JP2009018047A - Psychological state estimation apparatus - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a psychological state estimation apparatus for highly accurately estimating the psychological state of a subject. <P>SOLUTION: The psychological state estimation apparatus 1 for estimating the psychological state of the subject comprises: a stimulation setting means 5 for setting a stimulation imparting pattern to be given to the subject; stimulation imparting means 2 and 5 for imparting stimulation to the subject on the basis of the stimulation imparting pattern; a biological information acquisition means 3 for acquiring the biological information of the subject; a biological information change acquisition means 6 for acquiring the change of the biological information to the imparted stimulation by the acquired biological information; a learning means 6 for associating the set stimulation imparting pattern with the acquired change of the biological information to the stimulation and learning a parameter for estimating the psychological state of the subject; and estimation means 8, 9 and 11 for estimating the psychological state of the subject on the basis of the learned parameter for estimating the psychological state of the subject. <P>COPYRIGHT: (C)2009,JPO&INPIT

Description

  The present invention relates to a psychological state estimation device for estimating a psychological state of a subject such as an irritated state.

In order to improve safety during driving, various devices for estimating a driver's psychological state such as an irritated state have been proposed. As this estimation apparatus, for example, in the apparatus described in Patent Document 1, biological information such as a heart rate and blood pressure of a subject is detected, and an irritated state or the like is determined by the magnitude relationship between the detected biological information and a preset threshold value. judge. Moreover, in the apparatus described in Patent Document 2, an irritated state or the like is determined in consideration of not only biological information but also road load due to vehicle speed, weather, and the like.
JP 2006-34803 A JP 2003-61939 A JP-A-2005-157762 Japanese Patent Laid-Open No. 2004-24879

  When determining an irritated state based on biometric information, since the determination is based on a preset numerical standard (threshold), there is a possibility that the irritated state is erroneously determined. For example, since the heart rate and blood pressure rise when the frustrated state increases, the heart rate and blood pressure higher than normal are set as threshold values, respectively, and it is determined that the frustrated state is high when the heart rate and blood pressure exceed the respective threshold values. To do. However, the heart rate and blood pressure may increase not only when the annoyance is high, but also when it is in other states such as when laughing out loud. Therefore, even when the frustrated state is not high, it may be determined that the frustrated state is high. As described above, the estimation accuracy of the frustrated state is lowered only by the biological information.

  In addition, when the frustrated state is determined in consideration of the road load amount (external environmental factor) in addition to the biometric information, the response to such an environmental factor is different for each individual. is there. For example, if you encounter a traffic jam of a certain scale, not all drivers will show the same degree of frustration uniformly. Other drivers were in a hurry and can be frustrating. Therefore, in fact, even if the same environmental factors are encountered, the influence on the frustrated state varies depending on the individual situation. Thus, even when environmental factors are added to the biological information, the estimation accuracy of the frustrated state is lowered.

  Then, this invention makes it a subject to provide the psychological state estimation apparatus which estimates a test subject's psychological state with high precision.

  The psychological state estimation device according to the present invention is a psychological state estimation device that estimates the psychological state of a subject, based on a stimulus setting unit that sets a stimulus applying pattern to be given to the subject and a stimulus applying pattern set by the stimulus setting unit. A stimulus applying means for applying a stimulus to the subject, a biological information acquiring means for acquiring the biological information of the subject, and a change in the biological information with respect to the stimulus applied by the stimulus applying means based on the biological information acquired by the biological information acquiring means. Biometric information change acquisition means, learning means for learning the parameter for estimating the psychological state of the subject by associating the stimulus application pattern set by the stimulus setting means with the change of biometric information for the stimulus acquired by the biometric information change acquisition means, An estimation means for estimating the psychological state of the subject based on the parameters for estimating the psychological state of the subject learned by the learning means And features.

  In this psychological state estimation apparatus, a stimulus applying pattern to be given to the subject is set by the stimulus setting means, and the stimulus is given to the subject by the stimulus applying means based on the stimulus applying pattern. This stimulus applying pattern is a pattern that gives a stimulus that makes the subject to the state of the psychological state estimation parameter to be learned, and indicates the type of stimulus, the content of the stimulus, the size of the stimulus, the strength of the stimulus, etc. ing. When a stimulus is applied for learning, the psychological state estimation device acquires the biological information of the subject by the biological information acquisition unit, and the biological information change acquisition unit calculates the change in the biological information with respect to the stimulus applied from the acquired stimulus. get. At this time, since the stimulus that gives the state of the psychological state estimation parameter to be learned according to the stimulus application pattern is given, the change in the biological information of the subject when the state of the psychological state estimation parameter to be learned becomes . Therefore, in the psychological state estimation device, the learning unit learns the parameter for estimating the psychological state of the subject by associating the stimulus application pattern with the change in biological information with respect to the stimulus. In the psychological state estimation device, the psychological state of the subject is estimated based on the learned psychological state estimation parameter by the estimating means. As described above, in this psychological state estimation device, the psychological state estimation criterion for estimating the psychological state of the subject can be set for each subject by setting the psychological state estimation parameter for each subject, and the psychological state of each subject can be set. The state can be estimated with high accuracy. Furthermore, in this psychological state estimation apparatus, learning is performed by actively applying a stimulus by determining a given pattern, so that early and accurate learning can be performed.

  Note that the stimulus is a variety of stimuli that can control the pattern when applied and can be applied to the subject, such as sound and a display screen. The biological information is various information that shows a person's psychological state and can be measured from the person, such as blood pressure, heart rate, and skin potential. Psychological state estimation parameters are various parameters that serve as estimation criteria for a subject's psychological state. For example, the estimation criteria for biological information and behavior information indicating the psychological state, and the cognitive response of the subject when receiving an external event. Judgment criteria. Examples of the psychological state include an irritated state, an impatient state, and a panic state.

  The psychological state estimation apparatus of the present invention includes a cognitive response determination unit that determines a cognitive response to an external event for the subject, and the stimulus setting unit sets a stimulus application pattern such that the subject indicates the cognitive response, and learning The means learns the judgment criteria of the cognitive response to the external event, the cognitive reaction judgment means judges the cognitive response to the external event based on the judgment criteria learned by the learning means, and the estimation means is the cognitive response judgment means It is good also as a structure which estimates a test subject's psychological state based on the cognitive reaction determined by (1).

