Determination device, selection device, determination system, determination method, selection method, and program

Abstract

To determine a bond and teamwork in an activity in a cooperative activity by multiple persons.SOLUTION: A determination device includes a reception part and a determination part. The reception part receives time-series measurement values of physiological signals or non-verbal signals acquired by measuring a plurality of persons to be measured belonging to a group performing a cooperative activity. The determination part calculates a synchronous level of a change in the measurement value in the group using the time-series measurement values of the plurality of persons to be measured, and determines the strength of the teamwork or the psychological bond in the cooperative activity on the basis of the calculated synchronous level.SELECTED DRAWING: Figure 19

Description

[Technical field]

[0001] The present invention relates to a determination device, a selection device, a determination system, a determination method, a selection method, and a program. [Background technology]

[0002] Humans are social animals that always cooperate physically and intellectually with others to achieve common dreams and goals, thereby preserving the individual and the species, and embodying activities unique to humans, including economics, science, agriculture, education, and sports. Conventionally, there is a technology that detects the synchrony of a conversation by synchronizing the amount and timing of speech between two people (see, for example, Patent Document 1). On the other hand, there is a technology that detects stress from the synchronization between physiological indicators such as breathing, blood pressure, and heart rate variability within an individual, and promotes feedback and stress relief by sounds of specific judgment features (see, for example, Patent Document 2). There is also a technology that improves an individual's mood by stimulating them with music, etc., based on the individual's physiological information (see, for example, Patent Document 3). [Prior art documents] [Patent documents]

[0003] [Patent Document 1] JP 2013-52049 A [Patent Document 2] Patent No. 4410234 [Patent Document 3] Patent No. 4378957 Summary of the Invention [Problem to be solved by the invention]

[0004] Until now, teamwork and psychological bonds (hereafter referred to as bonds) during human self-other collaborative activities could only be evaluated ex post using questionnaires or team performance. Therefore, it was difficult to obtain the quality of teamwork or the strength of bonds instantly during collaborative activities involving many people. Therefore, it was not possible to visualize and provide feedback on the levels of teamwork and bonds in real time, or to adjust or strengthen team performance.

[0005] The present invention has been made in consideration of the above-mentioned circumstances, and provides a technique that makes it possible to assess teamwork and psychological bonds in collaborative activities involving a large number of people. [Means for solving the problem]

[0006] One aspect of the present invention is a judgment device that includes a receiving unit that receives time series measurement values ​​of physiological signals or non-verbal signals obtained by measuring each of a number of subjects belonging to a group engaging in a collaborative activity, and a judgment unit that calculates a synchronization level of changes in the measurement values ​​in the group using the time series measurement values ​​of the multiple subjects, and judges the strength of teamwork or psychological bonds in the collaborative activity based on the calculated synchronization level.

[0007] One aspect of the present invention is the above-mentioned judgment device, further comprising a feedback unit that outputs the strength of teamwork or psychological bonds judged by the judgment unit so that it can be perceived using one or more of vision, hearing, or touch.

[0008] One aspect of the present invention is the above-mentioned judgment device, which further includes a feedback unit that controls the measurement subject to receive visual, auditory, tactile, olfactory or gustatory stimuli to adjust the teamwork or psychological bond in the collaborative activity, depending on the strength of the teamwork or psychological bond judged by the judgment unit.

[0009] One aspect of the present invention is the above-mentioned determination device, wherein the physiological signal is heart rate, mental sweating, surface body temperature, pupil diameter, blood components or saliva components, and the non-verbal signal is facial expression, posture, gestural movement, breathing or vocalization volume.

[0010] One aspect of the present invention is a selection device that includes a memory unit that stores time series measurement values ​​of multiple types of physiological signals or non-verbal signals obtained from each of multiple measurement subjects belonging to a group engaging in collaborative activities, and an evaluation value that quantitatively represents teamwork or psychological bonds in the collaborative activity, and a selection unit that calculates a synchronization level of changes in the measurement values ​​in the group using the time series measurement values ​​of the multiple measurement subjects for each of the multiple types of physiological signals or non-verbal signals, and selects the type of physiological signal or non-verbal signal to use in determining the strength of teamwork or psychological bonds in the collaborative activity based on the strength of correlation between the calculated synchronization level and the evaluation value.

[0011] One aspect of the present invention is a judgment system comprising: a memory unit that stores time series measurement values ​​of multiple types of physiological signals or non-verbal signals obtained from each of multiple measurement subjects belonging to a group performing a collaborative activity, and an evaluation value that quantitatively represents teamwork or psychological bonds in the collaborative activity; a selection unit that calculates a synchronization level of changes in the measurement values ​​in the group using the time series measurement values ​​of the multiple measurement subjects for each of the multiple types of physiological signals or non-verbal signals, and selects a type of physiological signal or non-verbal signal to be used in judging the strength of the teamwork or psychological bonds in the collaborative activity based on the strength of correlation between the calculated synchronization level and the evaluation value; a receiving unit that receives time series measurement values ​​of the selected type of physiological signal or non-verbal signal obtained by measuring each of multiple measurement subjects belonging to a judgment group performing a collaborative activity; and a judgment unit that calculates a synchronization level of changes in the measurement values ​​in the judgment group using the time series measurement values ​​of the multiple measurement subjects received by the receiving unit, and judges the strength of the teamwork or psychological bonds in the collaborative activity of the judgment group based on the calculated synchronization level.

[0012] One aspect of the present invention is a method for determining the strength of teamwork or psychological bonds in the collaborative activity, comprising: a receiving step of receiving a time series of measurement values ​​of physiological signals or non-verbal signals obtained by measuring each of a number of subjects belonging to a group engaged in a collaborative activity; and a determining step of calculating a synchronization level of changes in the measurement values ​​in the group using the time series measurement values ​​of the multiple subjects, and determining the strength of teamwork or psychological bonds in the collaborative activity based on the calculated synchronization level.

[0013] One aspect of the present invention is a selection method having an acquisition step of acquiring time series measurement values ​​of multiple types of physiological signals or non-verbal signals obtained from each of multiple measurement subjects belonging to a group engaged in collaborative activity, and an evaluation value quantitatively representing teamwork or psychological bonds in the collaborative activity, and a selection step of calculating, for each of the multiple types of physiological signals or non-verbal signals, a synchronization level of changes in the measurement values ​​in the group using the time series measurement values ​​of the multiple measurement subjects, and selecting the type of physiological signal or non-verbal signal to be used for determining the strength of teamwork or psychological bonds in the collaborative activity based on the strength of correlation between the calculated synchronization level and the evaluation value.

[0014] One aspect of the present invention is a method for determining whether a group is performing a collaborative activity, the method comprising: a receiving step of receiving time series measurement values ​​of multiple types of physiological signals or non-verbal signals obtained from each of multiple subjects belonging to a group performing a collaborative activity, and an evaluation value that quantitatively represents teamwork or psychological bonds in the collaborative activity; a selection step of calculating, for each of multiple types of physiological signals or non-verbal signals, a synchronization level of changes in the measurement values ​​in the group using the time series measurement values ​​of the multiple subjects, and selecting a type of physiological signal or non-verbal signal to be used in determining the strength of the teamwork or psychological bonds in the collaborative activity based on the strength of correlation between the calculated synchronization level and the evaluation value; a receiving step of receiving time series measurement values ​​of the selected type of physiological signal or non-verbal signal obtained by measuring each of multiple subjects belonging to a group performing a collaborative activity; and a determination step of calculating a synchronization level of changes in the measurement values ​​in the group to be determined, using the time series measurement values ​​of the multiple subjects received in the receiving step, and determining the strength of the teamwork or psychological bonds in the collaborative activity of the group to be determined based on the calculated synchronization level.

[0015] One aspect of the present invention is a program for causing a computer to function as the above-described determination device.

