves at one or more frequencies, and may include other data. For example, EEG information Ib may include data indicating the magnitude of at least one component of alpha waves, beta waves, theta waves, delta waves, and gamma waves. 【0026】 Alpha waves may be further classified into high alpha waves, medium alpha waves, and low alpha waves depending on their frequency band. Beta waves may be classified into high beta waves and low beta waves. EEG information Ib may include data indicating the magnitude of at least one of the high alpha waves, medium alpha waves, and low alpha waves. EEG information Ib may also include data indicating the magnitude of at least one of the high beta waves and low beta waves. 【0027】 EEG information Ib may include information on the time waveform of one or more brain waves measured at one or more locations on the head, including the head and face of the subject person 110. For example, EEG information Ib may be obtained by measuring the time waveform of the potential of electrodes placed at equal intervals near the scalp of the subject person 110, as in the International 10-20 method, or by other methods. The electrodes placed on the scalp do not have to be at equal intervals. The electrodes may be provided on a wearable device attached to the head of the subject person 110, such as a headgear, headphones, earphones, or glasses. EEG information Ib may also be information obtained by wireless communication from electrical signals at electrodes implanted in the body of the subject person 110. In the example in Figure 1, EEG information Ib is transmitted wirelessly to the control unit 90. 【0028】 When the subject person 110 is in state S1, the electroencephalogram (EEG) information Ib of the subject person 110 is denoted as EEG information Ib1 (see Figure 1). When the subject person 110 is in state S2, the EEG information Ib of the subject person 110 is denoted as EEG information Ib2 (see Figure 1). The information acquisition unit 10 acquires EEG information Ib2. The information acquisition unit 10 may also acquire EEG information Ib1. The discrimination unit 20 determines the state S2 of the subject person 110 based on EEG information Ib2. The discrimination unit 20 may also determine the state S1 of the subject person 110 based on EEG information Ib1. 【0029】 The state presentation unit 30 may present the state S determined by the discrimination unit 20. The state presentation unit 30 may, for example, present an avatar of the target person 110 and present state S2 through the avatar's facial expressions, movements, etc. In the example in Figure 1, the state presentation unit 30 presents state S2 of the target person 110, who is feeling, "I'm happy that everyone came to see me today." This allows the living organism 120 to recognize state S2 of the target person 110, with whom dialogue is difficult. In the example in Figure 1, the living organism 120 feels, "Ah, they're happy," by recognizing state S2 of the target person 110. In the example in Figure 1, state S2 is presented through the facial expressions of the virtual animal displayed on the state presentation unit 30. 【0030】 The discrimination unit 20 may determine state S1 or state S2 based on the magnitude of specific frequency components of the brainwaves of the subject person 110. The discrimination unit 20 may determine state S1 or state S2 based on the magnitude of one or more components among alpha waves, beta waves, theta waves, delta waves, and gamma waves. 【0031】 The total amplitude As is defined as the sum of the amplitudes of alpha waves (8Hz to less than 14Hz), beta waves (14Hz to less than 26Hz), theta waves (4Hz to less than 8Hz), gamma waves (26Hz to less than 40Hz), and delta waves (less than 4Hz) at a given time. For example, if the proportion of delta wave amplitude in the total amplitude As of subject 110 is greater than the proportion of alpha wave amplitude, beta wave amplitude, theta wave amplitude, and gamma wave amplitude, subject 110 may be presumed to be in a sleep state. For example, if the proportion of theta wave amplitude in the total amplitude As of subject 110 during the meeting is greater than the proportion of theta wave amplitude in the total amplitude As of subject 110 before the meeting, it may be presumed that subject 110's fatigue and drowsiness are increasing. 【0032】 For example, if the proportion of the sum of the amplitudes of low alpha waves (8Hz to less than 10Hz) and medium alpha waves (10Hz to less than 12Hz) in subject 110 to the total amplitude As increases over time, it can be inferred that subject 110's level of relaxation is increasing. 【0033】 For example, if the proportion of the sum of the amplitudes of high alpha waves (12Hz to less than 14Hz) and low beta waves (14Hz to less than 18Hz) in subject 110 to the total amplitude As increases over time, it can be inferred that subject 110 is increasingly in a state of good balance between relaxation and concentration. A state of good balance between relaxation and concentration is what is commonly known as a state of immersion. 【0034】 The degree of relaxation of subject 110 is more likely to be higher the larger the proportion of the sum of the amplitudes of low alpha waves and medium alpha waves in the total amplitude As of subject 110. For this reason, the discrimination unit 20 may determine that subject 110 is more relaxed the larger the proportion of the sum of the amplitudes of low alpha waves and medium alpha waves in the total amplitude As of brainwave information Ib2. The degree of immersion of subject 110 is more likely to be higher the larger the proportion of the sum of the amplitudes of high alpha waves and low beta waves in the total amplitude As of subject 110. For this reason, the discrimination unit 20 may determine that subject 110 is more immersed the larger the proportion of the sum of the amplitudes of high alpha waves and low beta waves in the total amplitude As of brainwave information Ib2. 