  In this psychological state estimation apparatus, a stimulus application pattern is set such that the subject indicates a cognitive reaction to be learned by the stimulus setting unit, and a stimulus is applied to the subject based on the stimulus application pattern by the stimulus application unit. In the psychological state estimation device, the learning means learns the criterion for determining the cognitive response to the external event of the subject based on the change in the biological information with respect to the applied stimulus. Furthermore, in the psychological state estimation device, the cognitive response to the external event of the subject is determined based on the learned determination criterion by the cognitive reaction determination unit, and the psychological state of the subject is estimated based on the cognitive response determined by the estimation unit. . As described above, in this psychological state estimation device, the influence of the external event on the psychological state is accurately determined for each subject by learning the cognitive reaction to the external event of each subject and estimating the psychological state of each subject. It can be reflected and the psychological state of each subject can be estimated with high accuracy. Moreover, in this psychological state estimation apparatus, by using the cognitive response to the external event, the cognitive response to the external event appears instantaneously as a response, so that it can cope with an instantaneous change in the psychological state of the subject. .

  External events are various external events (factors) that affect the psychological state of the subject. For example, in the case of a driver, there are traffic jams, interruptions, and road conditions. A cognitive response is a subject's cognitive response to an external event when the subject receives an external event, such as a stereotactic response (a response to a familiar external event and a low response), protective response (Responding to external events and high response), no response (not aware of external events, no response).

  The psychological state estimation apparatus of the present invention includes an influence degree acquisition unit that acquires an influence degree on a change in the psychological state of the subject based on the cognitive reaction determined by the cognitive reaction determination unit, and the estimation unit includes the biological information acquisition unit. It is good also as a structure which estimates a test subject's psychological state based on the biometric information acquired by (1) and the influence degree acquired by the influence degree acquisition means.

  In this psychological state estimation device, the biological information of the subject whose psychological state appears is obtained by the biological information acquisition means. In addition, the psychological state estimation device acquires the degree of influence on the change in the psychological state of the subject based on the cognitive reaction to the external event determined by the influence degree acquisition unit. The psychological state estimation device estimates the psychological state of the subject based on the biological information indicating the psychological state by the estimating means and the degree of influence on the change in the psychological state due to the cognitive reaction to the external event. As described above, in this psychological state estimation device, the individual psychological state of each subject is further estimated by estimating the individual psychological state of the subject in consideration of the individual subject's cognitive reaction to the external event in addition to the individual biometric information of the subject. It can be estimated with high accuracy.

  The present invention learns psychological state estimation parameters for estimating the psychological state of each subject, and can thereby set a psychological state estimation criterion for each subject, and can estimate the individual psychological state with high accuracy. .

  Hereinafter, an embodiment of a psychological state estimation device according to the present invention will be described with reference to the drawings.

  In this Embodiment, the psychological state estimation apparatus which concerns on this invention is mounted in a vehicle, and is applied to the frustration estimation apparatus which estimates a driver | operator's frustration level. In the frustration estimation apparatus according to the present invention, the driver's frustration level is determined in consideration of the environmental factors themselves that affect the driver's frustration state in addition to the biological information representing the frustration state of the driver and the cognitive reaction to the environmental factor. presume. In the frustration estimation device according to the present invention, the estimated frustration level is provided to various driver support devices, alarm devices, frustration mitigation devices, and the like.

  The outline of the present embodiment will be described with reference to FIGS. FIG. 1 is a conceptual diagram of the present embodiment. FIG. 2 is a time change of biological information for each stimulation pattern. In the present embodiment, a criterion for determining a driver's cognitive response (defense response, localization response, no response) to external environmental factors that may increase the driver's frustration is learned. Therefore, the stimulus application patterns (stimulus type, large, strength, speed, content, etc.) that the driver is expected to show each cognitive response are set and intentionally created on the vehicle side according to the stimulus pattern. The driver is given a stimulus (alarm sound, horn, navigation screen, etc.). Then, the driver's biometric information (heart rate, skin potential) at that time is sensed, the feature quantity of the biometric information for each cognitive reaction is learned, and the biometric feature feature quantity learning data is updated. At this time, if the updated learning data of the feature amount satisfies the update condition for the determination criterion of the cognitive reaction, the determination criterion of the cognitive reaction is updated. Initial values are initially set as the criteria for determining the cognitive response.

  It also senses traffic jams, interrupts, and road conditions (winding roads, narrow streets, etc.) as environmental factors. When any environmental factor is detected, the driver's biological information is sensed, and based on each of the criteria for localization reaction, defense reaction, and no reaction stored in the learning database, Determine the driver's cognitive response. When an environmental factor is detected, the frustration influence level (reference value) for the environmental factor itself is set. Then, the final frustration influence degree is calculated by adding the cognitive reaction to the frustration influence degree (reference value). Furthermore, biometric information (heart rate, blood pressure) representing the driver's frustrated state is sensed, and the frustrated level (reference value) is set based on the biological reaction. Then, the final frustration level is estimated by reflecting the frustration influence level on the frustration level (reference value). In addition to biological information, driver behavior information (such as facial expressions, gestures, and vehicle operations) may also be used.

  When a person receives a stimulus such as an environmental factor, a cognitive reaction to the stimulus appears instantly in biological information. Therefore, in the case of cognitive reaction, compared with the case where the irritated state is determined based on the biological information, the section to be determined can be greatly shortened, and is in units of seconds (when the irritated state is determined based on the biological information) Also, the type of biometric information used for determination and the determination criteria are different. In this embodiment, in order to determine the cognitive response, a skin potential response and a heart rate that are likely to generate an instantaneous response are used.

  The defensive reaction is a reaction that warns against a stimulus, is a preparatory stage to an irritated state, and has a high response coefficient. Therefore, it can be predicted that a defense response appears to the driver by giving the driver a stimulus that is not normally given. The stereotaxic reaction is a response to a familiar stimulus, is difficult to shift to an irritated state, and has a low response coefficient. Therefore, it can be predicted that a stereotaxic reaction appears in the driver by giving the driver a normal stimulus. No reaction is a state in which the stimulus is not noticed, there is no influence on the frustrated state, and the response coefficient is zero. Therefore, it can be predicted that no response will appear to the driver when the driver is not stimulated.

  The intentionally applied stimulus is a stimulus that can be applied in the vehicle, and it is preferable to use one provided in the vehicle. When a stimulus is intentionally applied, the stimulus is set so as not to impair the safety by the applied stimulus, and the stimulus is set so as not to impair the merchantability of the equipment to which the stimulus is applied.

