WO2021046359A1

WO2021046359A1 - A game-theoretic approach to detecting changes in theory of mind

Info

Publication number: WO2021046359A1
Application number: PCT/US2020/049420
Authority: WO
Inventors: Michael Lindemann; David NOBBS; Cedric SIMILLION; Timothy KILCHENMANN; Florian LIPSMEIER; Christian Gossens; Joerg Hipp; Christopher CHATHAM
Original assignee: Hoffmann-La Roche Inc.; F. Hoffmann-La Roche Ag
Priority date: 2019-09-05
Filing date: 2020-09-04
Publication date: 2021-03-11

Abstract

Theory of Mind generally refers to the ability to interpret others' beliefs, intentions, and emotions. In order to participate in adaptive social interactions, individuals must be capable of inferring others' goals based on observed behavior. Theory of Mind, however, is often impaired in conditions like Autism Spectrum Disorder (ASD). Multiplayer games as described herein allow one to assess Theory of Mind with a model that accounts for cooperative and non-cooperative behaviors. The administration of the autism-relevant neuropeptide vasopressin may increase players' tendency to cooperate, and games that take into account both cooperative and non-cooperative behavior may then be used as a tool for measuring treatment effects on Theory of Mind in ASD.

Description

A GAME-THEORETIC APPROACH TO DETECTING CHANGES IN THEORY OF MIND

CROSS-REFERENCE TO RELATED APPLICATIONS

[0001] The instant application claims priority to U.S. Provisional Patent Application No. 62/896,058, titled “A Game-Theoretic Approach to Detecting Changes in Theory of Mind” and filed September 5, 2019, the disclosure of which is hereby incorporated by reference in its enthety.

TECHNICAL FIELD

[0002] Aspects described herein generally relate to digital health tools for medical diagnostics and analytics and more specifically to measuring treatment effects on Theory of Mind in Autism Spectrum Disorder (ASD) patients.

BACKGROUND

[0003] Autism spectrum disorders (ASD) include a variety of neurodevelopmental conditions, including autism and Asperger syndrome. Individuals with ASD may experience difficulties with social communication and interaction and also exhibit restricted and/or repetitive patterns of behavior. Those in the mild range of ASD may' function independently, while those in the moderate to severe range may require substantial support in their daily lives. Theory of Mind generally refers to the ability to interpret others' beliefs, intentions, and emotions.

SUMMARY

[0004] The following presents a simplified summary of various aspects described herein. This summary is not an extensive overview, and is not intended to identify key or critical elements or to delineate the scope of the claims. The following summary merely presents some concepts in a simplified form as an introductory prelude to the more detailed description provided below. Corresponding apparatus, systems, and computer-readable media are also within the scope of the disclosure.

[0005] As a general introduction to the subject matter described herein, in order to participate in adaptive social interactions, individuals must be capable of inferring others’ goals based on observed behavior. Theory of Mind, however, is often impaired in conditions like Autism Spectrum Disorder (ASD). Interactive media, e.g., multiplayer games, may allow one to assess Theory of Mind with a model that accounts for cooperative and non-cooperative behaviors. The administration of the autism-relevant neuropeptide vasopressin (and/or other treatments) may increase players' tendency to cooperate, and games that take into account both cooperative and non-cooperative behavior may then be used as a tool for measuring treatment effects on Theory of Mind in ASD.

[0006] These features, along with many others, are discussed in greater detail below.

BRIEF DESCRIPTION OF THE DRAWINGS

[0007] A more complete unders tanding of aspects described herein and the advantages thereof may be acquired by referring to the following description in consideration of the accompanying drawings, in which like reference numbers indicate like features, and wherein:

[0008] Figure 1 illustrates a custom computing system and architecture which may be used to implement one or more illustrative aspects described herein;

[0009] Figures 2A-V illustrates screenshots of a game playing board according to one or more illustrative aspects described herein;

[0010] Figure 3 illustrates simulation results for a ToMPlan agent according to one or more illustrative aspects described herein;

[0011] Figure 4 illustrates simulation results for a k-ToM agent according to one or more illustrative aspects described herein;

[0012] Figure 5A is a table conceptually illustrating a k-ToM’s agent interpretation of player goals according to one or more illustrative aspects described herein:

[0013] Figure 5B is a table conceptually illustrating user engagement with the custom computing system according to one or more illustrative aspects described herein; and

[0014] Figure 6 is a flow'chart of a process for measuring treatment effects according to one or more illustrative aspects described herein.

