Behavioral support system, behavioral support method, and behavioral support program

Abstract

To provide a technology for more effectively assisting with behavior modification of subject persons, without using personal information of the subject persons or acquiring, beforehand, knowledge such as information dissemination which is effective for the subject persons.SOLUTION: A behavior assistance system comprises: a storage unit that stores multiple message groups which are respectively associated with multiple behavior modification techniques that act on one or more of multiple behavior modification factors, with each of the message groups including one or more messages; a generation unit that extracts messages from each of one or more of the multiple message groups and generates for each subject person a message set composed of the extracted messages; and a control unit that performs control such that the messages within the message sets are output to the subject persons on one or more occasions.SELECTED DRAWING: Figure 5

Description

[Technical Field]

[0001] The present invention relates to an action support system, an action support method, and an action support program. [Background technology]

[0002] In recent years, we have developed services based on theories of behavioral science, which is the scientific study of human behavior. Efforts to utilize this technology for development are spreading to various fields, including public policy, healthcare, retail, and education. In addition, we will provide technical support for behavioral change and habit formation for subjects in various fields. Systems that realize this are also being considered (for example, Patent Document 1).

[0003] In the system described in Patent Document 1, various data measured on the behavior of multiple subjects are collected. Based on the results of the analysis of the behavioral data, the goal of forming a habit is Stage indicators are indicators that serve as standards for multiple stages that gradually lead to the desired behavior, and defining each of the plurality of stages according to a stage index; For each stage pair, identify the gap between the two stages that make up the pair. The reasons for the identified gaps and the high At least one of the measures to induce behavioral change to transition to the The reasons / measures for the gaps are identified from the relationship information that defines the relationship between the gaps and the reasons / measures. It is stated that the process for the reasons / measures identified for each stage pair will be carried out. There are. [Prior art documents] [Patent documents]

[0004] [Patent Document 1] Japanese Patent Application Publication No. 2020-140596 Summary of the Invention [Problem to be solved by the invention]

[0005] In order to encourage behavioral change in subjects, personal information such as the subjects' attributes, income, and residential area is used. In this case, it is thought that implementing individualized interventions according to the behavioral characteristics of the subjects would be effective. However, using the personal information of the subjects raises the following problems. Acquiring personal information requires high levels of security and compliance with laws and regulations. Second, it is difficult to actually identify the behavioral characteristics of the target person. The personal information required for this purpose is often not clear in advance. Even if personal information is obtained beyond the scope of the law, the personal information will not be used to design and operate behavior change programs. Third, the personal data obtained for behavior change interventions may not be utilized effectively. The information is personal information of the subject at a certain point in time in the past, and indicates the subject's condition at the time of intervention. Therefore, interventions based on such past personal information may not be as expected. There is a risk that the expected results will not be achieved.

[0006] Therefore, the present invention uses the personal information of the subject and provides knowledge of effective methods for the subject. Behavioral support that more effectively supports behavioral change in the target person without acquiring prior knowledge One of the objectives is to provide a system, a behavioral support method, and a behavioral support program. [Means for solving the problem]

[0007] A behavior support system according to one aspect of the present invention is a behavior support system that supports behavioral change of a subject. A system that includes multiple actions that affect one or more of the multiple behavioral change factors. A plurality of message groups each associated with a dynamic transformation technique, a storage unit for storing a plurality of message groups each including one or more messages; extracting a message from each of one or more of the message groups; A message generator generates a message set consisting of messages for each target person, and Control to output messages in the set to the target person once or multiple times It comprises a section.

[0008] According to this aspect, a message set that acts on a plurality of behavioral change factors is transmitted to a subject. Therefore, we may use the personal information of the subject and provide effective information to the subject. It is possible to effectively encourage behavioral change in subjects without having to acquire prior knowledge of become.

[0009] In the above aspect, the generation unit selects a specific behavior modification technique from among the plurality of behavior modification techniques. Behavior change techniques that only affect the behavioral factors were selected for all of the multiple behavior change factors. Select one behavior change technique from each of the messages linked to the selected technique. The first message set may be generated by extracting the two messages. According to the embodiment, a comprehensive set of messages that affect multiple behavioral change factors is sent to the subject. It is possible.

[0010] In the above embodiment, response information from the target person in response to the output message is acquired. and an acquisition unit for acquiring the response information from among the plurality of behavioral change factors based on the response information. an analysis unit that extracts the behavioral change factors to which the subject responded when the sponsor information was acquired; and The generation unit further generates a response information for the target person from whom the response information is acquired. Select a behavior change technique that will affect the behavior change factors to which the subject responded, and implement the selected behavior change. Extract one or more messages from each of the message groups associated with each technique. According to this aspect, the second message set may be generated by issuing the Interventions should be tailored to each individual, taking into account the influence of the passage of time and changes in the environment on factors that affect behavior change. This makes it possible to more effectively encourage behavioral change in the target person.

[0011] In the above aspect, the generation unit randomly extracts a message from each message group. That's fine.

[0012] In the above aspect, the plurality of behavioral change factors are abilities, opportunities, and , and motivation.

[0013] In the above aspect, the plurality of behavioral change factors include an experiential attitude, an instrumental attitude, an instructive norm, , descriptive norms, sense of behavioral control, self-efficacy, knowledge, skills, importance of behavior, environmental It may include at least two of the following: constraints and habituation.