【0035】 The discrimination unit 20 may determine that the subject 110's sense of security has increased if the ratio of the sum of the amplitudes of low alpha waves and medium alpha waves of the subject 110 to the total amplitude As after intervention En is greater than the ratio of the sum of the amplitudes of low alpha waves and medium alpha waves of the subject 110 to the total amplitude As before intervention En. The discrimination unit 20 may evaluate that the subject 110's level of immersion has increased if the ratio of the sum of the amplitudes of high alpha waves and low beta waves of the subject 110 to the total amplitude As after intervention En is greater than the ratio of the sum of the amplitudes of high alpha waves and low beta waves of the subject 110 to the total amplitude As before intervention En. 【0036】 The discrimination unit 20 may combine multiple components from the alpha, beta, theta, delta, and gamma waves of the subject person 110 to determine at least one of state S1 and state S2. For example, the larger the value obtained by dividing the magnitude of the alpha waves of the subject person 110 by the magnitude of the beta waves, the higher the probability that the subject person 110 is relaxed. Therefore, the discrimination unit 20 may determine that the larger the value obtained by dividing the magnitude of the alpha waves of the subject person 110 by the magnitude of the beta waves, the higher the degree of relaxation of the subject person 110. 【0037】 The electroencephalogram (EEG) information Ib1 may reflect the latent state S1 of the subject person 110. The EEG information Ib2 may reflect the latent state S2 of the subject person 110. The discrimination unit 20 discriminates the state S2 of the subject person 110 based on the EEG information Ib2. The support device 100 can discriminate the latent state S2 of the subject person 110. The state presentation unit 30 presents the state S2, allowing the living organism 120 to recognize the state S2 of the subject person 110. 【0038】 Figure 3 shows an example of the information acquisition unit 10. The information acquisition unit 10 may have an electroencephalograph (EEG) capable of measuring EEG information Ib, and may also have a communication device that acquires EEG information Ib measured by an external EEG. In this example, the information acquisition unit 10 is a headgear-type EEG. The information acquisition unit 10 may also be an earphone-type EEG. In this example, the living organism 120 interacts with the subject person 110, who is wearing a headgear-type or earphone-type EEG, by performing an action En. As a result, the information acquisition unit 10 acquires EEG information Ib2 when interacting with the subject person 110 by performing an action En. 【0039】 The discrimination unit 20 and the control unit 90 (see Figure 2) may or may not be housed in the housing of a headgear-type electroencephalograph. If the discrimination unit 20 and the control unit 90 are not housed in the housing, the electroencephalogram information Ib2 acquired by the information acquisition unit 10 may be transmitted wirelessly to the control unit 90. 【0040】 The status display unit 30 may or may not be housed in the housing of the headgear shown in Figure 3. If the status display unit 30 is not housed in the housing of the headgear, it may be a display, monitor, etc., installed separately from the housing of the headgear. Information relating to status S1 or status S2 may be transmitted to the status display unit 30 wirelessly. 【0041】 The biometric information of subject 110 is denoted as biometric information Ig. Biometric information Ig may include at least one of subject 110's heart rate information, sweat amount information, and body temperature information. The biometric information Ig of subject 110 may be acquired by a sensor provided on a wearable device worn by subject 110 (for example, a headgear-type electroencephalograph shown in Figure 3). 【0042】 The information acquisition unit 10 (see Figure 2) may further acquire the biological information Ig of the target person 110. The information acquisition unit 10 may acquire the biological information Ig when the biological body 120 interacts with the target person 110 and performs an action En. The discrimination unit 20 (see Figure 2) may determine the state S2 of the target person 110 based on the electroencephalogram information Ib2 and the biological information Ig. The biological information Ig tends to reflect the target person 110's latent state S. For example, if the target person 110 is experiencing stress, the sympathetic nervous system is likely to be more dominant than the parasympathetic nervous system. When the sympathetic nervous system is more dominant than the parasympathetic nervous system, the target person 110's heart rate variability tends to be smaller and the amount of sweating tends to be larger. Therefore, the discrimination unit 20 can appropriately determine the state S2 of the target person 110 based on the electroencephalogram information Ib and the biological information Ig. 【0043】 The discrimination unit 20 (see Figure 2) may generate state information Is based on electroencephalogram information Ib and biological information Ig. State information Is is information based on the potential state S of the subject person 110. The discrimination unit 20 may generate state information Is based on the change from electroencephalogram information Ib1 to electroencephalogram information Ib2 and biological information Ig. The discrimination unit 20 may determine the state S of the subject person 110 based on the state information Is. 【0044】 Let LF be the magnitude of the first power spectrum and HF be the magnitude of the second power spectrum in the heartbeat of subject 110. The frequency band of the second power spectrum is a higher frequency band than that of the first power spectrum. The frequency bands of the first and second power spectra do not need to overlap. For example, the frequency band of the first power spectrum is 0.04-0.15 Hz. For example, the frequency band of the second power spectrum is 0.15-0.4 Hz. 【0045】 Change C1 is defined as the change from the ratio of the amplitudes of high beta waves (18 to less than 26 Hz) and gamma waves to the total amplitude As in EEG information Ib1 to the ratio of the amplitudes of high beta waves and gamma waves to the total amplitude As in EEG information Ib2. The discrimination unit 20 may generate state information Is based on change C1 and the ratio of LF to HF (LF / HF). 