  Examples of the stimulus include alarm sound, horn, navigation display screen and sound, various other sounds, and automatic seat position. In the case of an alarm sound and a horn, for example, each cognitive reaction is learned with a normal volume, a volume higher than normal, and no occurrence. Each cognitive reaction is learned by a navigation display screen, for example, a route guidance screen display at a normal timing, a route guidance screen display at a timing earlier than normal, and no route guidance screen display. In the case of the navigation voice, for example, each cognitive reaction is learned without a route guidance voice output at a normal timing, a route guidance voice output at a timing earlier than usual, and no route guidance voice output. In the case of the automatic seat position, the sheet movement is performed at a normal change speed, the sheet movement is performed at a faster change speed than usual, and each cognitive reaction is learned after the sheet movement.

  FIG. 2 shows an example of changes in biological information when an alarm sound is used as a stimulus. When the alarm sound indicated by the symbol A is not generated, the biological information hardly changes. When an alarm sound having a normal volume indicated by the symbol B is generated, the biological information changes slightly. When an alarm sound having a louder volume than indicated by the symbol C is generated, the biological information changes greatly.

  The frustration level estimated in the present embodiment is a value from 0 to 2, may be an integer value (0, 1, 2), or may not be an integer value. The frustration level 0 is a state where there is no awareness of frustration and is a normal calm state. The frustration level 1 is a state in which the awareness of the frustration is small, and it is a state that makes it stuffy. The frustrating level 2 is a state in which the awareness of frustration is large, endures anger, and is a state of immediate touch. When the level is up to about this frustrating level 2, misjudgments and sudden operations are likely to occur during driving. Incidentally, the frustrating level 3 is a state in which anger is expelled. You may make it estimate to this irritation level 3.

The frustration estimation device will be described with reference to FIGS. FIG. 3 is a configuration diagram of the frustration estimation apparatus according to the present embodiment. FIG. 4 is an example of each stimulation pattern. FIG. 5 is an example of a time change of biometric information (skin electricity, heart rate) for each stimulus pattern of the driver. FIG. 6 is an example of a change over time in the determination section of the skin potential in response to a strong driver's stimulus. FIG. 7 is an example of a change over time in a heart rate determination interval in response to a strong driver's stimulus. FIG. 8 is an example of a frequency distribution table of the temporal change amount of the biological information with respect to the stimulus pattern. FIG. 9 is an example of learned determination criteria (reference range of biometric information (skin potential, heart rate) for each cognitive reaction). FIG. 10 is an example of a time change of biometric information when a defense response is shown to the driver's environmental factors, (a) is a time change of skin potential, and (b) is a time change of heart rate. It is. FIG. 11 is an example of a time change of biometric information when a localization reaction is shown with respect to the driver's environmental factors, (a) is a time change of skin potential, and (b) is a time change of heart rate. The frustration estimation device 1 estimates the frustration level based on the biological information indicating the frustration state and external environmental factors that affect the frustration state. In particular, the frustrated estimation device 1 learns the determination criteria for the driver's cognitive reaction by controlling the stimulus given to the driver in order to improve the estimation accuracy. Therefore, the frustration estimation device 1 includes a stimulus presentation device 2, biological information detection means 3, environment information detection means 4, stimulus control means 5, cognitive reaction learning means 6, learning data storage device 7, cognitive reaction determination means 8, and frustration. An influence degree calculation means 9, a reference data storage device 10, and an irritated level determination means 11 are provided. In particular, for the stimulus control means 5, the cognitive reaction learning means 6, the learning data storage device 7, the cognitive reaction determination means 8, the frustration influence calculation means 9, the reference data storage device 10, and the frustration level determination means 11, the frustration estimation device 1 ECU [Electronic Control Unit].

  In the present embodiment, the stimulus presentation device 2 and the stimulus control means 5 correspond to the stimulus applying means described in the claims, and the biological information detection means 3 corresponds to the biological information detection means described in the claims. The stimulus control means 5 corresponds to the stimulus setting means described in the claims, the cognitive reaction learning means 6 corresponds to the biological information change acquisition means and the learning means described in the claims, and the cognitive reaction determination The means 8 corresponds to the cognitive reaction determination means described in the claims, and the frustration influence calculation means 9 corresponds to the influence acquisition means described in the claims. The cognitive reaction determination means 8 and the frustration influence calculation The means 9 and the frustration level determination means 11 correspond to the estimation means described in the claims.

  The stimulus presentation device 2 is a device that can present various stimuli equipped in the vehicle, and is a device that can be controlled with respect to stimulus application. In the stimulus presentation device 2, when there is an application command from the stimulus control means 5, the stimulus is presented according to the application command. Examples of the stimulus presentation device 2 include an alarm sound generation device, a horn generation device, a navigation device, and an automatic seat position device.

  The biological information detection means 3 is means for detecting skin potential (biological information that electrically indicates mental sweating), heart rate, and blood pressure as biological information. Specifically, a skin potential sensor, a heart rate sensor (which may be an electrocardiogram sensor), or a blood pressure sensor.

  The environmental information detection means 4 is means for detecting information on environmental factors that affect an irritated state such as a traffic jam. For example, in the case of a traffic jam, a navigation device that acquires VICS information (such as traffic jam information), a vehicle speed sensor for detecting the vehicle speed and a processing unit that determines traffic jam based on the vehicle speed, or a combination thereof. In the case of an interrupt, a vehicle speed sensor for detecting the vehicle speed and a processing unit for determining the interrupt based on the vehicle speed, a radar sensor for detecting the inter-vehicle distance and a processing unit for determining the interrupt based on the inter-vehicle distance, and taking a front image A processing unit that determines an interruption based on the camera and image information, or a combination thereof. In the case of road conditions, a navigation device including a map database, a camera that captures a front image, a processing unit that determines road conditions based on image information, or a combination thereof. Each processing unit of the environment information detecting unit 4 may be configured in the ECU.

  The stimulus control means 5 is a means for determining a stimulus pattern for learning each cognitive reaction and controlling the stimulus presentation device 2 according to the stimulus pattern. The stimulus pattern for each stimulus (each stimulus presenting device 2) is set in advance. For each stimulus, a normal stimulus corresponding to the localization response learning, a different stimulus corresponding to the defense response learning, and a non-response learning There is no corresponding stimulus. For example, in the case of an alarm sound, as shown in FIG. 4, it is a stimulation pattern of a normal stimulus that generates an alarm sound of a normal volume for localization response learning, and a normal volume for a defense reaction learning. This is a stimulation pattern of an enhanced stimulus that generates an alarm sound with a large sound level, and a non-stimulus stimulation pattern that does not generate an alarm sound for unresponsive learning.