[0014] A behavior support method according to another aspect of the present invention is a behavior support method for supporting a behavior change of a subject. Multiple behavioral change factors that affect one or more of the multiple behavioral change factors Multiple message groups each associated with a technique, each containing one From each of one or more message groups, including the above messages, Messages are extracted from the message set, and a message set consisting of the extracted messages is created for each target person. and generating the messages in the message set one or more times in correspondence with each other. and a step of controlling the image data to be output to the subject.

[0015] A behavioral support program according to another aspect of the present invention is a computer program that supports behavioral change in a subject. A behavioral support program that is implemented by a patient, which involves the implementation of one or more of multiple behavioral change factors. Multiple messages linked to multiple behavior change techniques that act on behavior change factors A message group consisting of a plurality of message groups, each message group containing one or more messages. That is, extracting a message from each of one or more message groups, and generating a message set for each target person, the message set including the following messages; and controlling the device to output the message to the subject in one or more separate messages. do. [Effects of the Invention]

[0016] According to the present invention, the personal information of the subject can be used, and effective information can be provided to the subject. It is possible to more effectively support behavioral change in subjects without acquiring knowledge in advance. can. [Brief explanation of the drawings]

[0017] [Figure 1] 1 is a table illustrating a matrix format of frames that associate behavior change factors with behavior change techniques. [Figure 2] FIG. 10 is a diagram conceptually illustrating the data structure of a nudge message group. [Figure 3] 1 is a diagram illustrating an example of a schematic configuration of a behavior support system according to an embodiment of the present invention. [Figure 4]FIG. 1 is a diagram illustrating an example of the hardware configuration of devices in a behavior support system according to an embodiment of the present invention. [Figure 5] FIG. 2 is a diagram illustrating an example of the functional configuration of devices in the action support system according to the present embodiment. [Figure 6] 10 is a flowchart illustrating an example of the operation of the action support system according to the present embodiment. [Figure 7] FIG. 1 is a schematic diagram showing a specific example of a MECE-type nudge cocktail. [Figure 8] FIG. 1 is a schematic diagram showing a specific example of a MECE-type nudge cocktail. [Figure 9] FIG. 1 is a schematic diagram showing a specific example of an adaptive nudge cocktail. DETAILED DESCRIPTION OF THE INVENTION

[0018] An embodiment of the present invention will be described with reference to the accompanying drawings. Those marked with the same or similar symbols have the same or similar configurations.

[0019] (Overview of behavioral support methods) The behavioral support method according to this embodiment is a method for scientifically identifying important behavioral change factors for a person who is the subject of intervention. - Nudge messages that effectively influence the factors that change the behavior of the target person while objectively exploring them. The message is sent to the target person to encourage behavioral change. In the method, a plurality of behavioral change factors and one or more of the plurality of behavioral change factors are included. A framework is used that links multiple behavior change techniques that work on the following:

[0020] <frame> Here, behavioral change factors are factors that cause the target person to take a specific action. The factors for behavioral change are based on academic research or surveys such as attitude surveys, theories on behavioral science, and the subject's Based on persona setting or behavioral process maps, multiple actions in the target field can be The factors of change can be identified.

[0021] An example of a behavioral change factor is the " COM-B model, which is commonly used in behavioral science. List "Capacity," "Opportunity," and "Motivation" Alternatively, as another example of a behavioral change factor, the Integrated Behavior Model can be used. The factors defined in the IBM (Innovative Model) (also known as "IBM factors") are: experiential attitude, instrumental attitude, directive Injunctive norm, descriptive norm, and sense of control perceived control), self-efficacy, knowledge, skills ), salience of the behavior, environmental constraints In this embodiment, the following two methods are also known: You may identify behavior change drivers that include at least two of these.

[0022] The above IBM Factors are merely examples, and are not intended to be a comprehensive list of higher-level factors that include the above IBM Factors. Factors may be identified as behavior change factors. For example, the experiential and instrumental attitudes mentioned above are The above indicative and descriptive norms are included in "Attitude" and are classified as "Perceived Norms." The sense of control and self-efficacy are included in the "personal norm" and The above knowledge and skills are included in "Knowledge" and the agency of action. Importance is included in "Importance," and environmental constraints are included in "Friction." Alternatively, the lower level factors obtained by dividing the above IBM factors may be included in the behavior change requirements. It may be identified as a cause.

[0023] Of course, in this embodiment, the COM-B model and IBM are used as factors for behavioral change. What factors related to human behavioral intentions are adopted, such as factors determined by external behavioral models? is also good.

[0024] The means or techniques that act on (apply to) the above-mentioned factors for behavioral change are called "behavioral change." It is called "Behavior Change Technique (BCT)" or intervention. It is sometimes called an intervention function. In the following, this means or The technique is also referred to as "BCT."

[0025] BCTs (or interventions) that affect behavior change factors are theoretically known. For example, Lou Atkins and Susan Michie (UCL Center for Behavior Change, University College London), "Designing interventions to change eating behaviours" A matrix showing the relationship between behavior change factors and intervention effects in the B model is disclosed. ("Table 2. Matrix of links between COM-B model and intervention functions" reference).

[0026] The frame used in this embodiment is based on the theory exemplified above. In addition to theory, academic research or surveys such as opinion polls, persona setting for subjects, may be developed based on a behavioral process map or the like.