[0016] One aspect of the present invention is a program for causing a computer to function as the above-described selection device. Effect of the Invention

[0017] According to the present invention, it is possible to assess teamwork and psychological bonds in collaborative activities involving a large number of people. [Brief description of the drawings]

[0018] [Figure 1] FIG. 1 shows the measurement timing of each measurement item in experiment A. [Diagram 2] FIG. 13 is a diagram showing the experimental results of Experiment A. [Diagram 3] FIG. 13 is a diagram showing the experimental results of Experiment A. [Figure 4] FIG. 13 is a diagram showing the experimental results of Experiment A. [Diagram 5] FIG. 13 is a diagram showing the experimental results of Experiment A. [Figure 6] FIG. 13 is a diagram showing the experimental results of Experiment A. [Figure 7] FIG. 13 is a diagram showing the experimental results of Experiment A. [Figure 8] FIG. 13 is a diagram showing the experimental results of Experiment A. [Figure 9] FIG. 13 is a diagram showing the experimental results of Experiment A. [Figure 10] FIG. 13 is a diagram showing the experimental results of Experiment A. [Figure 11] FIG. 13 is a diagram showing the measurement timing of each measurement item in Experiment B. [Figure 12] FIG. 13 is a diagram showing the experimental results of Experiment B. [Figure 13] FIG. 13 is a diagram showing the experimental results of Experiment B. [Figure 14] FIG. 13 is a diagram showing the experimental results of Experiment B. [Figure 15] FIG. 13 is a diagram showing the experimental results of Experiment B. [Figure 16] FIG. 13 is a diagram showing the experimental results of Experiment B. [Figure 17] FIG. 13 is a diagram showing the experimental results of Experiment B. [Figure 18] FIG. 13 is a diagram showing the experimental results of Experiment B. [Figure 19] FIG. 1 is a diagram illustrating an example of the configuration of a determination system according to an embodiment. [Figure 20] FIG. 2 is a flowchart showing a process of the decision model generating device according to the embodiment. [Figure 21] FIG. 4 is a flowchart showing a process of the determination device according to the embodiment. [Figure 22] FIG. 2 is a diagram illustrating a hardware configuration of a determination model generating device and a determination device according to an embodiment. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0019] Hereinafter, an embodiment of the present invention will be described in detail with reference to the drawings. The embodiment relates to a technique for detecting teamwork and psychological ties during collaboration among many people, a technique for providing feedback on teamwork and psychological ties during collaboration among many people, and a technique for adjusting teamwork and psychological ties during collaboration among many people.

[0020] Although collaboration with others is a fundamental part of human activity, traditionally, its quality and quantity could only be evaluated ex post. For example, the technology in Patent Document 1 can evaluate the amount and timing of speech between two people in real time, but cannot be applied to a large number of people. In addition, it does not take into account non-verbal signals such as physiological and social signals that affect teamwork evaluation and psychological bonds more than linguistic information.

[0021] In addition, in order to detect the level of bonds and teamwork among many people and utilize it to strengthen them, it is important to visualize the current state of teamwork and psychological bonds detected for individuals or teams in real time. The technology in Patent Document 2 detects stress from the synchronization between physiological indices such as respiration, blood pressure, and heart rate variability within an individual, and encourages feedback and stress relief with a sound of a specific judgment feature. However, it cannot be applied to evaluation and feedback of teamwork between individuals.

[0022] In addition, in order to strengthen teamwork and psychological bonds among many people, it is the only way to create an environment that stimulates the five senses of individuals, both internal and external, and within and outside the team, without impeding work. The technology in Patent Document 3 improves the mood of individuals by providing stimuli such as music based on the physiological information of the individuals, but it cannot be applied to moods or teamwork performance when many people work together.

[0023] In this embodiment, the level of good teamwork or the strength of psychological ties when multiple participants working together as a team are detected instantly based on the synchronization level of features obtained from each participant. Physiological signals and non-verbal social signals are used as the features. Physiological signals include, for example, time-series heart rate, mental sweating, surface body temperature, pupil diameter, blood components, saliva components, etc. Non-verbal social signals include facial expressions, posture, gestures, breathing, vocalization volume, etc., and voluntary signals can also be used.

[0024] Furthermore, in this embodiment, the results of detection of the level of teamwork or psychological bonding detected in real time from physiological signals and social signals among multiple people are visualized and fed back using visual, auditory, tactile, etc. The feedback method may be not only visual expression, but also auditory or tactile feedback depending on the time, place, and occasion, or may be a combination of these.

[0025] In addition, in this embodiment, based on the levels of teamwork and psychological bonds detected in real time from physiological signals and social signals between multiple people, the system modifies and strengthens them by influencing the internal and external environments of individuals and teams through the five senses (sight, hearing, touch, smell, and taste). For example, the system automatically controls air conditioning, lighting, music, etc. during collaboration, and suggests nutritional intake, exercise, and rest strategies. Alternatively, the system modifies the synchronization level between participants to adjust and strengthen teamwork and psychological bonds.

[0026] According to this embodiment, it is possible to visualize and strengthen teamwork in various sports teams, including motorsports and e-sports. In addition, according to this embodiment, it is possible to improve the sense of unity in live events such as sports spectators, music and theater. Furthermore, it is possible to detect, feedback and adjust teamwork and psychological bonds that eliminate loneliness in all human behavioral situations, such as in the office, in a classroom, at a senior citizen's facility, in a hospital, in an online conference, or in the metaverse.

[0027] First, we explain two experiments, Experiment A and Experiment B, regarding the relationship between teamwork and ties and features.

[0028] First, we will explain Experiment A. The participants in the experiment were 22 people in total, 4 teams consisting of 5-6 male university students each. These 4 teams played a round-robin tournament of a first-person shooter game, a type of e-sports, offline. The items measured were behavior (game stats), physiological indicators (saliva and heart rate), and psychology (team flow and intimacy).

[0029] FIG. 1 shows the timing of each measurement item in Experiment A. As shown in FIG. 1, heart rate measurement was started at 9:00 a.m., saliva measurement was also conducted, and the teamwork level was measured by having participants fill out a questionnaire. Then, just before the match, the teamwork level was measured using the questionnaire. After that, a shooting game was played for 60-120 minutes, and game stats representing the results of individual performance were measured. The game consisted of two sets, Set 1 and Set 2, with a break (halftime: HT) between them. After the match, saliva measurement and teamwork level measurement using the questionnaire were conducted. In addition, 30 minutes after the end of the match, saliva measurement and teamwork level measurement using the questionnaire were also conducted, and then the heart rate measurement was completed.

[0030] Representative indicators of game stats include Combat Score and KDA. Combat Score is a score for each match that is obtained according to the performance in the match. KDA is calculated by dividing the total of kills and assists by deaths. Kills are the number of times an opponent was killed, deaths are the number of times an opponent killed you, and assists are the number of times you helped a teammate to kill an opponent. Oxytocin concentration was quantified from saliva using the ELISA method. Oxytocin is a bonding hormone, and high levels of oxytocin tend to create teamwork, and it is secreted when playing sports that require teamwork. For heart rates, the synchronization level between participants was quantified using a cross correlation function. To quantify the synchronization level of the waveforms, we used the CCF (Cross Correlation Function), which represents similarity as a number between -1.0 and 1.0, and Lag, which indicates the time lag at which the CCF peaks. The window size of Lag was set to ±30 seconds.

[0031] The questionnaire elicited responses on a team scale that quantitatively represents the level of team flow. Team flow is a flow that occurs when a team works on advanced tasks toward a goal. This flow is a state in which a person is deeply immersed in one activity and nothing else matters. The questionnaire also elicited responses on intimacy (IOS) (Inclusion of other in the self). This is a measure of the psychological distance and intimacy between the respondent and the target, which is presented with a choice of illustrations in which a circle representing the self and a circle representing an target overlap to different degrees, and the degree of overlap of the two circles in the illustration selected by the respondent is used to measure the degree of intimacy.

[0032] 2 to 12 are diagrams showing the results of Experiment A. FIG. 2(a) shows the Combat Scores of the winning team and the losing team, and FIG. 2(b) shows the KDA of the winning team and the losing team. These show that the winning team had higher performance.

[0033] Figure 3(a) shows the percentage of hits by headshots for the winning and losing teams, Figure 3(b) shows the percentage of hits by bodyshots for the winning and losing teams, and Figure 3(c) shows the percentage of hits by legshots for the winning and losing teams. A headshot causes large damage to the enemy by shooting the enemy's head. A bodyshot causes medium damage to the enemy by shooting the enemy's torso. A legshot causes small damage to the enemy by shooting the enemy's legs. Figure 3 shows that there is no difference in shot accuracy between the winning and losing teams (ns), and that shot accuracy, which is an individual skill, does not determine victory or defeat.