  Specifically, the stimulus control means 5 selects a stimulus to be applied according to the situation, and further determines a stimulus pattern of the selected stimulus. For the selection of the stimulus, for example, when the alarm sound is generated due to forgetting to fasten the seat belt, the alarm sound is selected, when the driver operates the horn, the horn is selected, and when the driver gets on, the automatic seat position is selected. When the route guidance output timing is selected, the route guidance screen or the route guidance voice is selected. For example, the stimulation pattern is determined so as to be alternately applied in the case of a normal stimulation and an enhanced stimulation. However, no stimulus for learning no response can be learned at any time when no stimulus is applied, so when no stimulus is applied, it is stimulated at regular intervals or at random times. Determine the pattern.

  Then, the stimulus control means 5 sets the intensity and timing of the stimulus in the application command according to the determined stimulus pattern, and outputs the application command to the selected stimulus presentation device 2. In addition, the stimulus control unit 5 includes time information for applying a stimulus to the cognitive reaction learning unit 6 (time information for performing no response when there is no stimulus for learning no response) and stimulus pattern information. Output. In addition, when giving normal stimulation, since normal stimulation is given by normal control with respect to the stimulus presentation apparatus 2, it is not necessary to output an application command to the stimulus presentation apparatus 2 in particular. In addition, when learning no response, no stimulus is given, so that no giving command is output to the stimulus presentation device 2.

  The cognitive reaction learning unit 6 is a unit that learns the judgment criterion of the cognitive reaction indicated in the stimulation pattern information based on the skin potential and the heart rate every time time information and stimulation pattern information are input from the stimulation control unit 5. is there. Here, every time learning is performed, the learning data stored in the learning data storage device 7 is updated. Learning data are set for stereotactic response, defense response, and no response, respectively. Analysis results of each feature amount and localization response obtained from skin potential and heart rate when stimulus is applied by each stimulus pattern or not. Includes criteria for defense response and no response.

  Specifically, in the cognitive reaction learning means 6, every time time information and stimulus pattern information are input from the stimulus control means 5, the cognitive reaction is started from the stimulus application time (or the start time of no response learning). A section t1 for cutting out biological information for learning is set. The interval t1 is an extremely short time range (second order) for determining the cognitive response, and is set in advance by an experiment or the like. The cognitive reaction learning means 6 cuts out the skin potential in the section t1 from the skin potential detected by the biological information detection means 3, and the heart rate in the section t1 from the heart rate detected by the biological information detection means 3. Cut out the number (see FIG. 5). FIG. 5 shows an example of changes in skin potential and heart rate when an alarm sound is used as a stimulus. A section t1 indicated by reference sign A is a change when no alarm sound is generated, a section t1 indicated by reference sign B is a change when an alarm sound having a normal volume is generated, and a section t1 indicated by reference sign C is normal. This is a change when a louder alarm sound is generated. FIGS. 6 and 7 show changes in skin potential and heart rate cut out from the section t1 when an alarm sound having a louder volume than usual is generated. When a large alarm sound is generated for the skin potential. At the same time, the skin potential rises momentarily, and the heart rate begins to rise rapidly when an alarm sound is generated.

  Since the skin potential represents the person's mental sweating, the value changes instantaneously only by changing the person's mental state (such as emotion). Therefore, the skin potential increases instantaneously only by the driver recognizing environmental factors and stimuli. On the other hand, the value of the heart rate changes instantaneously when an action such as a person moving the body or getting excited is taken. Therefore, when the driver recognizes environmental factors and stimuli and further takes some action, the heart rate increases instantaneously. Therefore, if only the skin potential increases instantaneously, it is presumed that the driver has only recognized, and this is regarded as a stereotaxic reaction. In addition, when the skin potential and the heart rate increase greatly instantaneously, the driver recognizes it, and further assumes that some action has been taken against environmental factors and stimuli, which is taken as a defense reaction.

  The cognitive reaction learning means 6 extracts feature amounts from the skin potential and heart rate in the cut section t1. Examples of the feature amount include increase / decrease (+, −, 0) in the section t1, maximum values, minimum values and average values in the section t1, and time change amounts in the section t1.

  In the cognitive reaction learning means 6, each time a feature amount is extracted, the feature amount for the cognitive reaction indicated in the stimulation pattern information in the learning data storage device 7 by the feature amount of the skin potential and the feature amount of the heart rate obtained in the current learning. Update the learning data. For example, when the feature amount is a time change amount, frequency distribution tables are prepared for the localization response, defense reaction, and no response for the time change amount of the skin potential and the time change amount of the heart rate as shown in FIG. Yes. This frequency distribution table divides the amount of time change by a fixed amount and indicates the frequency for each fixed amount. When the time variation of skin potential and heart rate is obtained, the frequency distribution table of skin potential and heart rate of the cognitive reaction to be learned this time is selected, and the frequency at this time variation is 1 in the frequency distribution table. The frequency distribution table for skin potential and heart rate is updated.

  The cognitive reaction learning means 6 each determines whether or not the update condition of the criterion is satisfied based on the updated learning data of each feature quantity of skin potential and heart rate each time the feature quantity learning data is updated. . The cognitive reaction learning means 6 updates the determination condition for the skin potential when the learning data of the skin potential feature amount satisfies the update condition, and updates the determination condition for the heart rate feature amount when the learning data of the heart rate feature amount satisfies the update condition. Updates the criteria for heart rate. For example, in the case of a frequency distribution table of time variation, a range from 5 percent tile to 95 percent tile (a range including 90% of all frequencies) in the frequency distribution table is determined, and an upper limit value and a lower limit in the section are determined. The value is used as a criterion. In the case of the example shown in FIG. 8, it is a frequency distribution table of changes in skin potential over time in defense response learning, and one frequency D is added to the time change in the range of α1 to α0 in this learning. The range from the 5 percent tile to the 95 percent tile is changed from the range of α0 to β to the range of α1 to β, and the criterion for the skin potential of the defense reaction is updated between α1 and β. Initial values are set in advance for each criterion of skin potential and heart rate for each cognitive reaction. Each initial value is set in advance by experiments with a large number of drivers.