【0046】 For example, if the ratio of the sum of the amplitudes of high beta waves and gamma waves of the target person 110 to the total amplitude As after intervention En is greater than the ratio of the sum of the amplitudes of high beta waves and gamma waves of the target person 110 to the total amplitude As before intervention En, and if the ratio of LF to HF (LF / HF) after intervention En is above a threshold, it can be inferred that the target person 110's irritability, nervousness, or stress state is increasing. In this case, the discrimination unit 20 (see Figure 2) may determine that the target person 110's irritability, nervousness, or stress state is increasing. 【0047】 If the ratio of LF to HF (LF / HF) is greater than or equal to the threshold, subject 110 may be judged to be in a state where the sympathetic nervous system is dominant over the parasympathetic nervous system. If the ratio of LF to HF (LF / HF) is less than the threshold, subject 110 may be judged to be in a state where the parasympathetic nervous system is dominant over the sympathetic nervous system. The threshold may be 2, 3, 4, or 5. 【0048】 For example, if the ratio of the sum of the amplitudes of high beta waves and gamma waves of the target person 110 to the total amplitude As after intervention En is greater than the ratio of the sum of the amplitudes of high beta waves and gamma waves of the target person 110 to the total amplitude As before intervention En, and the ratio of LF to HF (LF / HF) after intervention En is less than a threshold, then it can be inferred that the excitement state of the target person 110 is increasing. In this case, the discrimination unit 20 (see Figure 2) may determine that the excitement state of the target person 110 is increasing. 【0049】 The discrimination unit 20 may generate state information Is based on the relationship between the ratio of LF to HF (LF / HF) after intervention En, the threshold value of the ratio of LF to HF, and the change C1. The threshold value may be predetermined. The discrimination unit 20 (see Figure 2) may generate state information Is indicating that the sense of alertness of the target person 110 has increased if the ratio of the sum of the amplitudes of high beta waves and gamma waves of the target person 110 to the total amplitude As after intervention En is greater than the ratio of the sum of the amplitudes of high beta waves and gamma waves of the target person 110 to the total amplitude As before intervention En, and the ratio of LF to HF (LF / HF) after intervention En is greater than or equal to the threshold value. The discrimination unit 20 may generate state information Is indicating that the excitement level of the target person 110 is increasing if the ratio of the sum of the amplitudes of the high beta waves and gamma waves of the target person 110 to the total amplitude As after intervention En is greater than the ratio of the sum of the amplitudes of the high beta waves and gamma waves of the target person 110 to the total amplitude As before intervention En, and the ratio of LF to HF (LF / HF) after intervention En is less than a threshold. 【0050】 Figure 4 shows an example of state information Is. State information Is may include information relating to multiple states of the subject person 110 (from the first state Is-1 to the nth state Is-n). In this example, state information Is includes information relating to four states of the subject person 110 (from the first state Is-1 to the fourth state Is-4). In Figure 4, the low-frequency f1 brainwave refers to at least one of delta waves, theta waves, low alpha waves, and medium alpha waves, and the high-frequency f2 brainwave refers to at least one of high alpha waves, low beta waves, high beta waves, and gamma waves. 【0051】 Amplitude Af is the amplitude of the brainwaves of subject 110 in a predetermined frequency band. Amplitude Af1 is the amplitude of the brainwaves of subject 110 before intervention En. Amplitude Af2 is the amplitude of the brainwaves of subject 110 after intervention En. The brainwaves in the predetermined frequency band may be at least one of low alpha waves, medium alpha waves, high alpha waves, low beta waves, high beta waves, gamma waves, and theta waves. 【0052】 The discrimination unit 20 (see Figure 2) may generate state information Is based on the change from the ratio of amplitude Af1 to the total amplitude As to the ratio of amplitude Af2 to the total amplitude As, and the ratio of LF to HF (LF / HF). This state information Is may be state information Is related to one of the multiple states of the target person 110 (any of the first state Is-1 to the nth state Is-n). 【0053】 In this example, the first state Is-1 is the state of subject 110 when, in the electroencephalogram of low frequency f1, the ratio of amplitude Af2 to the total amplitude As is greater than the ratio of amplitude Af1 to the total amplitude As, and the ratio of LF to HF (LF / HF) after intervention En is above a threshold. When subject 110 is in the first state Is-1, it can be inferred that subject 110's fatigue and drowsiness are increasing. When subject 110 is in the first state Is-1, the discrimination unit 20 (see Figure 2) may generate state information Is indicating that subject 110's fatigue and drowsiness are increasing. 【0054】 In this example, the second state Is-2 is the state of subject 110 when, in the low-frequency f1 electroencephalogram, the ratio of amplitude Af2 to the total amplitude As is greater than the ratio of amplitude Af1 to the total amplitude As, and the ratio of LF to HF (LF / HF) after intervention En is below a threshold. When subject 110 is in the second state Is-2, it can be inferred that the relaxation level of subject 110 is increasing. When subject 110 is in the second state Is-2, the discrimination unit 20 (see Figure 2) may generate state information Is indicating that the level of reassurance of subject 110 is increasing. 【0055】 In this example, the third state Is-3 is the state of subject 110 when, in the high-frequency f2 electroencephalogram, the ratio of amplitude Af2 to the total amplitude As is greater than the ratio of electroencephalogram Af1 to the total amplitude As, and the ratio of LF to HF (LF / HF) after intervention En is above a threshold. When subject 110 is in the third state Is-3, it can be inferred that subject 110's irritability, nervousness, or stress state is increasing. When subject 110 is in the third state Is-3, the discrimination unit 20 (see Figure 2) may generate state information Is indicating that subject 110's irritability, nervousness, or stress state is increasing. 