[0034] Figure 4(a) shows the assist time of the winning and losing teams, and Figure 4(b) shows the average number of assists per round for the winning and losing teams. Figure 4 shows that teamwork is more closely related to winning and losing than individual skill. Figure 5 shows the correlation between assists and KDA. As Figure 5 shows, we can see that increased teamwork leads to higher performance.

[0035] Figure 6(a) shows the CCF of heart rate for the winning and losing teams before the match, Set 1, HT, Set 2, and 30 minutes after the match, respectively, and Figure 6(b) shows the Lag for the winning and losing teams before the match, Set 1, HT, Set 2, and 30 minutes after the match, respectively. It can be seen that the winning team has higher heart rate synchronization than the losing team.

[0036] Figure 7 shows the flow ratings of winning and losing teams, which are quantitative values ​​out of 7 for each flow question. The questions were: (1) did I have the skills necessary to play well in this match (skill-demand balance 1); (2) would I have enjoyed the match more if my opponent had been stronger (weaker)? (skill-demand balance 2); (3) did I feel in control while playing this match (feeling in control); (4) did I make accurate movements unconsciously without thinking (automaticity); (5) did I like the feeling of the match and would like to play again (enjoyment); (6) felt that time passed quickly (time perception); (7) was conscious of my teammates' movements (awareness of partner); (8) felt that I was playing as a team with my teammates (team work); (9) my movements were coordinated with my teammates' movements (coordination); (10) felt like I was competing with my teammates (competition); and (11) was distracted by my teammates' movements (distraction). Items (1) to (6) are related to individual flow, and items (7) to (11) are related to team flow. According to Figure 7, a deeper flow state as a team is believed to lead to victory.

[0037] Figure 8 shows the change in IOS of teammates before, after, and 30 minutes after the match for the winning and losing teams. Figure 8 shows that playing as a team increases intimacy with teammates.

[0038] Figure 9(a) shows the relationship between the average oxytocin level before the match (OT Pre average) and the teamwork evaluation score of item (8) of team flow, Figure 9(b) shows the relationship between the average oxytocin level before the match (OT Pre average) and IOS after the match, and Figure 9(c) shows the relationship between the average oxytocin level before the match (OT Pre average) and assists. According to Figure 9, oxytocin may be involved in the formation of team flow.

[0039] Fig. 10(a) shows the relationship between the average heart rate synchronization (CNN) during the game and the evaluation value of team flow, Fig. 10(b) shows the relationship between the average heart rate synchronization during the game and IOS, and Fig. 10(c) shows the relationship between the average heart rate synchronization during the game and assists. According to Fig. 10, heart rate synchronization may have a direct effect on team flow, intimacy, and teamwork.

[0040] The results of Experiment A confirmed that teamwork during e-sports play may be generated through oxytocin and heart rate synchronization.

[0041] Next, we will explain Experiment B. Experiment A was an experiment on participants in e-sports, while Experiment B was an experiment on participants in an event to watch sports. The participants in Experiment B were 10 members of the general public, 10 students other than those majoring in physical education, 10 students majoring in physical education, and 10 students affiliated with the basketball club. The participants watched a basketball game. The items measured included physiological indicators (saliva and heart rate) and psychology (level of support, IOS, excitement, loyalty), etc.

[0042] Figure 11 is a diagram showing the timing of measurement of each measurement item in Experiment B. As shown in Figure 11, participants wore a heart rate monitor 60 minutes before watching the game, and responded to psychological questions by filling out questionnaires and collected saliva samples 30 minutes before watching the game (Pre), after the end of the second quarter (Q2), after the end of the fourth quarter (Q4), and 15 minutes after the end of watching the game (Post).

[0043] 12 to 18 are diagrams showing the results of Experiment B. FIG. 12(a) and FIG. 12(b) are diagrams showing changes in CCF of heart rates and distribution of CCF among general citizens, students other than physical education majors, students majoring in physical education, and students related to the basketball club, respectively, 30 minutes before watching the game (Pre), during watching the game (Watching play), and 15 minutes after watching the game (Post). FIG. 13(a) and FIG. 13(b) are diagrams showing changes in Lag of heart rate synchronization and distribution of Lag among general citizens, students other than physical education majors, students majoring in physical education, and students related to the basketball club, respectively, 30 minutes before watching the game (Pre), during watching the game (Watching play), and 15 minutes after watching the game (Post). According to these, the waveforms of heart rates of students related to the basketball club are synchronized and the Lag is small while watching the game. On the other hand, for the general public, there was no change in the synchronization of heart rates while watching the game, and there was also no change in the time difference (Tag).For students other than those majoring in physical education and students majoring in physical education, it can be seen that the heart rate waveforms were synchronized 15 minutes after the end of the game (Post).

[0044] Figure 14 is a plot of the relationship between the average heart rate synchronization level during a game between students related to the basketball club (CCF_AVE_in_play) and the evaluation value of unity (IOS_TsukubaPlayer_Post), and Figure 15 is a plot of the relationship between the maximum heart rate synchronization level at halftime (CCF_MAX_HT) and the evaluation value of enjoyment after the game ended (Enjoyment_Post). In addition, Figure 16 shows the relationship between the average heart rate synchronization level at halftime (CCF_AVE_HT) and the recommendation intention (Loyalty_Recommend_Post) 15 minutes after the end of the game viewing, Figure 17 shows the relationship between the average heart rate synchronization level at halftime (CCF_AVE_HT) and the evaluation value of the intention to revisit the game (Loyalty_Revisit_Post) 15 minutes after the end of the game viewing, and Figure 18 shows the relationship between the average heart rate synchronization level at halftime (CCF_AVE_HT) and the friendliness (POMS2_Friendliness_Post) 15 minutes after the end of the game viewing.

[0045] The following can be derived from Figures 12 to 18. That is, watching a live game increases a sense of unity and loyalty with the players and the university, regardless of the spectator's sports experience. On the other hand, the heart rate synchronization level, which indicates a sense of unity and enthusiasm in a group, increases in the group of spectators who specialize in sports while watching the game, while the heart rate synchronization level of the group of spectators who do not specialize in sports increases during cheering performances at halftime and after the game and when reflecting on the game. However, heart rate synchronization outside of the game (at halftime and after the game) is related to the enjoyment, sense of unity with the players and the university, and increased loyalty, regardless of the spectator's sports expertise. In addition, the heart rate synchronization level of home game spectators is likely to be the main physiological mechanism that reduces the psychological distance between the players, the team, and the university.

[0046] From the above two experimental results, it can be seen that teamwork and psychological ties during collaboration among participants in a multi-person event, such as participating in an event as a sports player or participating in an event to watch a sport, are related to heart rate synchronization. In addition, it is considered that the synchronization level of other physiological signals and non-verbal social signals, in addition to heart rate, can be used to detect the level of teamwork and psychological ties. Therefore, the teamwork or psychological bond detection system of this embodiment specifies the type of feature amount, such as physiological signals and non-verbal social signals, that can efficiently detect the level of teamwork and psychological ties, and detects the level of teamwork and psychological ties using the specified feature amount. The teamwork or psychological bond detection system provides feedback on the detection results and works to adjust and strengthen the teamwork and psychological ties. In addition, since the situation in which the synchronization level is high may differ depending on the attributes of the participants, in this embodiment, when specifying the type of feature amount related to the level of teamwork or psychological ties, or when detecting the level of teamwork or psychological ties using the specified feature amount, the synchronization level of the feature amount obtained for each group of participants with the same attribute can also be used.

[0047] 19 is a diagram showing the configuration of a teamwork or psychological bond judgment system 1 (hereinafter referred to as the judgment system 1) according to one embodiment of the present invention. The judgment system 1 includes a judgment model generation device 2, a teamwork or psychological bond judgment device 3 (hereinafter referred to as the judgment device 3), and a measurement device 4.

[0048] The determination model generating device 2 includes a storage unit 21, a learning data acquiring unit 22, a selection unit 23, a model generating unit 24, and an output unit 25. The determination model generating device 2 is realized by a computer device such as a server computer or a personal computer.

[0049] The storage unit 21 stores various data including learning data and a judgment model. The learning data is data used to learn a judgment model used to judge teamwork or psychological bonds. The learning data indicates an activity type, a participant attribute, and participant data of each participant.

[0050] The activity type indicates the type of activity (event) in which many people work together. The activity type is, for example, but not limited to, participation in physical sports, participation in e-sports, participation in motorsports, watching sports, office work, driving, etc. It may be more detailed, for example, watching a sports game of the home team. In the following, a group of participants who work together to participate in an activity is referred to as a "team."