  In addition to the frequency distribution table, an appropriate means may be used as means for analyzing each feature quantity in consideration of the type of feature quantity. As an update condition of the criterion for the learning data of each feature quantity, an appropriate update condition may be used in consideration of the learning data of the feature quantity. In addition, the determination criterion for no reaction may not be set because it may be determined that there is no response when both the localization reaction determination criterion and the defense reaction determination criterion are not satisfied.

  The cognitive reaction determination means 8 is a means for determining the driver's cognitive reaction to the environmental factor based on the skin potential and the heart rate when an environmental factor that affects the driver's frustration is detected.

  Specifically, the cognitive reaction determination unit 8 determines whether or not an environmental factor such as a traffic jam is detected by the environment information detection unit 4 at regular intervals. When an environmental factor is detected, the cognitive reaction determination means 8 sets a section t1 for cutting out biometric information for determining the cognitive reaction, starting from the time when the environmental factor is detected. The cognitive reaction determination means 8 cuts out the skin potential in the section t1 from the detection time from the skin potential detected by the biological information detection means 3, and also detects the detection time from the heart rate detected by the biological information detection means 3. From this, the heart rate in the section t1 is cut out (see FIGS. 10 and 11). Further, the cognitive reaction determination unit 8 extracts feature amounts from the skin potential and the heart rate of the cut section t1 as in the cognitive reaction learning unit 6.

  The cognitive reaction determination means 8 extracts the determination criteria of the latest localization response, defense response, non-response skin potential and heart rate from the learning data storage device 7. However, there may be no determination criteria for no reaction. FIG. 9 shows an example of a determination criterion when the feature amount is a time change amount. The determination criterion of the localization response is that the skin potential time change amount ΔGSR is in the range of 0.05 to 0.3 and the heart rate. Time variation ΔHR is in the range of −10 to 10, and the criteria for determining the defense response are skin potential time variation ΔGSR in the range of 0.05 to 0.5 and heart rate time variation ΔHR greater than 10. It is a range, and the criterion for no response is a range that does not satisfy both the criterion for stereotaxic response and the criterion for defense response.

  The cognitive response determination means 8 determines whether or not the skin potential feature quantity and the heart rate feature quantity satisfy the criteria for localization response, and the skin potential feature quantity and the heart rate feature quantity determine the defense response. Determine whether the criteria are met. When the feature value of skin potential and the feature value of heart rate satisfy the criteria for the localization response, the cognitive response determination means 8 sets the localization response (1) as the cognitive response Res. When the feature value of the skin potential and the feature value of the heart rate satisfy the determination criterion of the defense response, the cognitive response determination unit 8 sets the defense response (2) as the cognitive response Res. When the feature value of the skin potential and the feature value of the heart rate do not satisfy both the localization response determination criterion and the defense response determination criterion, the cognitive reaction determination means 8 sets no response (0) as the cognitive response Res.

  For example, when the determination criterion of FIG. 9 is applied, the time change amount ΔGSR of skin potential is 0.25 and the time change amount ΔHR of heart rate is 11 in the example shown in FIG. In the example shown in FIG. 11, the time variation ΔGSR of the skin potential is 0.2 and the time variation ΔHR of the heart rate is −6.

  As a method for determining the cognitive reaction, in addition to determining whether or not the feature quantity of the biological information falls within the determination reference range, the biological information is determined based on the magnitude of the threshold and determined based on the degree of correlation with the reference waveform of the biological information. Other determination methods may be used. For example, when the magnitude is determined by the threshold, the threshold is learned, and when the degree of correlation with the reference waveform is determined, the reference waveform is learned.

  The frustration influence degree calculation means 9 is a means for calculating the frustration influence degree from the environmental factor itself and the driver's cognitive reaction to the environmental factor when an environmental factor affecting the driver's frustration state is detected. Here, reference data stored in the reference data storage device 10 is used. The reference data is data in which the frustration influence degree (reference value) Ie with respect to the environmental factor itself is set, and is set in advance by experiments for a large number of drivers. For example, as reference data, the frustration influence degree Ie is 1 in the case of a short traffic jam, the frustration influence degree Ie is 1.5 in the case of a moderate traffic jam, and the frustration influence degree in the case of a long traffic jam. Ie is 2, and in the case of a sudden interruption (for example, an interruption to a vehicle with a short distance from the preceding vehicle), the frustration influence Ie is 2, and in the case of a normal interruption, the frustration influence Ie is In the case of a gradual interruption (for example, an interruption to a vehicle having a sufficiently long distance from the preceding vehicle), the frustration influence Ie is 0.5.

  Specifically, the frustration influence degree calculating means 9 refers to the reference data in the reference data storage device 10 and extracts the frustration influence degree (reference value) Ie for the detected environmental factor. Furthermore, the frustration influence calculation means 9 sets k0 as the cognitive response coefficient K when the cognitive response Res is 0 (no response), and the cognitive response coefficient K when the cognitive response Res is 1 (localization response). K1 is set, and when the cognitive response Res is 2 (defense response), k2 is set as the cognitive response coefficient K. k0, k1, and k2 are values from 0 to 1, and a larger value is set as the driver's cognitive response to environmental factors increases (k0 <k1 <k2). For example, k0 = 0, k1 = 0.3, and k2 = 1 are set. Then, the frustration influence degree calculation means 9 multiplies the frustration influence degree (reference value) Ie by the cognitive reaction coefficient K and sets the multiplied value as the final frustration influence degree (correction value) If. For example, when the frustration influence degree (reference value) is 1.5 and the cognitive response coefficient K is 0.3, the frustration influence degree (correction value) is 0.45.

  The frustration level determination means 11 is a means for determining the final frustration level by reflecting the frustration influence level on the frustration level (reference value) based on the heart rate and blood pressure.

  Specifically, the frustration level determination unit 11 cuts out the heart rate in the section t2 from the heart rate detected by the biological information detection unit 3. The section t2 is a time range (a time range longer than the section t1 and minute order) necessary for determining the frustrating level from the biological information, and is set in advance by an experiment or the like. Then, the frustrating level determination means 11 determines whether or not the heart rate in the section t2 is less than the threshold value a1 (range HR0), whether it is greater than or equal to the threshold value a1 and less than the threshold value a2 (range HR1), greater than or equal to the threshold value a2 and less than the threshold value a3 ( It is determined whether or not the range is HR2. The thresholds a1, a2, and a3 are values set in advance by experiments with a large number of subjects, and a1 <a2 <a3.