【0056】 In this example, the fourth state Is-4 is the state of subject 110 when, in the high-frequency f2 electroencephalogram, the ratio of amplitude Af2 to the total amplitude As is greater than the ratio of electroencephalogram Af1 to the total amplitude As, and the ratio of LF to HF (LF / HF) after intervention En is below a threshold. When subject 110 is in the fourth state Is-4, it can be inferred that subject 110's state of immersion is increasing. When subject 110 is in the fourth state Is-4, the discrimination unit 20 (see Figure 2) may generate state information indicating that subject 110's state of immersion is increasing. 【0057】 Figure 5A shows an example of learning by the learning unit 40. When an action En is performed on a target person 110 with electroencephalogram (EEG) information Ib1, and the target person 110 enters state S2, the learning unit 40 learns the relationship between the EEG information Ib1, the action En, and state S2 using machine learning. By learning the relationship between the EEG information Ib1, the action En, and state S2, the learning unit 40 generates an action inference model 42. 【0058】 Figure 5B shows an example of inference by the intervention inference model 42. The intervention inference model 42 in Figure 5B has been trained to understand the relationship between electroencephalogram (EEG) information Ib1, intervention En, and state S2. In the inference stage shown in Figure 5B, the EEG information Ib1 of the subject person 110 is denoted as EEG information Ib1', and the predetermined state S2 of the subject person 110 is denoted as state S2'. State S2' may be a desired state for the subject person 110 from the perspective of the organism 120. This desired state may be, for example, the second state Is-2. State S2' may also be a desired state for the subject person 110. 【0059】 The intervention inference model 42 infers an intervention En' to bring the target person 110 to state S2' based on the electroencephalogram (EEG) information Ib1' and state S2'. Since the intervention inference model 42 has learned the relationship between the EEG information Ib1, the intervention En, and state S2 through machine learning, it can infer an intervention En to bring the target person 110 to state S2' based on the EEG information Ib1' and state S2'. As a result, the user of the support device 100 can infer an intervention En' that has a high probability of bringing the target person 110 to state S2'. The intervention inference model 42 may be stored in the memory unit 50 (see Figure 2). 【0060】 The learning unit 40 (see Figure 2) may learn the relationship between electroencephalogram information Ib1, intervention En, and state S2 for multiple living organisms 120. The learning unit 40 may generate an intervention inference model 42 by learning the relationship between electroencephalogram information Ib1, intervention En, and state S2 for multiple living organisms 120. 【0061】 The support device 100 may include a recognition unit 12 (see Figure 2) that recognizes living organisms 120. The recognition unit 12 may be included in the information acquisition unit 10. The recognition unit 12 may be, for example, an image sensor or a microphone. If the recognition unit 12 is an image sensor, the recognition unit 12 distinguishes one living organism 120 from another living organism 120 using the captured image. If the recognition unit 12 is a microphone, the recognition unit 12 distinguishes one living organism 120 from another living organism 120 using the frequency of the voice of one living organism 120 and the frequency of the voice of the other living organism 120. 【0062】 The learning unit 40 may use machine learning to learn the relationship between the electroencephalogram information Ib1 and the intervention En and the state S2 for the living organism 120 recognized by the recognition unit 12. The learning unit 40 may also generate an intervention inference model 42 for each living organism 120. 【0063】 Figure 6 shows an example of a situation in which the living organism 120 interacts with the target person 110 through interaction En'1 and interaction En'2. In this example, the learning unit 40 (see Figure 2) is learning the relationship between electroencephalogram information Ib1, interaction En, and state S2. In this example, interaction En'1 is when the living organism 120, which is a human, shows itself to the target person 110 and says, "I came to visit you." In this example, the target person 110, who has received interaction En'1, feels, "Thank you for coming. But today is kind of boring..." The discrimination unit 20 discriminates this state S2 of the target person 110. In this example, the state presentation unit 30 presents the state S2 using the facial expressions of a virtual animal. 【0064】 The state presentation unit 30 may present an action En'2 that can bring the target person 110 into state S2'. This allows the living organism 120 to recognize the action En' that will bring the target person 110 into a predetermined state S2'. In this example, the action En'2 is the gifting of a souvenir. In this example, the living organism 120 gives the souvenir to the target person 110 while saying, "I brought you your favorite, ○○." The learning unit 40 (see Figure 2) has learned the relationship between the electroencephalogram information Ib1 and the action En and state S2. Therefore, the living organism 120 has a high probability that the action En' will bring the target person 110 into state S2'. 