[0051] The participant attribute indicates the attributes of all participants belonging to a team. For example, if the activity is watching a sport, the participant attribute may be whether or not the participant has experience of that sport, or whether or not the participant is a spectator of the home team. Note that the participant attribute may be omitted.

[0052] The participant data indicates data measured when a participant having participant attributes participated in an activity indicated by an activity type. The participant data includes data on a measurement feature type, a measurement feature, a feature measurement time, an evaluation value of teamwork or psychological bonding, and a time when the evaluation value was measured. One piece of participant data may include measurement features and feature measurement times for multiple types of measurement feature types.

[0053] The measurement feature type indicates the type of feature measured for the participant, and the measurement feature is a measurement value obtained by measuring the feature of the type indicated by the measurement feature type. The measurement feature types are heart rate, mental sweating, surface body temperature, pupil diameter, blood components, saliva components (e.g., oxytocin), facial expression, joints, pressure, breathing, and vocalization volume. The heart rate, mental sweating, surface body temperature, pupil diameter, blood components, and saliva components are physiological signals, and the facial expression, joints, pressure, breathing, and vocalization volume are non-verbal social signals (voluntary signals). When the measurement feature type is a joint, the measurement feature indicates the position of the joint. For example, the joint is the neck, the waist, etc., and the measurement feature indicates the position of the joint. The position of the joint indicates the gesture movement or posture. For example, the pressure is measured by one or more pressure gauges arranged on the back of a chair on which the participant sits. By using measurements from one or more pressure gauges placed on the back of a chair, it is possible to measure postures such as when a participant leans forward while watching a sporting event and then leans back to catch their breath.

[0054] The evaluation value of teamwork or psychological bond quantitatively represents the teamwork or psychological bond of a team. Hereinafter, teamwork or psychological bond may be simply referred to as teamwork. The evaluation value of teamwork or psychological bond is obtained, for example, by participants filling out a questionnaire. Alternatively, if an activity is performed in which the better the teamwork or the stronger the psychological bond, a higher score can be obtained, the score can be used as the evaluation value of teamwork or psychological bond. The evaluation value measurement time is the time when the evaluation value of teamwork or psychological bond is measured.

[0055] The judgment model includes judgment feature types and level classification conditions. The judgment feature types are types of features used to judge the quality of teamwork or the strength of psychological bonds. The judgment feature types are selected from among the measurement feature types. A feature of a judgment feature type is described as a judgment feature. The level classification conditions indicate the correspondence between a synchronization level and a judgment level. The synchronization level indicates the level at which the judgment features between teams are synchronized. The judgment levels indicate the quality of teamwork or the strength of bonds in multiple stages.

[0056] The synchronization level of the feature in a team is calculated as follows. First, all combinations of two participants in the team are generated. For example, if a team consists of three people, participant A, participant B, and participant C, three pairs of participants A and B, participants B and C, and participants A and C are generated. Next, the synchronization level of the feature is calculated for each generated combination, and the average or median of the synchronization levels of all the combinations is set as the synchronization level of the team. The synchronization level between two participants is obtained by the cross-correlation coefficient of the feature in time series or wavelet coherence analysis. When these are used, the synchronization level is expressed in the range of -1.0 to 1.0. In this case, in the level classification conditions, for example, a team synchronization level of less than 0.2 corresponds to a teamwork judgment level of "low", a team synchronization level of 0.3 to 0.4 corresponds to a judgment level of "medium", and a team synchronization level of 0.4 or more corresponds to a judgment level of "high".

[0057] The learning data acquisition unit 22 acquires a plurality of pieces of learning data. For example, the learning data acquisition unit 22 may receive the learning data from another computer device connected via a network, or may read the learning data from a computer-readable recording medium. The learning data acquisition unit 22 stores the acquired plurality of pieces of learning data in the storage unit 21.

[0058] The selection unit 23 selects, for each activity type or for each combination of activity type and participant attribute, a measurement feature type that is highly correlated with the evaluation value of teamwork or psychological bond from among the measurement feature types included in the learning data as a feature type for determination. That is, the selection unit 23 calculates a team synchronization level for each piece of learning data using the measurement feature of each measurement feature type, and calculates a correlation value representing the strength of correlation between the team synchronization level calculated for each piece of learning data and the evaluation value of teamwork or psychological bond set in the learning data. The selection unit 23 selects a measurement feature type with a correlation value representing a strong correlation as a feature type for determination.

[0059] The model generation unit 24 generates a judgment model using the judgment feature type selected by the selection unit 23 for each activity type, or for each activity type and participant attribute. The output unit 25 reads out the judgment model generated by the model generation unit 24 from the storage unit 21 and outputs it. The output judgment model may be transmitted to the judgment device 3 or another computer device connected to the judgment device 3, or may be written to a recording medium.

[0060] The determination device 3 includes a storage unit 31, a display unit 32, an input unit 33, an acquisition unit 34, a receiving unit 35, a determination unit 36, and a feedback unit 37. The determination device 3 can be realized by a computer device such as a server computer, a personal computer, a smartphone, a tablet, or a dedicated device.

[0061] The storage unit 31 stores the judgment model, the feature amounts of each participant of the team measured by the measurement device 4, and output control information corresponding to the output conditions. The output conditions are a combination of an activity type and a judgment level, and a combination of an activity type, a participant attribute, and a judgment level. The output control information indicates feedback corresponding to the output conditions, or an action for adjusting teamwork or psychological bonds.

[0062] As an example, the output control information corresponding to the output condition of the activity type being participation in sports and the judgment level being low is set to output a message including a message that teamwork is declining and useful advice for improving teamwork, and the output control information corresponding to the output condition of the activity type being participation in sports and the judgment level being high is set to output a message including a message that teamwork is high and useful advice for maintaining high teamwork. In addition, there are various methods for improving teamwork, such as, for example, encouraging physiological signals to be adjusted, playing music, letting people smell fragrances, exercising together as a team, taking a break together as a team, eating sweets together as a team, and the like, but the effective method differs depending on the activity type and the participant attributes. Therefore, the output control information can be determined using the activity type and the participant attributes in addition to the judgment level.

[0063] The output condition may further include an additional condition. The additional condition may be, for example, the state of the participant based on the feature amount measured by the measurement device 4, the environment of the team, the operating state of the equipment used by the team participants, the elapsed time from the start of the activity, etc. As a result, even if the activity type is participation in e-sports and the judgment level is low, for example, based on the feature amount such as the pupil diameter measured by the measurement device 4, when the degree of eye fatigue is low, an action such as playing a predetermined music is performed, and when the degree of eye fatigue is high, an announcement encouraging a break is played or a message is output. It is possible to change the content of the feedback.

[0064] The input unit 33 is a keyboard, a mouse, a button, a touch panel, or the like, and receives information input by a user's operation.

[0065] The acquisition unit 34 acquires the judgment model generated by the judgment model generating device 2 and stores it in the storage unit 31. The acquisition unit 34 may receive the judgment model from the judgment model generating device 2 or a computer device connected via a network, or may read it from a recording medium. The acquisition unit 34 may acquire a judgment model according to the activity type and participant attributes input by the input unit 33, and store it in the storage unit 31.

[0066] The receiving unit 35 receives data on the features and measurement times obtained by measuring each participant in the team from the measurement device 4, and writes the data as the received features and the received feature measurement times in the storage unit 31. A received feature type indicating the type of the received feature is added to the data on the received feature and the received feature measurement times.

[0067] The judgment unit 36 ​​reads out from the storage unit 31 a judgment model corresponding to the activity type input by the input unit 33, or the activity type and participant attributes. The judgment unit 36 ​​reads out from the storage unit 31 in real time the reception feature amount and the reception feature amount measurement time of the reception feature amount type that is the same as the judgment feature amount type set in the read judgment model. The judgment unit 36 ​​calculates the synchronization level of the team using the reception feature amount and the reception feature amount measurement time of each participant of the team that have been read out. The judgment unit 36 ​​reads out from the judgment model a judgment level that corresponds to the calculated synchronization level, and sets it as a judgment result.