[0027] Figure 1 shows a matrix of the framework that links behavior change factors and behavior change techniques. In Figure 1, there are three behavioral change factors: "ability," "opportunity," and Five BCTs for "motivation": "motion instruction," "social support," "social norms," ​​and "step-by-step tasks." The strength of the effects of "Screen Setting" and "Behavioral Observation" is expressed in vector form. Figure 1 In the matrix shown in Figure 1, the sign "+" indicates that an effect is observed, and the numerical value indicates the degree of effect. It shows strength.

[0028] In Figure 1, for example, the BCT's "action instructions" only apply to the behavioral change factor "ability." It has also been shown that "step-by-step task setting" has an effect on the behavioral change factor "ability." It has been shown that it has a strong effect on "opportunity" (+1) and also on "opportunity" (+1). It has been shown that "behavioral observation" does not affect any of the three factors for behavioral change shown in Figure 1. There are.

[0029] Below, we refer to the vector that represents the effect of each BCT on the behavioral change factors as the “BCT vector.” Also, a specific BCT vector is referred to as a "movement vector" using the name of the BCT. As shown in Figure 1, three The action instruction vector for the behavioral change factors "ability," "opportunity," and "motivation" is <+1,0,0> It is expressed as:

[0030] <Nudge message> In the behavioral support method according to this embodiment, a plurality of behavioral changes related to the target field of intervention are Several BCTs that affect one or more of the factors that contribute to behavioral change are identified, and these BCTs A message group is linked to each of the BCTs. It includes one or more embodied nudge messages (hereinafter simply referred to as messages). For example, the BCT "operation instructions" could be something like "Please do ____ by ____ date." A message instructing an action is linked.

[0031] FIG. 2 is a diagram conceptually illustrating the data structure of a nudge message group. In Figure 1, multiple messages are linked to each of the five types of BCTs. Each message is assigned an identification code (e.g., message ID). In addition, the messages associated with each behavior change technique are A code (for example, a group ID) may be assigned to identify the message group to which it belongs. stomach.

[0032] When implementing an intervention, it is important to consider the status of the intervention among the multiple behavior change factors related to the target area. One or more BCTs that act on the identified behavior change factors are selected according to the message, and the selected A message is extracted from each message group associated with the BCT. A message set consisting of the messages is generated.

[0033] The selection of messages from each message pool is preferably random for each subject. By incorporating randomness in the generation of message sets, intervention designers This avoids selection bias, allowing intervention designers to select nudges based on intuition and experience. Compared to conventional intervention designs, it is possible to obtain stable intervention effects. Many intervention designers are motivated by messages they have used before or are well-known for, rather than by scientific evidence. This is because they tend to use different types of messages.

[0034] However, some messages may have a stronger impact on the target audience than others. It is possible that messages that are considered to be significantly larger (best messages) may be included. In such a message set, the best message is treated equally to the other messages. Randomly selecting messages may lead to missed opportunities for intervention. Therefore, we use a reinforcement learning method such as the Epsilon-Greedy method to extract messages from each message group. Here, the Epsilon-Greedy method is to act randomly with probability ε. This is a method of searching by probability (1-ε) and selecting the action with the highest expected value. Specifically, from a set of messages, the best message is selected with probability (1-ε) and the probability Alternatively, we can randomly extract messages from other messages with a probability of (1-ε). Randomly extract one message from the best messages above a certain threshold, and select it with probability ε. The latest message may be extracted.

[0035] In addition, there may be cases where only one message is associated with a BCT. If so, that one message is extracted from the BCT.

[0036] <Nudge Cocktail> As mentioned above, implementing an intervention involves sending messages linked to one or more BCTs. A message set is generated by extracting messages from each group. Such a message set is referred to as a "nudge cocktail."

[0037] In this embodiment, two types of nudge cocktails are generated depending on the stage of intervention. The first nudge cocktail (first set of messages) was designed to target all participants in the early stages of the intervention. This is the nudge cocktail that is generated for the target person. The term MECE (Mutually Exclusive and Collectively Exhaustive) type nudging is used to refer to It is written as Kutell.

[0038] MECE-type nudge cocktails are a combination of multiple BCs that act on behavioral change factors related to the target field. From the T, messages linked to BCTs that only affect specific behavioral change factors were extracted. This consists of the message issued when the BCT is accepted. This is to make it easier to determine whether the transformation factors have worked.

[0039] Such MECE-type nudge cocktails are not BCTs that only affect specific behavioral change factors. All of the behavior change factors related to the target area were selected, and each of the selected BCTs was It is generated by extracting one message from each of the linked messages. BCTs that only affect specific behavior change factors can be applied to all behavior change factors related to the target field. The selection of factors is made to cover all the behavioral change factors to which the subjects may respond. do.

[0040] The second nudge cocktail (second message set) is a behavior change factor tailored to the target audience. It is a nudge cocktail generated based on the MECE-type nudge cocktail. This second nudge cocktail is generated for subjects who responded as follows: This is referred to as an adaptive nudge cocktail.

[0041] Adaptive nudge cocktails are designed to provide multiple lines of advice relevant to a target area based on responses from the target audience. Among the behavioral change factors, identify the behavioral change factors to which the subject responded, and Select behavior change techniques that affect the behavior change factors, and then By extracting one or more messages from each of the linked message groups, is generated.

[0042] In both MECE and adaptive nudge cocktails, the nudges from each message group The nudge cocktail varies depending on the target audience, as the message is randomly selected for each target audience. It becomes what it is.