  Further, the frustrating level determination means 11 cuts out the blood pressure in the section t2 from the blood pressure detected by the biological information detection means 3. Then, the frustration level determination means 11 determines whether or not the blood pressure in the interval t2 is less than the threshold value b1 (range BP0), whether the blood pressure is greater than or equal to the threshold value b1 and less than the threshold value b2 (range BP1), greater than or equal to the threshold value b2 and less than the threshold value b3 (range). BP2) is determined. The threshold values b1, b2, and b3 are values set in advance by experiments with a large number of subjects, and b1 <b2 <b3.

  Then, the frustration level determination means 11 refers to the evaluation table and sets the frustration level (reference value) Ip corresponding to the determined heart rate range and the determined blood pressure range. The evaluation table is a table in which an irritated level (reference value) Ip is set for each combination of a heart rate range (HR0, HR1, HR3) and a blood pressure range (BP0, BP1, BP2). Is preset. As an evaluation table, for example, there are three stages of frustration levels of 0, 1, 2 integer values for each combination of heart rate range (HR0, HR1, HR3) and blood pressure range (BP0, BP1, BP2). A table to be set, a table in which five levels of frustration of 0, 0.5, 1, 1.5, and 2 are set, and the frustration level increases as the heart rate and blood pressure increase.

  In addition, about the determination method of the irritation level based on biometric information, it is not limited to said method, Various methods are applicable. For example, although the number of classifications of each piece of biometric information is set to three ranges, it may be two or four or more ranges, or a change amount in a section may be determined as a threshold value. Further, although the frustration level is set in the evaluation table, it may be determined using a predetermined determination formula.

  Then, the frustration level determination means 11 determines whether or not the frustration influence degree If is greater than zero. When the frustration influence level If is 0 (when the cognitive response is no response), there is no influence on the frustration state due to environmental factors, so the frustration level determination means 11 sets the frustration level (reference value) Ip to the final frustration level. Set to I as it is. On the other hand, when the irritability influence level If is greater than 0 (when the cognitive response is a localization response or a defense response), there is an influence on the irritability state due to environmental factors, so the irritability level determination means 11 uses the irritability level (reference value) Ip In order to reflect the frustration influence level If, the final frustration level I is calculated by the frustration judgment formula F (If, If). Various formulas can be applied to the frustration determination formula F (Ip, If). For example, F (Ip, If) = (a × Ip + b × If) / 2, and coefficients a and b are set in advance. Usually, it is 1.

  When the frustration judgment formula F (Ip, If) in the above example is used, the frustration level (reference value) Ip = 2, the frustration influence If = 0.45 (medium traffic congestion has been encountered, but the cognitive reaction is the stereotaxic reaction In the case of (1), the frustrating level I = (1 × 2 + 1 × 0.45) /2=1.225, which is the frustrating level 1. In this case, although a moderate traffic jam is encountered, it is determined that the frustrated state is light because it is a familiar traffic jam usually encountered. Incidentally, when the irritated level is determined only by the biometric information, the irritated level (reference value) Ip = 2 becomes the irritated level I as it is, and there is a risk of erroneous determination. In this case, although the frustrated state is light with respect to a moderate traffic jam, the heart rate and blood pressure increase due to conversation with a passenger or listening to hard music. Further, when the frustration level is determined by the biometric information and the environmental factor itself, the frustration influence degree (reference value) Ie = 1.5 is used instead of the frustration influence degree If = 0.45, and the frustration level I = (1 × 2 + 1 × 1.5) /2=1.75, which leads to an irritated level 2, which may reduce the determination accuracy. In this case, although the influence on the frustrated state is low because it is a familiar traffic jam, the influence on the frustrated state is set by the moderate traffic jam itself, so the frustrated influence level is higher than the actual.

  With reference to FIGS. 3-11, operation | movement of the irritation | stimulation estimation apparatus 1 is demonstrated. Here, a case where an alarm sound is used as a stimulus and a time change amount in the section t1 is used as a feature amount of biological information will be described. In particular, the processing of the stimulus control means 5 and the cognitive reaction learning means 6 will be described along the flowchart of FIG. 12, and the processing of the cognitive reaction determination means 8 will be described along the flowchart of FIG. FIG. 12 is a flowchart showing the flow of processing of the stimulus control means and the cognitive reaction learning means of FIG. FIG. 13 is a flowchart showing the flow of processing of the cognitive reaction determination means of FIG.

  The biological information detection means 3 always senses skin potential, heart rate, and blood pressure. In addition, the environmental information detection means 4 always senses environmental factors such as traffic jams and interruptions.

  The stimulus control means 5 selects a stimulus to be used for learning, and determines a stimulus pattern for the stimulus (S10). Here, since a warning sound is generated when the driver forgets the seat belt, the warning sound is selected as a stimulus, and a normal stimulus pattern and a stimulus pattern stronger than usual are determined by an appropriate method. Further, the stimulus control means 5 determines a stimulus pattern for learning no response at an appropriate time (S10).

  The stimulus control means 5 determines whether or not to give a stimulus based on the determined stimulus pattern (S11). When it is determined that no stimulus is applied in S11, the stimulus control unit 5 outputs learning start time information for learning no response and stimulus pattern information without applying stimulus to the cognitive reaction learning unit 6, and in S15 Transition to processing.

  On the other hand, when it determines with giving a stimulus in S11, the stimulus control means 5 determines whether a stimulus pattern is a normal pattern (S12). If the normal pattern is determined in S12, the stimulus presenting device 2 (alarm sound generating device) generates a normal stimulus (normal sound volume alarm sound) by the normal control of the alarm sound (S13). On the other hand, if it is determined in S12 that the pattern is not a normal pattern, the stimulus control means 5 sets an application command (applying command for generating an alarm sound with a louder volume than normal) according to the stimulus pattern and selects the selected stimulus presentation device 2 (alarm A grant command is output to the sound generator (S14). When this giving command is input, the stimulus presentation device 2 (warning sound generating device) generates a stronger stimulus (warning sound with a louder volume than usual) according to the giving command (S14). Then, the stimulus control means 5 outputs the stimulus occurrence time information and the stimulus pattern information for giving the stimulus to the cognitive reaction learning means 6.