[0068] The feedback unit 37 outputs the judgment result obtained by the judgment unit 36. That is, the feedback unit 37 reads out output control information corresponding to the activity type, participant attributes, additional conditions, and judgment result from the storage unit 31. The participant attributes and additional conditions may not be used. The feedback unit 37 outputs the judgment result and takes measures to improve teamwork and psychological bonds in accordance with the read output control information.

[0069] For example, the feedback unit 37 visually, audibly, tactilely, etc. provides feedback on the determination result. That is, the feedback unit 37 may provide feedback not only visually, but also through hearing or touch depending on the time, place, and occasion, or may provide feedback by combining these.

[0070] Also, for example, the feedback unit 37 may modify or strengthen the teamwork or psychological bond by influencing the internal and external environments of individuals and teams through the five senses (sight, hearing, touch, smell, taste) based on the judged level of teamwork or psychological bond detected in real time from physiological signals or social signals between multiple people. For example, the feedback unit 37 may automatically control air conditioning, lighting, music, etc. during collaboration, or output suggestions for nutritional intake, exercise, and rest strategies. Alternatively, the feedback unit 37 may modify the synchronization level to adjust and strengthen the teamwork or psychological bond.

[0071] The measurement device 4 measures the feature amounts of each participant of the team. Any existing device can be used as the measurement device 4. For example, the measurement device 4 is a measuring device for measuring heart rate, respiration, physiological sweating, blood components, etc., attached to the participant, a measuring device for measuring surface body temperature using a thermograph, a pressure gauge placed on the back of the participant's chair, etc. Also, for example, the measurement device 4 may be a computer terminal that calculates the scores of each facial expression such as pupil diameter, joint position, posture, gesture, and smile based on an image taken by a camera, or may be a device that measures sound volume. A plurality of different feature amounts may be measured for one participant by a plurality of measurement devices 4. The measurement device 4 outputs the measured feature amounts to the judgment device 3 in real time.

[0072] The judgment model generating device 2 may be realized by a plurality of computer devices connected to a network. In this case, it is possible to arbitrarily determine which of the plurality of computer devices each functional unit of the judgment model generating device 2 is realized by. For example, the storage unit 21, or the storage unit 21 and the learning data acquiring unit 22, and other functional units may be realized by different computers connected by a network. Also, the same functional unit of the judgment model generating device 2 may be realized by a plurality of computer devices.

[0073] Similarly, the determination device 3 may be realized by a plurality of computer devices connected to a network. In this case, it is possible to arbitrarily determine which of the plurality of computer devices each functional unit of the determination device 3 is realized by. For example, the memory unit 31 of the determination device 3 may be realized by a server computer on a network such as the Internet, and other functional units of the determination device 3 may be realized by a personal computer, tablet, smartphone, or the like of a user. In addition, the memory unit 31, the acquisition unit 34, the determination unit 36, and the feedback unit 37 of the determination device 3 may be realized by a server computer on a network such as the Internet, and the display unit 32, the input unit 33, and the receiving unit 35 of the determination device 3 may be realized by a personal computer, smartphone, tablet, or the like of a subject. In addition, the display unit 32 of the determination device 3 may be provided in the measurement device 4.

[0074] Moreover, the determination device 3 and the measurement device 4 may be integrated into one device. Moreover, the determination model generating device 2 and the determination device 3 may be integrated into one device, or the determination model generating device 2, the determination device 3, and the measurement device 4 may be integrated into one device. In this case, too, as described above, the functions of the determination model generating device 2 and the functions of the determination device 3 may be realized by multiple computer devices connected to a network.

[0075] Next, a description will be given of the operation of the determination system 1. Fig. 20 is a flow diagram showing the process of the determination model generating device 2.

[0076] First, the learning data acquisition unit 22 of the determination model generating device 2 acquires a plurality of pieces of learning data (step S11). The learning data acquisition unit 22 writes the acquired learning data into the storage unit .

[0077] The selection unit 23 selects one activity type for which a model is to be generated (step S12). The selected activity type is described as a target activity type. Furthermore, the selection unit 23 selects one participant attribute for which a model is to be generated from among participant attributes that can be taken according to the target activity type (step S13). The selected participant attribute is described as a target participant attribute.

[0078] Next, the selection unit 23 selects one of the unselected measurement feature types that are candidates for use in the judgment model as a processing target (step S14). For example, heart rate synchronization is considered to be an effective feature for expressing teamwork and psychological bonding in any event. Therefore, at least the heart rate may be included in the measurement feature types that are candidates for use in the judgment model. The selected measurement feature type is described as a target measurement feature type. Note that one or more activity types for which a model is to be generated, one or more participant attributes for which a model is to be generated according to the activity type, and one or more measurement feature types that are candidates for use in the judgment model may be set in the judgment model generation device 2 by inputting them in advance by the user, or may be all activity types, participant attributes, and measurement feature types read out from the learning data.

[0079] The selection unit 23 reads out the learning data in which the target activity type, the target participant attribute, and the target measurement feature type are set (step S15). For each of the learning data read out in step S15, the selection unit 23 reads out the measurement feature of the target measurement feature type and the feature measurement time from each participant data, and performs a process of calculating the synchronization level for each predetermined interval in each team (step S16). Specifically, the selection unit 23 generates all combinations consisting of two different participant data using each participant data set in the learning data. For each generated combination, the selection unit 23 calculates the synchronization level of the measurement feature for each predetermined interval using the measurement feature of the target measurement feature type of each participant data. The selection unit 23 determines the average or median of the synchronization levels for each predetermined interval obtained for all combinations as the synchronization level for each predetermined interval in the team. It is assumed that the predetermined interval is set in advance in the judgment model generation device 2.

[0080] Next, for each of the learning data read out in step S15, the selection unit 23 associates the team synchronization level at each predetermined interval obtained in step S16 with the evaluation value of teamwork or psychological bond at each evaluation value measurement time (here, described as teamwork evaluation value) based on the time information. The selection unit 23 calculates a correlation value representing the correlation between the synchronization level of the measurement feature of the target measurement feature type and the teamwork evaluation value based on the correspondence between the synchronization level at each predetermined interval obtained for each learning data and the teamwork evaluation value (step S17). The selection unit 23 uses any one of correlation analysis, multiple regression analysis, and canonical correlation formula to calculate the correlation value depending on the case. In the correlation analysis, Pearson or Spearman correlation analysis is performed on the relationship between the synchronization level of one type of feature and one type of teamwork evaluation value, and a correlation coefficient (R value) and a risk ratio (P value) are obtained as the degree of association. In the multiple regression analysis, a multiple regression analysis is performed between the synchronization level of one type of feature or one type of teamwork evaluation value and multiple types of teamwork evaluation values ​​or synchronization levels of multiple types of feature values, and a multiple correlation coefficient (R2 value), standard partial regression coefficient, and partial correlation coefficient are obtained as the degree of association. In the canonical correlation analysis, a canonical correlation analysis is performed between a synchronization index group consisting of the synchronization levels of multiple types of feature values ​​and a teamwork index group consisting of multiple types of teamwork evaluation values, and the degree of association between each group of canonical correlation coefficients is obtained.

[0081] A specific example using correlation analysis will be described. It is assumed that the measurement feature of the target measurement feature type is feature X of feature type X. The selection unit 23 calculates the average value of the synchronization level of feature X calculated in step S16 for each of the learning data for each time period separated by the evaluation value measurement time. Specifically, for a certain learning data D1, it is assumed that the synchronization levels of the team's feature X are obtained every minute as L1, L2, L3, L4, L5, L6, ..., and that the learning data D1 has a teamwork evaluation value of type V (e.g., team work, coordination, IOS, etc.) set for each of N participants every 30 minutes. The teamwork evaluation values ​​of type V for the nth participant (n is an integer between 1 and N) every 30 minutes are V1n, V2n, V3n, .... In this case, the selection unit 23 calculates an average value A1 of the synchronization levels L1 to L30 from 1 to 30 minutes from the start, an average value A2 of the synchronization levels L31 to L60 from 31 to 60 minutes from the start, an average value A3 of the synchronization levels L61 to L90 from 61 to 90 minutes from the start, ..., and calculates an average value B1 of the teamwork evaluation values ​​V11 to V1N after 30 minutes from the start, an average value B2 of the teamwork evaluation values ​​V21 to V2N after 60 minutes from the start, an average value B3 of the teamwork evaluation values ​​V31 to V3N after 90 minutes from the start, .... As a result, a correspondence of (average value of the synchronization level of the feature X, average value of the teamwork evaluation value of type V) = (A1, B1), (A2, B2), (A3, B3), ... is obtained for each time period separated by the evaluation value measurement time. The selection unit 23 similarly obtains an average value of the synchronization level of the feature X and an average value of the teamwork evaluation value of type V for each time period separated by the evaluation value measurement time for other learning data. The selection unit 23 performs a Pearson or Spearman correlation analysis using the correspondence between the average value of the feature X obtained from each learning data read in step S15 and the average value of the teamwork evaluation value of type V, and obtains a correlation coefficient (R value) and a risk ratio (P value) as the relevance of the feature X.