[0043] (Details of behavioral support methods) Next, a method for generating a nudge cocktail and evaluation of intervention in the behavioral support method according to this embodiment Here's how.

[0044] <Coordinate system> In conventional behavioral support methods, it takes time for a subject to change their behavior through intervention. The process and reasons were judged qualitatively. However, this method does not allow for accurate analysis of each subject. It has been difficult to design individualized interventions for these patients.

[0045] In contrast to this, in this embodiment, a coordinate system is generated with multiple behavioral change factors as axes, In this coordinate system, the target coordinates or target area (hereinafter collectively referred to as the target area) are Then, whether or not the subject has undergone behavioral change is determined by the subject's position in the coordinate system. The evaluation is based on whether the target has approached the target area. This is referred to as "Behavioral Scientific Distance (BSD)."

[0046] In this coordinate system, for each subject, the current coordinate and the target area to which the intervention should be directed are By setting the distance (BSD) from the current coordinate to the target area, the distance is converted into a multidimensional vector. This allows for a more objective and clear understanding of the BCTs that should be adopted to bridge this gap. It becomes possible to visualize it.

[0047] Below, we will explain the steps for intervention based on behavioral distance using the framework shown in Figure 1 as an example. . (1) Selection of coordinate axes First, academic research or opinion polls, or theories on behavioral science, personas of the subjects, Identify multiple factors that can change behavior in the target field based on the setting, behavioral process map, etc. These are defined as coordinate axes. Among the factors for behavioral change shown in Figure 1, "ability" The coordinate axis showing "opportunity" is the C axis, the coordinate axis showing "motivation" is the O axis, and the coordinate axis showing "motivation" is the M axis. Write down.

[0048] Here, the units of the multiple coordinate axes do not need to be the same, and do not need to be integers. For example, when designing an intervention to promote behavioral change toward vaccination, the C axis is a 5-point scale. The O axis represents the number of visits to the vaccination reservation site, and the M axis represents the number of seconds spent on the reservation site. The coordinate axes may be set as follows. The dimensional space becomes a coordinate system that represents the process of behavioral change in the subject.

[0049] (2) Setting the target area (coordinates) Next, a target area is set in a coordinate system consisting of the selected coordinate axes. In the field of intervention, what condition are the subjects in and what kind of intervention have they received? The target area is set as pinpoint coordinates. For example, "any value of the coordinates (c, o, m) is 3 or more" By setting a target area in this way, it is possible to Depending on the behavioral science characteristics and human characteristics of each subject, we will determine the shortest path to behavioral change for each subject. It can be clearly shown.

[0050] The target area is set in common for all subjects. It also changes over time and with changes in the environment. The influence of factors that change behavior such as these can be easily determined by changing the setting of the target area in the coordinate system. The impact on behavioral change factors can be expressed as, for example, the vaccination rate. Changes in standards, improvements to user interface (UI) and user experience (UX) These include the increased convenience of the new system and the approaching deadline for free medical consultations.

[0051] (3) Setting the target's coordinates In the above coordinate system, the initial coordinates of the subject are basically set to (c,o,m) = (0,0,0). And the past Visit For subjects whose history and attributes are available, a certain coordinate transformation rule is used. For example, the initial coordinates may be set individually based on the rule. Visit For subjects with a history Regarding this, Visit Since we understand the method (i.e., we have capacity), the initial coordinates are (c,o,m) =(1,0,0).

[0052] In this embodiment, intervention actions (messages) are performed based on the coordinate system set in this way. Decisions about the appropriate interventions (such as sending a message) and evaluation of intervention actions are made.

[0053] (4) Creation of MECE-type nudge cocktails BCTs that act only on specific behavioral change factors have a value for only one coordinate (non-zero). ), and the other coordinate values ​​are represented by a vector with zero. It is generated based on a collection of BCT vectors such as

[0054] Specifically, the sum of the BCT vectors is <+1,+1,+1>, and the constituent BCT vectors are The aggregate with the smallest number of blocks is extracted. If so, one of the groups is randomly selected for each subject. For example, see Figure 1. In the frame shown in Fig. 1, the action instruction vector is <+1,0,0>, the social support vector is <0,+1,0>, and the social A set of three BCT vectors of the canonical vector <0,0,+1> (a trivalent BCT vector) Furthermore, for each subject, the BCT vectors included in this collection can be extracted. By extracting messages one by one from the message group linked to A MECE-type nudge cocktail (trivalent nudge cocktail) consisting of sage is generated. The sum of the BCT vectors in the set does not have to be <+1,+1,+1>. For example, in Figure 1 In this frame, the action instruction vector <+1,0,0>, the social support vector <0,+1,0>, and the social norm vector The vector is <0,0,+1>, but the action instruction vector is <+1,0,0>, the social support vector is <0,+2,0>, and the social The social norm vector is <0,0,+3>, and the sum of the action instruction vector is <+1,+2,+3>. The social support vector and the social norm vector may be extracted as a set. The set includes a predetermined number of BCT vectors, and the sum of the predetermined number of BCT vectors is a predetermined The n-valent nudge cocktail consists of n (n > 0 integer) types (i.e., The message may be configured to include a message corresponding to a BCT vector of the dynamic change factor.