  When the time information and the stimulation pattern information are input, the cognitive reaction learning means 6 determines the section t1 from the occurrence time (or the start time for learning no reaction) (S15), and the biometric information detection means 3 detects it. The skin potential and heart rate in the section t1 are cut out from the existing skin potential and heart rate (S16). And the cognitive reaction learning means 6 extracts the feature quantity from the skin potential in the section t1, and also extracts the feature quantity from the heart rate in the section t1 (S17).

  The cognitive reaction learning means 6 recognizes which reaction of the cognitive reactions is to be learned this time from the stimulus pattern information. Then, the cognitive reaction learning means 6 refers to the learning data in the learning data storage device 7 and extracts learning data of each feature quantity (time change amount) of skin potential and heart rate for the cognitive reaction to be learned this time (S18). ). Then, the cognitive reaction learning means 6 updates the learning data (feature distribution table of the time change amount) of the feature amount for the cognitive reaction to be learned this time, based on the feature amount of the skin potential and the feature amount of the heart rate extracted this time ( S19). Furthermore, the cognitive reaction learning means 6 determines whether or not the updated learning data of the feature value of the updated cognitive response this time satisfies the criteria update condition, and the learned data of the skin potential feature value satisfies the update condition. If the condition is satisfied, the skin potential determination criterion is updated. If the heart rate feature amount learning data satisfies the update condition, the heart rate determination criterion is updated (S19).

  The cognitive reaction determination means 8 determines whether or not an environmental factor such as a traffic jam has been detected by the environment information detection means 4 at regular intervals (S20). When it is determined that the environmental factor is detected in S20, the cognitive reaction determination unit 8 determines the section t1 from the detected time (S21), and from the skin potential and heart rate detected by the biological information detection unit 3 Cut out the skin potential and heart rate in the section t1 (S22). Further, the cognitive reaction determination means 8 extracts each feature amount (time change amount) from the skin potential and heart rate in the cut-out section t1 (S23). The cognitive reaction determination means 8 refers to the learning data in the learning data storage device 7 and extracts the latest determination criteria for the localization response and the cognitive reaction, respectively (S24). The latest criterion for no response is that it does not satisfy both the latest criterion for stereotactic response and the latest criterion for cognitive response.

  The cognitive reaction determination means 8 determines whether or not the temporal change amount of the skin potential satisfies the localization reaction determination criteria (S25). If it is determined in S25 that the temporal change amount of the skin potential satisfies the localization response determination criterion, the cognitive response determination means 8 determines whether or not the heart rate temporal change amount satisfies the localization response determination criterion ( S26). When it is determined in S26 that the temporal change amount of the heart rate satisfies the determination criterion for the localization response, the cognitive response determination unit 8 sets the localization response (1) as the cognitive response Res (S27).

  If it is determined in S25 that the temporal change amount of the skin potential does not satisfy the localization reaction criterion, the cognitive reaction determination means 8 determines whether or not the temporal change amount of the skin potential satisfies the defense reaction criterion. (S28). When it is determined in S28 that the time variation of the skin potential satisfies the criteria for determining the defense response, the cognitive response determination means 8 determines whether or not the time variation of the heart rate satisfies the criteria for determining the defense response ( S29). When it is determined in S29 that the amount of change in heart rate with time satisfies the criteria for determining the defense response, the cognitive response determination means 8 sets the defense response (2) as the cognitive response Res (S30).

  If it is determined in S26 that the temporal change in heart rate does not meet the criteria for localization response, or if it is determined in S28 that the temporal change in skin potential does not meet the criteria for defense response, or if the heart rate is determined in S29 When it is determined that the number of temporal change amounts does not satisfy the criteria for determining the defense response, the cognitive response determination unit 8 sets no response (0) as the cognitive response Res (S31).

  Subsequently, the frustration influence calculation means 9 refers to the reference data in the reference data storage device 10 using the detected environmental factor as a key, and acquires the frustration influence degree Ie for the detected environmental factor.

  Next, the frustration influence calculation means 9 determines whether or not the cognitive response Res = 0 (no response). When it is determined that the cognitive response Res = 0, the frustration influence calculation unit 9 sets k0 (= 0) as the cognitive response coefficient K. On the other hand, when it is determined that the cognitive response Res = 0 is not satisfied, the frustration influence calculation means 9 determines whether or not the cognitive response Res = 1 (localization response). When it is determined that the cognitive reaction Res = 1, the frustration influence calculation means 9 sets k1 (> k0) as the cognitive reaction coefficient K. On the other hand, when it is determined that the cognitive response Res = 1 is not satisfied, the frustration influence calculation means 9 sets k2 (> k1) as the cognitive response coefficient K. Then, the frustration influence calculation means 9 multiplies the frustration influence degree Ie by the cognitive reaction coefficient K and sets the multiplied value as the frustration influence degree If.

  The frustration level determination means 11 cuts out the heart rate in the section t2 from the heart rate detected by the biometric information detection means 3, and whether the heart rate in the section t2 is less than the threshold value a1 (range HR0) or more than the threshold value a1. In addition, it is determined whether or not it is less than the threshold value a2 (range HR1), and whether or not it is greater than or equal to the threshold value a2 and less than the threshold value a3 (range HR2). Further, the frustration level determination means 11 cuts out the blood pressure in the section t2 from the blood pressure detected by the biological information detection means 3, and determines whether or not the blood pressure in the section t2 is less than the threshold value b1 (range BP0). It is determined whether or not the threshold value is less than b2 (range BP1), and whether or not the threshold value b2 is greater than or equal to threshold value b3 (range BP2). Then, the frustrated level determining means 11 sets the frustrated level Ip corresponding to the determined heart rate range and blood pressure range.

  Next, the frustration level determination means 11 determines whether or not the frustration influence If is greater than zero. When the frustration influence level If is greater than 0, the frustration level determination means 11 calculates the final frustration level I by the frustration determination formula F (Ip, If) based on the frustration level Ip and the frustration influence level If. On the other hand, when the frustration influence level If is 0, the frustration level determination means 11 sets the frustration level Ip as the final frustration level I.

  Then, the frustration estimation device 1 outputs the estimated final frustration level I to various driver support devices, alarm devices, frustration mitigation devices, and the like.

  According to the frustration estimation device 1, by learning the judgment criteria of the cognitive reaction for estimating the frustration state of the driver for each driver, the estimation criteria for the frustration state can be set individually for each driver. Can be estimated with high accuracy. Furthermore, since this irritation estimation apparatus 1 determines a stimulation pattern for learning a cognitive reaction and learns by actively applying a stimulation, early and accurate learning can be performed.