[0082] The selection unit 23 judges whether or not all of the measurement feature types that are candidates for use in the teamwork judgment model have been selected (step S18). If it is judged that there is an unselected measurement feature type (step S18: NO), the selection unit 23 repeats the process from step S14.

[0083] When the selection unit 23 determines that all candidate measurement feature types to be used in the teamwork judgment model have been selected (step S18: YES), it determines whether or not all participant attributes for which a model is to be generated according to the target activity type have been selected (step S19).When the selection unit 23 determines that there are unselected participant attributes (step S19: NO), it repeats the process from step S13.

[0084] When the selection unit 23 determines that all participant attributes for which a model is to be generated according to the target activity type have been selected (step S19: YES), the selection unit 23 determines whether all activity types for which a model is to be generated have been selected (step S20). When the selection unit 23 determines that there is an unselected activity type (step S20: NO), the selection unit 23 repeats the process from step S12.

[0085] When the selection unit 23 determines that all activity types for which a model is to be generated have been selected (step S20: YES), it selects one from among all combinations of the activity types for which a model is to be generated and the participant attributes for which a model is to be generated corresponding to the activity types (step S21). For the combination of the activity type and the participant attribute selected in step S21, the selection unit 23 selects a measurement feature type for which the correlation value calculated in step S17 indicates a higher correlation than a predetermined value, and sets the measurement feature type as a determination feature type (step S22). For example, the selection unit 23 may set a measurement feature type for which a correlation value indicating a higher correlation than a predetermined value has been obtained as the determination feature type, or may set a predetermined number of measurement feature types (one or more) from those for which a correlation value indicating the highest correlation has been obtained as the determination feature type.

[0086] The model generation unit 24 generates a judgment model of the activity type and participant attribute selected in step S21 by using the judgment feature type selected in step S22 (step S23). Note that, when multiple judgment feature types are selected in step S22, the model generation unit 24 generates a judgment model for each judgment feature type.

[0087] For example, the evaluation value of teamwork or psychological bond is divided into judgment levels of low, medium, and high teamwork in advance, and the evaluation value a1 or less is low teamwork, the evaluation value a1 to a2 is medium teamwork, and the evaluation value a2 or more is high teamwork. The model generation unit 24 estimates a relational expression representing the relationship between the evaluation value of teamwork or psychological bond and the judgment feature from the correspondence between the synchronization level used when calculating the correlation value in step S17 and the evaluation value of teamwork or psychological bond, and calculates the synchronization level corresponding to the judgment level by substituting the evaluation value of teamwork or psychological bond corresponding to the judgment level into the estimated relational expression. When the model generation unit 24 calculates the synchronization levels b1 and b2 of the judgment feature corresponding to the evaluation values ​​a1 and a2, the synchronization level b1 or less of the judgment feature is low teamwork, the synchronization levels b1 to b2 of the judgment feature are medium teamwork, and the synchronization level b1 or less is high teamwork. The model generation unit 24 generates a judgment model in which the judgment feature type and the level classification condition indicating the range of the synchronization level corresponding to each judgment level are set. The model generation unit 24 further adds the activity type and the participant attributes to the judgment model, and stores it in the storage unit 21. Note that the model generation unit 24 may set the calculation used to calculate the synchronization level in step S16 or the type of the calculation (e.g., cross-correlation coefficient, Wavelet coherence analysis, etc.) for the measurement feature of the measurement feature type selected as the judgment feature type.

[0088] The selection unit 23 determines whether or not all combinations of the activity type and the participant attributes that can be taken according to the activity type have been selected (step S24). If the selection unit 23 determines that there are unselected combinations (step S24: NO), it repeats the process from step S21. If the selection unit 23 determines that all combinations have been selected (step S24: YES), it outputs the generated judgment model (step S25). The acquisition unit 34 of the judgment device 3 acquires the judgment model output from the judgment model generation device 2 and stores it in the memory unit 31.

[0089] In the above, the model generation unit 24 sets the level classification conditions indicating the range of the synchronization level corresponding to each judgment level in the judgment model, but the user may manually generate the level classification conditions and set them in the judgment model. Also, when a feature amount effective for judging teamwork or psychological bonds in collaborative activities, such as heart rate, is known in advance, the judgment model may be manually created using the feature amount and provided to the judgment device 3.

[0090] Furthermore, the judgment model generating device 2 may provide the user with information for determining whether or not features of a predetermined feature type are effective for determining teamwork or psychological bond. In this case, in step S11, learning data obtained by a demonstration experiment is input to the judgment model generating device 2. The judgment model generating device 2 performs processing of steps S12 to S21 on the input learning data. For the combination of the activity type and participant attribute selected in step S21, the judgment model generating device 2 presents the correlation value calculated for each measured feature type in step S17 to the user, for example by displaying it on a display. The user checks whether the feature type used in the demonstration experiment is effective for determining the level of teamwork or psychological bond based on the correlation value for each measured feature type.

[0091] If there is only one activity type, the processes of steps S12 and S20 can be omitted. If there is only one participant attribute type, the processes of steps S13 and S19 can be omitted. If there is only one measurement feature type, the processes of steps S14 and S18 can be omitted.

[0092] 21 is a flow diagram showing an example of the processing of the determination device 3. The acquisition unit 34 of the determination device 3 displays the activity type, participant attribute, and measurement feature type of each determination model stored in the storage unit 31 on the display unit 32. The user refers to the displayed activity type, participant attribute, and measurement feature type, selects a determination model according to the activity of the team to be evaluated, the attributes of the participants belonging to that team, and the type of feature measured by the available measurement device 4, and inputs the selection result by the input unit 33 (step S31). The acquisition unit 34 reads out the selected determination model from the storage unit 31 and outputs it to the determination unit 36.

[0093] In addition, when only one judgment model is stored, the process of step S31 may be omitted. In addition, the acquisition unit 34 may display the activity type, participant attribute, and measurement feature type of each judgment model stored in the storage unit 21 of the judgment model generation device 2 on the display unit 32, and read out the judgment model selected by the user via the input unit 33 from the judgment model generation device 2.

[0094] The receiving unit 35 of the determination device 3 starts receiving the feature of each participant measured by the measurement device 4 (step S32). That is, the measurement device 4 transmits data of the feature measured for each participant belonging to the team to the determination device 3 in real time. The type of feature measured by the measurement device 4 is the determination feature type used in the determination model selected in step S31. The receiving unit 35 of the determination device 3 writes the received feature, which is the feature received from the measurement device 4, in the storage unit 31 in association with the received feature type and the received feature measurement time.

[0095] When the input unit 33 inputs a command to start the judgment (step S33), the judgment unit 36 ​​continuously calculates a moving average of the synchronization level of the team for each predetermined period by using the reception feature of the same reception feature type as the judgment feature type indicated by the judgment model. The predetermined period is, for example, 1 to 5 minutes. Specifically, the judgment unit 36 ​​calculates the synchronization level of the team every time a predetermined time has elapsed (step S34) after the judgment start is input (step S35). That is, the judgment unit 36 ​​reads out the reception feature of each participant belonging to the team for the synchronization level calculation target period from the storage unit 31. The synchronization level calculation target period is a period from the end of the period in which the reception feature for calculating the previous team synchronization level was obtained to the time when a predetermined time has elapsed. The judgment unit 36 ​​generates all combinations of two participants for the participants belonging to the team, and calculates the synchronization level of the reception feature in the synchronization level calculation target period for each generated combination. The judgment unit 36 ​​sets the average or median of the synchronization level obtained for each combination of participants as the synchronization level of the team.