[0055] (5) Update the target person's coordinates When a nudge cocktail intervention is implemented, the response to the nudge cocktail is determined. The behavioral change factors to which the subject responded are identified, and the subject's coordinates in the coordinate system are updated. The behavioral change factors that the subject responded to can be identified from the BCT that the subject received. For example, if a target person accesses a reservation site that is directed to a message linked to a certain BCT, When the message is accessed, the BCT corresponding to the message is identified and the frame shown in Figure 1 is created. This will allow us to identify the behavioral change factors that this BCT will affect.

[0056] For example, for the MECE-type nudge cocktail mentioned above, if a subject accepts all BCTs, When this happens, the subject's state in the coordinate system is a trivalent coordinate system with respect to the initial coordinate system (0,0,0). It is considered to have changed by the BCT vector (<+1,0,0>, <0,+1,0>, <0,0,+1>). That is, the total change due to the trivalent BCT vector is <+1,0,0>+<0,+1,0>+<0,0,+1>=<+1,+1,+1> Therefore, the coordinates of the subject are calculated by adding this change to the initial coordinates (0,0,0). , changes to (1,1,1).

[0057] In this way, by updating the subject's coordinates in the coordinate system, the line on which the subject responded can be determined. It is possible to objectively grasp the dynamic change factors and the distance to the target area.

[0058] (6) Creating adaptive nudge cocktails The adaptive nudge cocktail is generated based on each subject's current coordinates in the coordinate system. In detail, first, a vector representing the shortest route from the subject's current coordinates to the nearest target area is generated. If there are multiple candidates for the shortest route, In this case, the route may be selected randomly for each subject. Depending on the situation, it may be possible to prioritize and select those who will continue to receive intervention. For example, Subjects who are further away than a certain distance may be excluded from the intervention.

[0059] Next, we decompose the intervention vector and find the set of BCT vectors whose sum is equal to the intervention vector. Generate a field. For example, if the intervening vector is <+2,+2,+2>, [<+1,0,0>×2, <0,+1,0>×2, <0,0,+1>×2], and [<+2,+1,0>, <0,+1,0>, <0,0,+1> x 2] These two types of aggregates are candidates.

[0060] If there are multiple candidates for a set of BCT vectors equivalent to the intervention vector, In addition, there are constraints on intervention actions, and the BCT vector When the number is limited (for example, there is a cost constraint on sending messages, and the number of messages that can be sent is limited) If there is a limit to the number of messages, a group that meets the requirements is randomly selected from the candidates. Extracted from each of the messages associated with the BCT vectors that make up the selected set The set of messages delivered becomes the adaptive nudge cocktail for that individual.

[0061] Once an adaptive nudge cocktail intervention is implemented, as described above, Depending on the response to the message, the behavioral change factors to which the subject responded are identified, and the corresponding The coordinates of the subject are updated. When the subject's coordinates reach the target area, intervention is initiated. On the other hand, if the target coordinates have not yet reached the target area, the movement is stopped. Further adaptive nudge cocktails are generated based on the coordinates of

[0062] (Effect of nudge cocktail) The reasons for creating two types of nudge cocktails, MECE and adaptive, are as follows: In other words, the factors that change the behavior of the subjects are thought to be dynamic and change daily. Therefore, nudges are generated based on static personal information that indicates the state of the target person at a certain point in time. With cocktails, it can be difficult to encourage behavioral change in subjects. MECE nudge cocktails are a set of multiple actions identified based on objective research and theory. The intervention is a combination of BCTs that act on dynamic change factors without omission or overlap. This will enable a multifaceted approach to all related behavioral change factors.

[0063] On the other hand, adaptive nudge cocktails are designed to improve the response of subjects who receive MECE-type nudge cocktails. It consists of BCTs that act on the factors that change behavior. By using this, we can dynamically grasp the factors that change the behavior of the subject based on the subject's response. This will enable us to grasp the individual's needs and take an approach that encourages behavioral change in a concentrated manner.

[0064] By combining such a cocktail of MECE and adaptive nudges, Even if the factors that change the behavior of the subjects are unknown, intervention using these nudge cocktails can be implemented in a timely manner. In the next implementation process, it is possible to understand the factors that cause behavioral change in the target person and effectively promote behavioral change. It becomes Noh.

[0065] Here, the number of messages that make up the nudge cocktail, in other words, the number of messages that make up the aggregate, The number of BCT vectors depends on the target level of behavioral change rate of the subjects, the budget available for intervention, and The intervention design conditions, such as the duration, time (period), and ethical constraints, can be determined. Reasonable constraints include, for example, that too many interventions may lead to complaints from the subjects, or that the subjects may be This is an ethical consideration, such as giving the impression of being overbearing. For example, a trivalent nudge The cocktail is expected to be effective for approximately 60% of the target population, while the pentavalent nudge cocktail If the effect is expected to be seen in approximately 90% of all subjects, based on the above-mentioned intervention design conditions, , it is decided whether to use a trivalent or pentavalent.

[0066] In addition, when generating nudge cocktails from the second time onwards, the nudge cocktails sent to the target person in the past will be generated. The message is extracted from the remaining message set excluding the filtered message.

[0067] (Configuration of behavior support system) Next, the configuration of the action support system according to this embodiment will be described. As an intervention, messages that make up the nudge cocktail are sent via email or short message. The notification will be sent to the target person via SM.

[0068] FIG. 3 is a diagram showing an example of a schematic configuration of the action support system according to this embodiment. As shown, the action support system 1 includes an action support device 10 and a terminal 20. Position 10 is a message that uses nudges based on behavioral change factors to encourage behavioral change in the target person. It is an information processing device that provides information to the terminal 20.