【0065】 The information acquisition unit 10 (see Figure 2) may acquire electroencephalogram (EEG) information Ib2 after the living organism 120 has performed an action En'. The discrimination unit 20 (see Figure 2) may determine the state S2' of the target person 110 based on the EEG information Ib2. The state presentation unit 30 may present the state S2' of the target person 110, which has been determined by the discrimination unit 20 (see Figure 2). In this example, the state presentation unit 30 presents the state S2' of the target person 110, which is "You brought me ○○, I'm so happy." As a result, the living organism 120 can recognize the state S2' of the target person 110. In this example, by recognizing the state S2' of the target person 110, the living organism 120 feels, "Oh, they look so happy, so happy." In this example, the state S2' is presented by the facial expression of a virtual animal displayed on the state presentation unit 30. The living organism 120 can confirm whether the intervention En' presented by the state presentation unit 30 was appropriate by recognizing the state S2' of the target person 110. 【0066】 The information acquisition unit 10 (see Figure 2) may acquire attribute information indicating the attributes of the living organism 120. This attribute information is referred to as attribute information Ia. If the living organism 120 is a human, attribute information Ia may include information regarding at least one of the human's age, gender, occupation, and preferences. Attribute information Ia may also include information regarding whether the living organism 120 is a relative of the target person 110. If the living organism 120 is a relative of the target person 110, attribute information Ia may include information regarding the degree of kinship between the target person 110 and the living organism 120. 【0067】 The learning unit 40 (see Figure 2) may learn by machine learning the relationship between the attribute information Ia of the living organism 120, the electroencephalogram (EEG) information Ib1, the intervention En by the living organism 120, and the state S2. The intervention inference model 42 infers the intervention En' to bring the target person 110 to state S2' based on the EEG information Ib1', the state S2', and the attribute information Ia. When the EEG information Ib1' of the target person 110 is the same EEG information Ib1, the intervention En' to bring the target person 110 to state S2' may differ depending on the attributes of the living organism 120. Since the intervention inference model 42 has learned by machine learning the relationship between the attribute information Ia, the EEG information Ib1, the intervention En, and the state S2, the intervention En to bring the target person 110 to state S2' can be inferred based on the EEG information Ib1', the state S2', and the attribute information Ia. 【0068】 The state presentation unit 30 may present the intervention En' inferred by the intervention inference model 42 for each attribute of the living organism 120. For example, the state presentation unit 30 presents different intervention En' depending on whether the living organism 120 is a nurse or a visitor such as a friend of the target person 110. This allows the living organism 120 to determine the optimal intervention En' according to its own attributes. 【0069】 If the recognition unit 12 (see Figure 2) is an image sensor, the recognition unit 12 may acquire attribute information Ia based on the captured facial image of the living organism 120. If the recognition unit 12 is a microphone, the recognition unit 12 may acquire attribute information Ia based on the frequency of the voice of the living organism 120. 【0070】 The state presentation unit 30 (see Figure 2) may present the state S of the target person 110 based on the attribute information Ia. The state presentation unit 30 may choose whether or not to present the state S of the target person 110 based on the attribute information Ia. The storage unit 50 may store attribute information Ia that may or may not present the state S. 【0071】 The state presentation unit 30 (see Figure 2) may change the presentation method of the state S of the target person 110 based on attribute information Ia. For example, if the state S of the target person 110 is a first state Is-1 (e.g., fatigue) or a third state Is-3 (e.g., irritation), and the living body 120 is a nurse, the state presentation unit 30 may present the first state Is-1 or the third state Is-3 of the target person 110 to the living body 120. When the living body 120 is a nurse, it is preferable that the living body 120 accurately recognizes the state S of the target person 110. For this reason, the state presentation unit 30 may present the first state Is-1 or the third state Is-3 of the target person 110 to the living body 120. 【0072】 For example, if the state S of the subject person 110 is the first state Is-1 (e.g., fatigue) or the third state Is-3 (e.g., irritation), and the living organism 120 is a visitor, the state presentation unit 30 may or may not present the first state Is-1 or the third state Is-3 of the subject person 110 to the living organism 120 in a discreet manner. In this case, the visitor may refer to a visitor other than a relative, such as a friend of the subject person 110. A discreet presentation would be, for example, to make the expression of the virtual animal displayed on the state presentation unit 30 expressionless, or to make it look listless, or not to change its expression at all. If the living organism 120 is a visitor, the living organism 120 may be shocked by accurately recognizing the first state Is-1 or the third state Is-3 of the subject person 110. Therefore, the status display unit 30 may discreetly present the first status Is-1 or third status Is-3 of the target person 110 to the living body 120, or it may not present it at all. The status display unit 30 may also present the first status Is-1 or third status Is-3 of the target person 110 to someone other than the living body 120, for example, a nurse. 【0073】 The discrimination unit 20 (see Figure 2) may determine the attributes of the biological organism 120 based on the biological organism 120 recognized by the recognition unit 12 (see Figure 2). The information acquisition unit 10 (see Figure 2) may acquire attribute information Ia based on the determined attributes. The discrimination unit 20 may determine whether to present the state S of the target person 110 to the biological organism 120 based on the attribute information Ia. If the discrimination unit 20 determines that state S should be presented, the state presentation unit 30 may present the state S of the target person 110 to the biological organism 120. 