[0096] The judgment unit 36 ​​reads out the judgment level of the teamwork and psychological bond corresponding to the synchronization level of the team from the level classification conditions set in the judgment model, and sets it as the judgment result (step S36). For example, the synchronization level takes a value of -1.0 to 1.0. The judgment unit 36 ​​judges that the teamwork and psychological bond are low level when the synchronization level of the team is less than 0.2, medium level when the synchronization level of the team is 0.3 to 0.4, and high level when the synchronization level of the team is more than 0.4. The judgment unit 36 ​​writes the synchronization level calculation target period and the judgment result in the storage unit 31 in association with each other.

[0097] The feedback unit 37 outputs the determination result acquired by the determination unit 36 ​​(step S37). The feedback unit 37 reads out output control information corresponding to the activity type, participant attributes, additional conditions, and the determination result from the storage unit 31. The participant attributes and additional conditions may not be used. The feedback unit 37 outputs the determination result and takes measures to improve teamwork and psychological bonds according to the read output control information. When using additional conditions, the feedback unit 37 calculates the additional conditions based on the received feature amount during the synchronization level calculation target period during which the determination result was obtained. Alternatively, the feedback unit 37 may obtain information on the team's environment, such as temperature, humidity, brightness, sound volume, and movement speed, as additional conditions from equipment that controls or measures the team's environment, or may obtain information on the state of the equipment and the elapsed time from the start of the activity from equipment used by the team participants, as additional conditions.

[0098] For example, the feedback unit 37 visually, audibly, tactilely, or the like, and then provides feedback on the judgment result. That is, the feedback unit 37 may not only provide feedback on the judgment result in a visual manner, but may also provide feedback through hearing or touch, or may combine these, depending on the time, place, and occasion. For example, the feedback unit 37 displays the judgment result on a device used by each participant or on a display installed in a position visible to all participants, using a message or color corresponding to the judgment result. Alternatively, the feedback unit 37 may output the judgment result by sound, or may play music corresponding to the judgment result. The feedback unit 37 may vibrate devices such as a controller, mouse, or steering wheel used by each participant, or a chair on which the participant is sitting, with a rhythm or magnitude corresponding to the judgment result. The feedback unit 37 may also combine these.

[0099] In addition to or instead of the above, the feedback unit 37 may act on each participant and team from the internal and external environment by one or more of sight, hearing, touch, smell, and taste based on the judgment level. For example, the feedback unit 37 controls the air conditioning, lighting, music, etc. of the room where the team is working together based on the judgment level. For example, when the judgment result is a low judgment level, the feedback unit 37 may calculate the frequency of the change in the reception feature amount (for example, the frequency of the heart rate) based on the reception feature amount obtained during the period when the judgment level was high, and change the brightness of the lighting according to the calculated frequency, output music with a rhythm of the calculated frequency, or vibrate the equipment used by the participant or the chair on which the participant is sitting at the calculated frequency. The feedback unit 37 may also emit a scent that brings about a relaxing effect, a refreshing effect, etc. to the place where the participant is located according to the judgment level.

[0100] Furthermore, the feedback unit 37 does not perform control when the judgment level of the teamwork or psychological bond obtained as the judgment result is medium or high, and when the judgment level is low, the feedback unit 37 may display a message on the display proposing strategies such as nutrition and fluid intake, exercise, rest, and eating snacks for the team according to additional conditions such as fatigue, or may output such suggestions by voice. In this way, the feedback unit 37 modifies the synchronization level and adjusts and strengthens the teamwork and psychological bond.

[0101] In addition, the feedback unit 37 may continue the process of step S37 currently being performed when the same judgment level as the previous judgment level is obtained in step S36, and may perform the process of step S37 using the new judgment level when the judgment level different from the previous judgment level is obtained. The judgment device 3 judges whether or not an end instruction has been input (step S38). When the judgment device 3 judges that an end instruction has not been input (step S38: NO), it repeats the process from step S34, and when it judges that an end instruction has been input (step S38: YES), it ends the process of FIG. 21.

[0102] The determination device 3 may perform the processes from step S32 onward for each team. This makes it possible to output an evaluation result for each team competing in a sport. Also, by dividing participants in the same live event into teams based on their location in the venue, and the determination device 3 controlling the lighting for each team based on the evaluation result, it is possible to visualize which locations are unified and excited.

[0103] In addition, the user may select judgment models of different judgment feature types for the same set of activity type and participant attribute in step S31. The judgment device 3 may perform the processes of steps S32 to S36 for each of the different judgment models to obtain a judgment level for the team, and may combine the obtained judgment levels to determine the judgment level for the team. For example, when two judgment models are selected, the judgment unit 36 ​​may determine the better of the judgment levels using these models as the judgment level for the team.

[0104] Note that the user may rewrite, for example, the judgment feature type and the level classification conditions of the judgment model stored in the storage unit 31 by the input unit 33. This makes it possible to judge teamwork and psychological bonds using new types of feature values ​​and level classification conditions without generating a new judgment model by the judgment model generation device 2.

[0105] According to this embodiment, the determination device 3 acquires physiological signals such as heart rate, surface body temperature, and pupil diameter, as well as non-verbal social signals such as facial expressions and postures, using wearable sensors and cameras in collaborative situations involving a large number of people in physical sports, team play in e-sports, and watching live sporting events. The determination device 3 performs cross-correlation function and wavelet coherence analysis on the acquired data to calculate the synchronization level in real time and evaluate teamwork and psychological bonds. The background to this is that in each situation, the level of physiological and social synchronization is related to the level of psychological distance, unity, loyalty, and teamwork (the number of assist plays and the overall game index) between collaborators.

[0106] When this embodiment is applied to e-sports or sports watching, the determination device 3 measures the physiological signals and non-verbal social signals of each e-sports team member or each spectator at a sports watching event in real time using measurement devices 4 such as sensors and cameras at the venue. When this embodiment is applied to an office or a drive, the determination device 3 measures the physiological signals and non-verbal social signals of each employee in the office or passenger in the car in real time using measurement devices 4 such as sensors and cameras in the office or vehicle. The determination device 3 enables detection and adjustment of teamwork and psychological bonds by feeding back the analysis results through the five senses or by influencing the internal and external environments.

[0107] The determination system 1 of this embodiment can be used to visualize and strengthen teamwork in sports teams, including motorsports and e-sports. The determination system 1 or the determination device 3 of this embodiment can be implemented as a new function in an analysis system for a heart rate system or an activity meter. The determination system 1 of this embodiment can be used to improve the sense of unity of participants in live events such as sports watching, music, and theater. For example, it is useful for developing a real-time visualization and strengthening system for the fun, reputation, and loyalty of live events. The determination system 1 of this embodiment can be used to detect, feedback, and adjust teamwork and psychological bonds in all human behavioral situations, such as in a car, an office, a classroom, an elderly care facility, a hospital, an online conference, and a metaverse. For example, it can be used in a communication promotion system aimed at eliminating loneliness, and a system for building an online environment, cyberspace, and metaverse that are rich in physicality.

[0108] 22 is a device configuration diagram showing an example of the hardware configuration of the determination model generating device 2 and the determination device 3. Each of the determination model generating device 2 and the determination device 3 includes a processor 71, a storage unit 72, a communication interface 73, and a user interface 74.

[0109] The processor 71 is a central processing unit that performs calculations and control. The processor 71 is, for example, a CPU. The processor 71 reads out and executes programs from the storage unit 72. The storage unit 72 further has a work area and the like for the processor 71 to execute various programs. The communication interface 73 is for connecting to other devices so as to be able to communicate with them. The user interface 74 is an input device such as a keyboard, a pointing device (mouse, tablet, etc.), a button, a touch panel, etc., a display device such as a display, and an audio output device such as a speaker or headphones.

[0110] At least some of the functions of the learning data acquisition unit 22, the selection unit 23, and the model generation unit 24 of the judgment model generation device 2, and the functions of the acquisition unit 34, the reception unit 35, the judgment unit 36, and the feedback unit 37 of the judgment device 3 are realized by the processor 71 reading and executing a program from the storage unit 72. The above program may be transmitted via an electric communication line or may be recorded on a portable recording medium. The above program may be for realizing some of the above-mentioned functions, and may further be capable of realizing the above-mentioned functions in combination with a program already recorded in a computer system. All or some of the above functions may be realized using hardware such as an ASIC (Application Specific Integrated Circuit), a PLD (Programmable Logic Device), or an FPGA (Field Programmable Gate Array). The storage unit 21 of the judgment model generation device 2 and the storage unit 31 of the judgment device 3 are realized by the storage unit 72.