  Furthermore, according to this irritation estimation device 1, by estimating the irritation state in consideration of the driver's individual cognitive response to the environmental factors that affect the irritation state of the driver, the influence of the environmental factors on the irritation state is estimated. Each driver can be accurately reflected, and the frustrated state of each driver can be estimated with higher accuracy. Moreover, in this irritation estimation apparatus 1, since the cognitive reaction appears as a change in biometric information instantly by using the cognitive reaction with respect to environmental factors, it is possible to cope with an instantaneous change in the irritation state. Therefore, this estimated frustration state can be used in order to prevent momentary recognition and misjudgment during driving (and in order to prevent collision).

  In this irritation estimation apparatus 1, the irritation level of each driver is estimated by estimating the irritation state of each driver in consideration of the driver's individual cognitive response to environmental factors at the irritation level based on the individual biometric information of the driver. It can be estimated with higher accuracy. Further, the frustration estimation apparatus 1 considers the influence of the environmental factors on the frustrated state in consideration of the influence based on the cognitive reaction to the individual environmental factors of the driver. Can be reflected more accurately, and the psychological state of each driver can be estimated with higher accuracy.

  As mentioned above, although embodiment which concerns on this invention was described, this invention is implemented in various forms, without being limited to the said embodiment.

  For example, in this embodiment, the present invention is applied to a frustration estimation device that is mounted on a vehicle and estimates a driver's frustration level, but can also be applied to a device that estimates other states such as impatience, boredom, awakening, and panic. It can also be applied to other objects such as drivers of other vehicles, various plant supervisors, and night workers.

  In this embodiment, a cognitive reaction is determined from changes in biological information according to environmental factors, and an irritating influence level (reference value) is set based on the environmental factors themselves, and the irritating influence level (reference value) is recognized. The final irritability level is calculated by taking the reaction into account, and the final irritability level is estimated by adding the irritability level to the irritability level estimated from the biological information. The frustration level may be directly estimated based on the cognitive reaction determined from the change in the biometric information according to the environmental factor without obtaining the frustrated level, or from the change in the biometric information according to the environmental factor The frustration level may be estimated based on the determined cognitive reaction and the environmental factor itself, or from the change of biological information according to the environmental factor It may be estimated final frustrating level based on frustrated levels estimated from boss was cognitive reaction and the biological information.

  Also, in this embodiment, the three-stage reaction of the stereotactic reaction, the defense reaction, and the no-response is used as the cognitive reaction, but only two of these reactions may be used, or other reactions may be added. There may be four or more stages.

  In this embodiment, skin potential and heart rate are used as biometric information to determine cognitive reaction, and heart rate and blood pressure are used as biometric information to determine annoyance level. One or three or more may be used, other biological information may be used, or other information such as a driver's behavior may be used.

  Further, in the present embodiment, the present invention is applied to the case where the judgment criterion of the cognitive reaction is learned as the psychological state estimation parameter, but other cases such as the case where the judgment criterion is learned for determining the frustration level based on the biological information. The present invention can also be applied when learning a psychological state estimation parameter.

It is a conceptual diagram of this invention. It is a time change of biological information with respect to each stimulation pattern. It is a block diagram of the frustration estimation apparatus which concerns on this Embodiment. It is an example of each stimulation pattern. It is an example of the time change of the biometric information (skin electricity, heart rate) with respect to each irritation | stimulation pattern of a driver | operator. It is an example of the time change of the determination area of the skin potential with respect to a driver | operator's strong irritation | stimulation. It is an example of the time change of the determination interval of the heart rate with respect to a driver | operator's strong irritation | stimulation. It is an example of the frequency distribution table of the amount of time change of living body information to a stimulus pattern. It is an example of the learned criteria (reference range of biometric information (skin potential, heart rate) for each cognitive reaction). It is an example of the time change of the biometric information at the time of showing a defense reaction with respect to a driver | operator's environmental factor, (a) is a time change of skin potential, (b) is a time change of heart rate. It is an example of the time change of the biometric information at the time of showing a stereotaxic response with respect to a driver | operator's environmental factor, (a) is a time change of skin potential, (b) is a time change of heart rate. It is a flowchart which shows the flow of a process of the stimulus control means and cognitive reaction learning means of FIG. It is a flowchart which shows the flow of a process of the cognitive reaction determination means of FIG.

Explanation of symbols

  DESCRIPTION OF SYMBOLS 1 ... Irritating estimation apparatus, 2 ... Stimulus presentation apparatus, 3 ... Biological information detection means, 4 ... Environmental information detection means, 5 ... Stimulation control means, 6 ... Cognitive reaction learning means, 7 ... Learning data storage apparatus, 8 ... Cognitive reaction Determining means, 9 ... Irritated influence degree calculating means, 10 ... Reference data storage device, 11 ... Irritated level determining means

Claims (3)

A psychological state estimation device that estimates a subject's psychological state,
A stimulus setting means for setting a stimulus applying pattern to be given to the subject;
A stimulus applying means for applying a stimulus to the subject based on the stimulus applying pattern set by the stimulus setting means;
Biometric information acquisition means for acquiring biometric information of the subject;
Biometric information change acquisition means for acquiring a change in biometric information with respect to the stimulus applied by the stimulus applying means based on the biometric information acquired by the biometric information acquisition means;
Learning means for learning a parameter for estimating a psychological state of a subject by associating a stimulus application pattern set by the stimulus setting means with a change in biological information with respect to a stimulus acquired by the biological information change acquisition means;
A psychological state estimation device comprising: estimation means for estimating the psychological state of the subject based on the parameter for estimating the psychological state of the subject learned by the learning means. Cognitive response determination means for determining a cognitive response to an external event for the subject,
The stimulus setting means sets a stimulus application pattern such that a subject shows a cognitive reaction,
The learning means learns a criterion for a cognitive response to an external event,
The cognitive reaction determination means determines a cognitive response to an external event based on the determination criterion learned by the learning means,
The psychological state estimation apparatus according to claim 1, wherein the estimation unit estimates the psychological state of the subject based on the cognitive reaction determined by the cognitive reaction determination unit. An influence degree acquisition means for acquiring an influence degree on a change in the psychological state of the subject based on the cognitive reaction determined by the cognitive reaction determination means;
The psychological state estimation according to claim 2, wherein the estimating unit estimates the psychological state of the subject based on the biological information acquired by the biological information acquiring unit and the influence level acquired by the influence level acquiring unit. apparatus.

Images