【0074】 The discrimination unit 20 may determine the manner in which the state S is presented based on the attribute information Ia. The manner of presentation refers to presenting the state S as is, presenting it discreetly, not presenting it, etc. For example, if the discrimination unit 20 determines that the living organism 120 is a nurse or a relative, it may determine that the state S is presented as is. For example, if the discrimination unit 20 determines that the living organism 120 is a friend of the target person 110, it may determine that the state S is presented discreetly or not presented at all. 【0075】 Figure 7 is a flowchart illustrating an example of a support method according to one embodiment of the present invention. A support method according to one embodiment of the present invention will be explained using the support device 100 shown in Figure 2 as an example. The support method comprises an information acquisition step S100 and a discrimination step S104. The support method may also include an information acquisition step S90, a state learning step S102, and a state presentation step S106. 【0076】 The information acquisition step S100 is a step in which the information acquisition unit 10 acquires brainwave information Ib of the target person 110 when the biological device 120 interacts with the target person 110 in a state where dialogue is difficult. The discrimination step S104 is a step in which the discrimination unit 20 determines the state S of the target person 110 based on the brainwave information Ib acquired in the information acquisition step S100. 【0077】 The information acquisition step S100 may be a step in which the information acquisition unit 10 further acquires the biological information Ig of the target person 110. The discrimination step S104 may be a step in which the state S of the target person 110 is determined based on the electroencephalogram information Ib and biological information Ig acquired in the information acquisition step S100. 【0078】 The information acquisition step S90 is a step in which the information acquisition unit 10 acquires electroencephalogram (EEG) information Ib1 of the target person 110 before the intervention En. The discrimination step S104 may be a step in which the discrimination unit 20 generates state information Is indicating the state S of the target person 110 based on the change from the EEG information Ib1 before the intervention En to the EEG information Ib2 after the intervention En and the biological information Ig, and then discriminates the state S of the target person 110 based on the generated state information Is. 【0079】 The discrimination step S104 may be a step in which the discrimination unit 20 generates state information Is based on the change from the ratio of the amplitude of brain waves in a predetermined frequency band to the total amplitude As in the brainwave information Ib1 before intervention En, to the ratio of the amplitude of brain waves in the said frequency band to the total amplitude As in the brainwave information Ib2 after intervention En, and the ratio of LF to HF (LF / HF) in the heartbeat of the subject person 110. 【0080】 The discrimination step S104 may be a step in which the discrimination unit 20 generates state information Is based on the change from the ratio of the amplitude of the brain waves in a predetermined frequency band to the total amplitude As in the brain wave information Ib1 before the intervention En, to the ratio of the amplitude of the brain waves in the said frequency band to the total amplitude As in the brain wave information Ib2 after the intervention En, and the relationship between the ratio of LF to HF (LF / HF) after the intervention En and a predetermined threshold for the ratio of LF to HF. 【0081】 The state information Is may include information relating to multiple states of the target person 110. The discrimination step S104 may be a step in which the discrimination unit 20 generates state information Is relating to one of the multiple states based on the change from the ratio of the amplitude of the brain waves in a predetermined frequency band to the total amplitude As in the brainwave information Ib1 before intervention En, to the ratio of the amplitude of the brain waves in the said frequency band to the total amplitude As in the brainwave information Ib2 after intervention En, and the ratio of the magnitude of the first power spectrum to the magnitude of the second power spectrum. 【0082】 The state learning step S102 is a step in which the state learning unit 40 generates an action inference model 42. The state learning step S102 is a step in which the state learning unit 40 generates an action inference model 42 that infers an action En' to bring the target person 110 to state S2' based on the brainwave information Ib1' and state S2' by machine learning the relationship between the brainwave information Ib1 and the action En and state S2. 【0083】 The discrimination step S104 may be a step in which the discrimination unit 20 determines an action En to bring the target person 110 to a predetermined state, based on the state S of the target person 110 inferred in the state learning step S102. 【0084】 The state presentation step S106 is a step in which the state presentation unit 30 presents the state S of the target person 110 that was identified in the discrimination step S104. The information acquisition step S100 may be a step in which the information acquisition unit 10 further acquires attribute information Ia indicating the attributes of the biological organism 120. The state presentation step S106 may be a step in which the state presentation unit 30 presents the state S of the target person 110 based on the attribute information Ia. The state presentation step S106 may also be a step in which the state presentation unit 30 changes the presentation mode of the state S of the target person 110 based on the attribute information Ia. 