[0111] According to the embodiment described above, the teamwork or psychological bond determination system includes a selection device and a teamwork or psychological bond determination device. For example, the selection device corresponds to the determination model generation device 2 of the embodiment.

[0112] The selection device includes a storage unit and a selection unit. The storage unit stores time-series measurement values ​​of multiple types of physiological signals or non-verbal signals obtained from multiple measurement subjects belonging to a group performing a collaborative activity, and an evaluation value quantitatively expressing teamwork or psychological bonds in the collaborative activity. The selection unit calculates a synchronization level of changes in the measurement values ​​in the group using the time-series measurement values ​​of the multiple measurement subjects for each of the multiple types of physiological signals or non-verbal signals, and selects a type of physiological signal or non-verbal signal to be used for determining the strength of teamwork or psychological bonds in the collaborative activity based on the strength of correlation between the calculated synchronization level and the evaluation value.

[0113] The teamwork or psychological bond assessment device includes a receiving unit and a judging unit. The receiving unit receives time series measurement values ​​of physiological signals or non-verbal signals obtained by measuring each of a plurality of subjects belonging to a group to be assessed who are engaged in collaborative activities. The physiological signals or non-verbal signals are of a type selected by a selecting unit of a selecting device. The judging unit calculates a synchronization level of changes in the measurement values ​​in the group using the time series measurement values ​​of the plurality of subjects belonging to the group to be assessed, and judges the strength of the teamwork or psychological bond in the collaborative activity in the group to be assessed based on the calculated synchronization level.

[0114] The teamwork or psychological bond determination device may further include a feedback unit that outputs the strength of the teamwork or psychological bond determined by the determination unit so that it can be perceived using one or more of vision, hearing, and touch. Alternatively, the feedback unit controls to provide the measurement subject with a visual, auditory, tactile, olfactory, or gustatory stimulus for adjusting the teamwork or psychological bond in the collaborative activity according to the strength of the teamwork or psychological bond determined by the determination unit.

[0115] For example, the physiological signal is a heart rate, mental sweating, surface body temperature, pupil diameter, blood components, or saliva components. For example, the non-verbal signal is a facial expression, a posture, a gesture, a respiration, or a voice volume.

[0116] Although an embodiment of the present invention has been described in detail above with reference to the drawings, the specific configuration is not limited to this embodiment, and designs that do not deviate from the gist of the present invention are also included. [Industrial Applicability]

[0117] It can be used to visualize and strengthen teamwork in sports teams, improve the sense of unity at live events such as sports, music, and theater, and detect, provide feedback on, and adjust teamwork and psychological bonds in any human behavior situation. [Explanation of symbols]

[0118] 1 Judgment system 2. Decision model generation device 3 Judgment device 4. Measurement equipment 21 Memory section 22 Learning data acquisition unit 23 Selection section 24 Model Generation Unit 25 Output section 31 Storage section 32 Display section 33 Input section 34 Acquisition Department 35 Receiving section 36 Judgment section 37 Feedback Section 71 Processor 72 Memory section 73 Communication Interface 74 User Interface

Claims

1. a receiving unit that receives time series measurement values ​​of physiological signals or non-verbal signals obtained by measuring each of a plurality of subjects belonging to a group that is engaged in a collaborative activity; a determination unit that calculates a synchronization level of changes in the measurement values ​​in the group using the time-series measurement values ​​of the plurality of measurement subjects, and determines the strength of teamwork or psychological bonds in the collaborative activity based on the calculated synchronization level; A determination device comprising:

2. A feedback unit that outputs the strength of the teamwork or psychological bond determined by the determination unit so that the strength of the teamwork or psychological bond can be perceived using one or more of visual, auditory, and tactile senses. The determination device according to claim 1 .

3. A feedback unit controls the measurement subject to provide a visual, auditory, tactile, olfactory or gustatory stimulus for adjusting the teamwork or psychological bond in the collaborative activity according to the strength of the teamwork or psychological bond determined by the determination unit. The determination device according to claim 1 .

4. The physiological signal is a heart rate, mental sweating, surface temperature, pupil diameter, blood constituents, or saliva constituents; The non-verbal signal is a facial expression, a posture, a gesture, a breathing or a vocalization volume. The determination device according to claim 1 .

5. a storage unit that stores time series measurement values ​​of multiple types of physiological signals or non-verbal signals obtained from multiple measurement subjects belonging to a group performing collaborative activities, and an evaluation value that quantitatively represents teamwork or psychological bonds in the collaborative activities; a selection unit that calculates a synchronization level of changes in the measurement values ​​in the group using the time-series measurement values ​​of the plurality of measurement subjects for each of the plurality of types of physiological signals or non-verbal signals, and selects a type of physiological signal or non-verbal signal to be used for determining the strength of teamwork or psychological bond in the collaborative activity based on the strength of correlation between the calculated synchronization level and the evaluation value; A selection device comprising:

6. a storage unit that stores time series measurement values ​​of multiple types of physiological signals or non-verbal signals obtained from multiple measurement subjects belonging to a group performing collaborative activities, and an evaluation value that quantitatively represents teamwork or psychological bonds in the collaborative activities; a selection unit that calculates a synchronization level of changes in the measurement values ​​in the group using the time-series measurement values ​​of the plurality of measurement subjects for each of the plurality of types of physiological signals or non-verbal signals, and selects a type of physiological signal or non-verbal signal to be used for determining the strength of teamwork or psychological bond in the collaborative activity based on the strength of correlation between the calculated synchronization level and the evaluation value; a receiving unit that receives time series measurement values ​​of the selected types of physiological signals or non-verbal signals obtained by measuring each of a plurality of measurement subjects belonging to a group to be judged who are engaged in a collaborative activity; a determination unit that calculates a synchronization level of changes in the measurement values ​​in the group to be determined using the time series measurement values ​​of the plurality of measurement subjects received by the receiving unit, and determines the strength of teamwork or psychological bonds in the collaborative activity of the group to be determined based on the calculated synchronization level; A determination system comprising:

7. A receiving step of receiving a time series measurement value of a physiological signal or a non-verbal signal obtained by measuring each of a plurality of subjects belonging to a group performing a collaborative activity; a determination step of calculating a synchronization level of changes in the measurement values ​​in the group using the time-series measurement values ​​of the plurality of measurement subjects, and determining the strength of teamwork or psychological bonds in the collaborative activity based on the calculated synchronization level; The determination method has the following.

8. an acquisition step of acquiring time series measurement values ​​of multiple types of physiological signals or non-verbal signals obtained from each of multiple measurement subjects belonging to a group performing a collaborative activity, and an evaluation value quantitatively expressing teamwork or psychological bonds in the collaborative activity; a selection step of calculating a synchronization level of changes in the measurement values ​​in the group using the time series measurement values ​​of the plurality of measurement subjects for each of the plurality of types of physiological signals or non-verbal signals, and selecting a type of physiological signal or non-verbal signal to be used for judging the strength of teamwork or psychological bond in the collaborative activity based on the strength of correlation between the calculated synchronization level and the evaluation value; A selection method having the following structure:

9. a receiving step of receiving time series measurement values ​​of multiple types of physiological signals or non-verbal signals obtained from each of multiple subjects belonging to a group performing a collaborative activity, and an evaluation value quantitatively representing teamwork or psychological bonding in the collaborative activity; a selection step of calculating a synchronization level of changes in the measurement values ​​in the group using the time series measurement values ​​of the plurality of measurement subjects for each of the plurality of types of physiological signals or non-verbal signals, and selecting a type of physiological signal or non-verbal signal to be used for judging the strength of teamwork or psychological bond in the collaborative activity based on the strength of correlation between the calculated synchronization level and the evaluation value; A receiving step of receiving time series measurement values ​​of the selected types of physiological signals or non-verbal signals obtained by measuring each of a plurality of measurement subjects belonging to a group to be determined who are engaged in a collaborative activity; a determination step of calculating a synchronization level of changes in the measurement values ​​in the group to be determined using the time-series measurement values ​​of the plurality of measurement subjects received in the receiving step, and determining the strength of teamwork or psychological bonds in the collaborative activity of the group to be determined based on the calculated synchronization level; The determination method has the following.

10. Computer, A program for causing the determination device according to claim 1 to function.

11. Computer, A program for causing the selection device according to claim 5 to function.

Images