[0069] The terminal 20 is a terminal used by a person who is a target of behavioral change, and is, for example, a smartphone, a mobile phone, The terminal 20 may be any of various information devices such as a telephone, a tablet, a personal computer, etc. The function to receive messages such as short messages and e-mails, for example, LINE (registered) It is assumed that various social networking services (SNS) such as (trademark) can be used. do.

[0070] FIG. 4 is a diagram showing an example of the hardware configuration of devices in the action support system 1. As shown in the figure, each device in the action support system 1 (for example, the action support device 10, the terminal 20) Each of the processors 11 includes a central processing unit (CPU) or the like, which corresponds to a computing device. , a storage device 12, a communication device 13, and an input / output device 14. Each of these components is They are connected via a bus so that they can send and receive data to each other.

[0071] The processor 11 is, for example, a CPU (Central Processing Unit), and the storage device 1 2 is a control unit that controls the execution of the program stored in the The processor 11 receives various input data from the input / output device 14 and / or the communication device 13. The input data is received, and the calculation result is output (for example, displayed) to the input / output device 14, and the data is stored in the memory device. The information is stored in the device 12 or transmitted via the communication device 13 .

[0072] The storage device 12 includes a memory, a hard disk drive (HDD), and a solid state drive (SSD). The memory device 12 of the action support device 10 is configured to store at least one of the following data. The storage device 12 may be called a "storage unit" or the like. That's fine.

[0073] The communication device 13 is a device that communicates via a wired and / or wireless network, e.g. For example, it may include a network card, a communication module, a chip, an antenna, etc. Device 13 may also be referred to as a "transmitter" or a "receiver", etc.

[0074] The input / output device 14 may be, for example, a keyboard, a touch panel, a mouse, and / or a microphone. It includes input devices and output devices such as a display and / or a speaker. The device may be referred to as an "input section" or an "output section", etc.

[0075] The hardware configuration described above is merely an example. Each device in the action support system 1 Some of the hardware shown in FIG. 4 may be omitted, or the hardware not shown in FIG. The hardware shown in FIG. 4 may be one or more chips. It may be configured by:

[0076] FIG. 5 is a diagram showing the functional configuration of the devices in the action support system 1. Note that FIG. 5 is an example. Furthermore, each device in the action support system 1 may of course have functions not shown. do.

[0077] As shown in FIG. 5, the action support device 10 includes a setting unit 101, a message generating unit 102, a recording unit 103, and a message generating unit 104. A memory unit 103, a nudge cocktail generation unit 104, a transmission control unit 105, an acquisition unit 106, and an analysis unit At least a part of the functions realized by the acquisition unit 106 is implemented by the communication device 1. 3. Also, the setting unit 101, the message generating unit 102, the nut The signal cocktail generating unit 104, the transmission control unit 105, and the analysis unit 107 are configured by the processor 11. This can be realized by executing the behavior support program stored in the storage device 12. The behavioral support program can be stored on a storage medium. The storage medium storing the program is a computer-readable non-transitory storage medium (Non-transitory The non-transitory storage medium may be a non-transitory computer readable medium. However, it may be a storage medium such as a USB memory or a CD-ROM. The memory unit 103 can be realized using the storage device 12.

[0078] In this embodiment, the action support device 10 is shown as an integrated device. For example, the setting unit 101 and the message generating unit 102 may be separated into separate units. and configure the separate device to operate over a wired and / or wireless network. It may be connected to the support device 10.

[0079] The setting unit 101 sets a plurality of behavioral change factors and one or more actions of the plurality of behavioral change factors. It sets up a framework that links multiple behavioral change techniques that act on behavioral change factors. At the same time, a coordinate system is generated based on multiple behavioral change factors, and the target area and , and perform various settings such as initial settings of the target person's coordinates.

[0080] The message generation unit 102 generates a plurality of message groups each associated with a plurality of BCTs. is generated and stored in the storage unit 103. Each message group is specifically It contains one or more messages to represent.

[0081] The message generator 102 generates messages included in each message group by wired or wireless communication. It may be collected via a communication network or storage media, or automatically generated by machine learning. Alternatively, an operator can edit the automatically generated text and send it as a message. The message generator 102 may also receive text from the input / output device 14 and store it as a message. You can store the messages you have entered, or you can create messages using pre-prepared templates. You may also generate a page.

[0082] The memory unit 103 stores a frame (see FIG. 1) that associates behavioral change factors with BCTs. The coordinate system generated by 01 and various setting information in that coordinate system, and the information linked to each BCT The received nudge messages (see Figure 2) are stored.

[0083] The nudge cocktail generator 104 generates nudge cocktails stored in the storage unit 103 according to the stage of intervention. By extracting messages from the message group, nudge cocktails (MECE-type nudges) can be created. Generate adaptive nudge cocktails.

[0084] The transmission control unit 105 controls the nudge cocktail generated by the nudge cocktail generation unit 104. The included message is controlled to be output to the target person in one or multiple instalments. Under the control of the control unit 105, the communication device 13 of the action support apparatus 10 communicates with the terminal 20 of the subject. Messages can be sent using email or SM (short message). Here, the message transmission control can be performed using, for example, nudge control. The order in which multiple messages in a message were sent, the date each message was sent, and the time period in which each message was sent , the time interval (transmission frequency) for sending the multiple messages, the stop of sending each message, This may include controlling the number of times a message is sent, etc.