【0085】 Figure 8 shows an example of a computer 2200 in which an assist device 100 according to one embodiment of the present invention may be fully or partially embodied. A program installed on the computer 2200 can cause the computer 2200 to function as an operation associated with the assist device 100 according to an embodiment of the present invention, or as one or more sections of the assist device 100, or to execute such operation or one or more sections, or to cause the computer 2200 to execute each step (see Figure 7) of the method of the present invention. The program may be executed by the CPU 2212 to cause the computer 2200 to perform a specific operation associated with some or all of the blocks in the flowchart (Figure 7) and block diagram (Figure 2) described herein. 【0086】 A computer 2200 according to one embodiment of the present invention includes a CPU 2212, RAM 2214, a graphics controller 2216, and a display device 2218. The CPU 2212, RAM 2214, graphics controller 2216, and display device 2218 are interconnected by a host controller 2210. The computer 2200 further includes input / output units such as a communication interface 2222, a hard disk drive 2224, a DVD-ROM drive 2226, and an IC card drive. The communication interface 2222, hard disk drive 2224, DVD-ROM drive 2226, and IC card drive are connected to the host controller 2210 via an input / output controller 2220. The computer further includes legacy input / output units such as a ROM 2230 and a keyboard 2242. The ROM 2230 and keyboard 2242 are connected to the input / output controller 2220 via an input / output chip 2240. 【0087】 The CPU 2212 controls each unit by operating according to programs stored in the ROM 2230 and RAM 2214. The graphics controller 2216 retrieves the image data generated by the CPU 2212 and places it in the frame buffer or other location provided in RAM 2214, or in RAM 2214 itself, so that the image data is displayed on the display device 2218. 【0088】 The communication interface 2222 communicates with other electronic devices via a network. The hard disk drive 2224 stores programs and data used by the CPU 2212 in the computer 2200. The DVD-ROM drive 2226 reads programs or data from the DVD-ROM 2201 and provides the read programs or data to the hard disk drive 2224 via the RAM 2214. The IC card drive reads programs and data from or writes programs and data to the IC card. 【0089】 ROM2230 stores boot programs executed by computer 2200 upon activation, or programs that depend on the computer 2200's hardware. The input / output chip 2240 may connect various input / output units to the input / output controller 2220 via parallel ports, serial ports, keyboard ports, mouse ports, etc. 【0090】 The program is provided on a computer-readable medium such as a DVD-ROM 2201 or an IC card. The program is read from the computer-readable medium and installed on a hard disk drive 2224, RAM 2214, or ROM 2230, which are also examples of computer-readable medium, and executed by the CPU 2212. The information processing described within these programs is read by the computer 2200, resulting in coordination between the program and the various types of hardware resources described above. The apparatus or method may be configured to realize the manipulation or processing of information in accordance with the use of the computer 2200. 【0091】 For example, when communication is performed between a computer 2200 and an external device, the CPU 2212 may execute a communication program loaded into RAM 2214 and instruct the communication interface 2222 to perform communication processing based on the processing described in the communication program. Under the control of the CPU 2212, the communication interface 2222 reads transmission data stored in a transmission buffer processing area provided in a recording medium such as RAM 2214, a hard disk drive 2224, a DVD-ROM 2201, or an IC card, transmits the read transmission data to the network, or writes received data received from the network to a reception buffer processing area provided on the recording medium. 【0092】 The CPU 2212 may read all or necessary parts of a file or database stored on an external recording medium such as a hard disk drive 2224, a DVD-ROM drive 2226 (DVD-ROM 2201), or an IC card into the RAM 2214. The CPU 2212 may perform various types of processing on the data in the RAM 2214. The CPU 2212 may then write the processed data back to the external recording medium. 【0093】 Various types of information, such as various types of programs, data, tables, and databases, may be stored on the recording medium and processed. The CPU 2212 may perform various types of processing on the data read from the RAM 2214, including various types of operations, information processing, conditional decisions, conditional branching, unconditional branching, information retrieval, or replacement, as specified by the program instruction sequence described in this disclosure. The CPU 2212 may write the results back to the RAM 2214. 【0094】 CPU2212 may search for information in files, databases, etc., within the recording medium. For example, if multiple entries are stored in the recording medium, each having an attribute value of a first attribute associated with the attribute value of a second attribute, CPU2212 may search among the multiple entries for an entry that matches the specified condition for the attribute value of the first attribute, read the attribute value of the second attribute stored within that entry, and thereby obtain the attribute value of the second attribute associated with the first attribute that satisfies a predetermined condition. 【0095】 The program or software module described above may be stored on or on a computer-readable medium of the computer 2200. A recording medium such as a hard disk or RAM provided within a server system connected to a dedicated communication network or the Internet can be used as a computer-readable medium. The program may be provided to the computer 2200 via such a recording medium. 【0096】 Although the present invention has been described above using embodiments, the technical scope of the present invention is not limited to the scope described in the above embodiments. It will be apparent to those skilled in the art that various modifications or improvements can be made to the above embodiments. It will be clear from the claims that such modified or improved forms may also be included in the technical scope of the present invention. 【0097】 It should be noted that the execution order of operations, procedures, steps, and stages in the apparatus, systems, programs, and methods shown in the claims, specifications, and drawings is not explicitly stated as "before," "prior to," etc., and that these can be implemented in any order unless the output of a previous process is used in a later process. Even if the operation flow in the claims, specifications, and drawings is described using phrases such as "first," "next," etc. for convenience, it does not mean that it is essential to perform the operations in that order. [Explanation of Symbols] 【0098】 10...Information acquisition unit, 12...Recognition unit, 20...Discrimination unit, 30...State presentation unit, 40...Learning unit, 42...Interactive inference model, 50...Memory unit, 90...Control unit, 100...Support device, 110...Target person, 120...Biological device, 2200...Computer, 2201...DVD-ROM, 2210...Host controller, 2212...CPU, 2214...RAM, 2216...Graphics controller, 2218...Display device, 2220...Input / output controller, 2222...Communication interface, 2224...Hard disk drive, 2226...DVD-ROM drive, 2230...ROM, 2240...Input / output chip, 2242...Keyboard

Claims

[Claim 1] An information acquisition unit that acquires the brainwave information of a person in a state where dialogue is difficult when a biological device interacts with the person, the brainwave information of the person before the interaction, and biological information including the heart rate of the person. A discrimination unit that determines the state of the subject person based on the electroencephalogram information and the biological information, Equipped with, The aforementioned discrimination unit is The change from the electroencephalogram information before the intervention to the electroencephalogram information after the intervention, The ratio of the magnitude of the first power spectrum in the heartbeat of the subject to the magnitude of the second power spectrum and Based on this, state information indicating the state of the target person is generated, and the state of the target person is determined based on the generated state information. The frequency band of the second power spectrum is a higher frequency band than the frequency band of the first power spectrum. The discrimination unit generates the state information based on the change from the ratio of the amplitude of brain waves in a predetermined frequency band to the total amplitude in the brainwave information before the intervention, to the ratio of the amplitude of brain waves in the frequency band to the total amplitude in the brainwave information after the intervention. The total amplitude is the sum of the amplitudes of alpha waves, beta waves, theta waves, gamma waves, and delta waves. Support equipment. [Claim 2] The discrimination unit generates the state information based on the change, the ratio of the magnitude of the first power spectrum to the magnitude of the second power spectrum after the intervention, and a predetermined threshold value for the ratio of the magnitude of the first power spectrum to the magnitude of the second power spectrum. The support device according to claim 1. [Claim 3] The aforementioned status information includes information relating to multiple states of the subject person, The discrimination unit generates state information relating to one of the plurality of states based on the change and the ratio of the magnitude of the first power spectrum to the magnitude of the second power spectrum. The support device according to claim 1. [Claim 4] The support device according to claim 3, wherein the brainwaves in the frequency band are at least one of delta waves, theta waves, low alpha waves, and medium alpha waves. [Claim 5] The support device according to claim 3, wherein the brainwaves in the aforementioned frequency band are at least one of high alpha waves, low beta waves, high beta waves, and gamma waves. [Claim 6] The system further comprises a state presentation unit that presents the state of the target person identified by the discrimination unit, The information acquisition unit further acquires attribute information indicating the attributes of the living organism, The state presentation unit presents the state of the target person based on the attribute information. The support device according to any one of claims 1 to 5. [Claim 7] The support device according to claim 6, wherein the state presentation unit changes the presentation mode of the state of the target person based on the attribute information. [Claim 8] An information acquisition step in which the information acquisition unit acquires, when a biological device interacts with a target person who is in a state where dialogue is difficult, the target person's brainwave information, the target person's brainwave information before the interaction, and biological information including the target person's heart rate. The discrimination unit performs a discrimination step in which it determines the state of the target person based on the electroencephalogram information and biological information acquired in the information acquisition step, Equipped with, In the aforementioned determination step, The change from the electroencephalogram information before the intervention to the electroencephalogram information after the intervention, The ratio of the magnitude of the first power spectrum in the heartbeat of the subject to the magnitude of the second power spectrum and Based on this, state information indicating the state of the target person is generated, and the state of the target person is determined based on the generated state information. The frequency band of the second power spectrum is a higher frequency band than the frequency band of the first power spectrum. Based on the change from the ratio of the amplitude of brain waves in a predetermined frequency band to the total amplitude in the brainwave information before the intervention, to the ratio of the amplitude of brain waves in the frequency band to the total amplitude in the brainwave information after the intervention, the state information is generated. The total amplitude is the sum of the amplitudes of alpha waves, beta waves, theta waves, gamma waves, and delta waves. How to help. [Claim 9] A support program for causing a computer to perform the support method described in claim 8.

Images