[0085] The acquisition unit 106 acquires response information from the target person in response to the output message. Response information is information that indicates the reaction of the person who received the message. For example, whether or not you have viewed short messages (SM) or emails containing messages, Click-Through Rate (CTR) for the links provided on the website, The acquisition unit 106 receives the information from the terminal 20 of the target person, such as the length of time the target person stayed at the site. Response information may be acquired in real time. may include information identifying the corresponding message or messages (e.g., message IDs); This allows you to determine the message group to which the target person responded. .

[0086] The analysis unit 107 analyzes the response information from each subject and determines the continuation of the intervention according to the analysis result. Decide whether to continue or not, and what type of nudge cocktail to generate (MECE type or adaptive type). Specifically, the analysis unit 107 analyzes the behavioral change factors for which response information has been acquired. Extract the factors that caused the subjects to change their behavior.

[0087] (Behavioral support system operation) 6 is a flowchart showing the operation of the action support system according to this embodiment. The operation of the action support system shown in FIG. 6 is merely an example, and is not limited to the illustrated example. 7 and 8 are schematic diagrams showing specific examples of MECE-type nudge cocktails. FIG. 1 is a schematic diagram showing a specific example of a adaptive nudge cocktail.

[0088] First, in advance, we use various surveys and papers to create a persona and behavioral process map for the target person. Through these measures, multiple factors for behavior change in the area of ​​intervention will be identified. The identification of multiple behavioral change factors is not limited to the above. For example, MECE type with high valence Nudge cocktails may be used to identify more than one behavior change driver, i.e. MECE The first intervention using a nudge cocktail narrows down the factors that change behavior (determining the number of coordinate axes) ), initial coordinates may be set by a second intervention using a MECE-type nudge cocktail. In addition, the target persona and / or behavioral process map can be used in payment apps and social media. May be set based on data collected through network services (SNS) .

[0089] As shown in FIG. 6, the behavior support device 10 calculates a coordinate system In other words, the behavioral change factors and BCT are related to each other. Based on the frame (see Figure 1), we define the BCT vector and perform multiple behavioral changes. A coordinate system is generated using the capacity factors as coordinate axes, and the initial coordinates of the target area and the subject in the coordinate system are calculated. Set.

[0090] Next, the behavior support device 10 generates a plurality of messages corresponding to each BCT (behavior change technique). The generated message is linked to the corresponding BCT (step S102). The data is stored in the storage unit 103 together with the image data.

[0091] Next, the action support device 10 creates a MECE-type nudge cocktail for each subject ( As mentioned above, in the MECE-type nudge cocktail, multiple BCs Since messages are randomly selected from each message group associated with T, Therefore, a different set of messages is created for each target audience.

[0092] Specifically, as shown in Figure 7, for target X, the nudge message group shown in Figure 2 Among them, sentence B was extracted from the group of messages linked to "action instructions" and sentence B was extracted from the group of messages linked to "social support" " and messages linked to "social norms." A MECE-type nudge cocktail is generated consisting of sentences J extracted from the message group. Similarly, for target Y, the text extracted from the message group linked to "operation instructions" B, sentence E extracted from the group of messages linked to "social support," and sentence E extracted from the group of messages linked to "social norms" " and a sentence H extracted from a group of messages linked to " The same is true for subject Z.

[0093] Next, the action support device 10 transmits the generated MECE-type nudge cocktail to each subject. At this time, all messages included in one nudge cocktail are You can send them all at once, or you can send one or more messages sequentially over a period of time. is also good.

[0094] In addition, the message set for one intervention (opportunity to generate a nudge cocktail) was sent. There may be cases where the number of BCTs that make up the nudge cocktail is small compared to the number of times. In such a case, the number of messages extracted from the message group associated with one type of BCT For example, if the number of transmissions set is 3 and the BCT is 2, Two messages were selected from the group of messages associated with BCT, which is thought to have a relatively large impact. Alternatively, messages can be selected from the message group associated with each BCT. The messages were randomly extracted one by one, and then the entire two message groups (the extracted messages) were A message may be randomly selected from the list (excluding messages).

[0095] In addition, instead of sending messages contained in the nudge cocktail via email or SM, Instead, they deliver nudge cocktails in the form of videos or websites that allow for heatmap analysis. You can believe it.

[0096] If there is no opportunity to send another message after that (step S105: No), On the other hand, if a message is to be sent again ( Step S105: Yes), the action support apparatus 10 performs the following steps for each subject: The response to the behavioral change factors of the type nudge cocktail is confirmed (step S106).

[0097] The behavioral support device 10 is for subjects who do not respond to the MECE-type nudge cocktail. If the answer is "No" in step S107, the MECE-type nudge cocktail is created again (step However, in the MECE-type nudge cocktails generated from the second time onwards, From each message group to be extracted, remove messages that have already been sent to the target person. For example, response information is obtained from subject Y. If this is not done, then for the person Y, it will be linked to the "operation instruction" as shown in Figure 8. Among the messages sent (see Figure 2), a random message was selected from the messages other than the sent message B. One message (e.g., sentence C) extracted from the One message was randomly selected from the messages other than the sent message E. Sage (e.g., text F) and messages linked to "social norms" that have already been sent One message (e.g., message J) randomly extracted from messages other than message H A MECE-type nudge cocktail consisting of

[0098] On the other hand, the action support device 10 performs the following for the subject from whom some kind of reaction was obtained (step S10 7: Yes), the coordinates in the coordinate system are updated (step S108).

[0099] Next, the action support device 10 generates an adaptive nudge cocktail for the subject ( Step S109). For example, based on the updated coordinates of the subject X, the intervention vector <+2, +2, + 2> is calculated, and the set of BCT vectors equivalent to this intervention vector [<+2,+1,0>, <0,+1,0> , <0,0,+1>×2] is selected for subject X, as shown in Figure 9. Among the messages linked to " (see Figure 2), messages other than the sent text D A randomly extracted message (e.g., message E) and a message linked to "social norms" were Among the messages sent, two randomly extracted messages were sent, except for the sent message J. Two messages (e.g., text H and I) and a message linked to "step-by-step task setting" An adaptive nudge consisting of a randomly selected message (e.g., message K) from the target page. The valence n of the intervention vector is determined by the number of BCT vectors included in the ensemble. The number of types of behavioral change factors indicated by the BCT vector may be equal to the number of types of behavioral change factors indicated by the BCT vector. As shown, if the collection contains multiple BCT vectors of the same type (e.g., the above <0,0,+1>), In this case, from the viewpoint of diminishing returns and spam detection avoidance, the number of BCT vectors is equal to the number of the BCT vectors. A number of messages may be extracted, or a number of messages less than the number of BCT vectors may be extracted. A message may be extracted.

[0100] Next, the action support device 10 transmits the generated adaptive nudge cocktail to each subject ( Step S110) The number of times and method of transmission are the same as in step S104.

[0101] If there is no opportunity to send another message after that (step S111: No), On the other hand, if a message is to be sent again ( Step S111: Yes), the action support apparatus 10 performs the adaptive navigation process for each subject. The reaction to the behavioral change factors of the cocktail is confirmed (step S112).

[0102] The behavioral support device 10 provides a method for treating subjects who do not respond to the adaptive nudge cocktail. (Step S113: No), the MECE-type nudge coordinates sent to the target person are The reaction of the user to the behavior change factor is confirmed (step S106).

[0103] On the other hand, the behavior support device 10 detects whether or not a certain response was obtained to the adaptive nudge cocktail. For the subject (step S113: Yes), update the coordinates in the coordinate system (step S114). If the updated coordinates reach the target area (step S115: If the updated coordinates are not correct, the action support device 10 stops intervention for the subject. If the target area has not been reached (step S115: No), the action support apparatus 10 An adaptive nudge cocktail is again created for the subject (step S109).

[0104] As described above, according to this embodiment, the subject can be identified without using personal information of the subject. This will make it possible to effectively encourage behavioral change in individuals.

[0105] Furthermore, according to this embodiment, a network without overlaps can be created from the beginning by using the MECE-type nudge cocktail. This allows for comprehensive intervention, resulting in behavioral changes in subjects in a shorter time than conventional methods. This becomes possible.

[0106] Furthermore, according to this embodiment, the influence of the passage of time and changes in the environment on factors for behavioral change occurs. Even if this happens, adaptive nudge cocktails can flexibly incorporate such influences. Cut.

[0107] Furthermore, according to this embodiment, randomness can be introduced into the process of generating nudge cocktails. Compared to traditional intervention designs where intervention designers select nudges based on intuition and experience, This will enable us to achieve the desired intervention effect.

[0108] Furthermore, according to this embodiment, factors defined by the COM-B model and IBM are used. By identifying factors that lead to behavioral change, quality can be ensured from a behavioral science perspective. Furthermore, we randomly extract messages from each message group to generate nudge cocktails. Therefore, behavioral science can help diversify risks caused by uncertainties that cannot be controlled. become.

[0109] The above-described embodiments are intended to facilitate understanding of the present invention and are not intended to limit the present invention. The flowcharts, sequences, and implementations described in the embodiments are not intended to be interpreted as a single entity. The elements of the embodiment, as well as their arrangement, materials, conditions, shapes, sizes, etc., are merely examples. It is not limited to this and can be changed as appropriate. [Explanation of symbols]

[0110] 1...Behavior support system, 10...Behavior support device, 11...Processor, 12...Storage device, 1 3...communication device, 14...input / output device, 20...terminal, 101...setting unit, 102...message generator generation unit, 103...storage unit, 104...nudge cocktail generation unit, 105...transmission control unit, 106... Acquisition section, 107...Analysis section

Claims

1. a coordinate system generation unit that generates a coordinate system based on one or more behavior change factors; a target area setting unit that sets target coordinates or a target area in the coordinate system; a subject coordinate update unit that updates where the subject is located in the coordinate system; an intervention action determination unit that determines an intervention action for the subject, the intervention action determination unit determines an intervention action for the subject based on a positional relationship between the target area and the position of the subject in the coordinate system; Behavioral support system.

2. The target area setting unit appropriately changes the set target area based on the passage of time or changes in the environment. The behavior support system according to claim 1 .

3. the intervention action for the subject is outputting a message to the subject, calculating an intervention vector representing the shortest route from the subject's current coordinates to the nearest target area; generating a collection of BCT vectors whose sum equals said intervening vector; Selecting a message to be output to the target person from a group of messages linked to the BCT vectors that make up the collection; The behavior support system according to claim 1 .

4. Calculating an intervention vector for each of the plurality of subjects, and selecting subjects for whom the intervention action will be continued from the plurality of subjects according to each intervention vector; The behavior support system according to claim 2 .

Images