DETAILED DESCRIPTION [0015] In the following description of the various embodiments, reference is made to the accompanying drawings, which form apart hereof, and in which is shown by way of illustration various embodiments in which aspects described herein may be practiced. It is to be understood that a variety of embodiments may be utilized and structural and functional modifications may^¬ be made without departing from the scope of the described aspects and embodiments. Aspects described herein are capable of other embodiments and of being practiced or being carried out in various ways. Also, it is to be understood that the phraseology and terminology used herein are for the purpose of description and should not be regarded as limiting. Rather, the phrases and terms used herein are to be given their broadest interpretation and meaning. The use of “including” and “comprising” and variations thereof is meant to encompass the items listed thereafter and equivalents thereof as well as additional items and equivalents thereof. The use of the terms “mounted,” “connected,” “coupled,” “positioned,” “engaged” and similar terms, is meant to include both direct and indirect mounting, connecting, coupling, positioning, and engaging.

[0016] As a general introduction to the subject matter described herein, Theory- of Mind generally refers to the ability to interpret others' beliefs, intentions, and emotions. In order to participate in adaptive social interactions, individuals must be capable of inferring others’ goals based on observed behavior. Theory of Mind, however, is often impaired in conditions like Autism Spectrum Disorder (ASD). Interactive media, e.g , multiplayer games, may allow one to assess Theory- of Mind with a model that accounts for cooperative and non-cooperative behaviors. The administration of the autism-relevant neuropeptide vasopressin may increase players’ tendency to cooperate, and games that take into account both cooperative and non- cooperative behavior may then be used as a tool for measuring treatment effects on Theory- of Mind in ASD. Aspects described herein allow for identification of an optimal learning agent to estimate Theory of Mind changes over time in a mobile version of a coordination game. Aspects described herein also ensure acceptability of task in clinical trials by individuals with ASD. A first aspect described herein provides a learning agent for a tool for measuring treatment effects on Theory of Mind in ASD, which may be integrated into a device to create a custom and/or special purpose computing device, such as a modified smartphone. According to an aspect, multiplayer games may be played by a single player by using a computer simulation (or AI) for the second player. The games may allow the measurement of Theory- of Mind in an environment (e.g., home, private space, “safe” space, medical facility) in which the user feels comfortable. Based on the measured Theory- of Mind, particular treatments may be recommended and/or provided. For example, a therapeutically effective dose of vasopressin may be determined based on a difference between the measured Theory of Mind and a threshold Theory of Mind score. The therapeutically effective dose of vasopressin may he administered to the patient. The patient may then replay the multiplayer games to determine the effectiveness of the administered dose.

[0017] Figure 1 illustrates one example of a custom network architecture and data processing devices that may be used to implement one or more illustrative aspects described herein. Various network nodes 103, 105, 107, and 109 may be interconnected via a wide area network (WAN) 101, such as the Internet. Other networks may also or alternatively be used, including private intranets, corporate networks, LANs, wireless networks, personal networks (PAN), and the like. Network 101 is for illustration purposes and may be replaced with fewer or additional computer networks. A local area network (LAN) may have one or more of any known LAN topology and may use one or more of a variety of different protocols, such as Ethernet. Devices 103, 105, 107, 109 and other devices (not shown) may be connected to one or more of the networks via twisted pair wires, coaxial cable, fiber optics, radio waves or other communication media.

[0018] The term “network” as used herein and depicted in the drawings refers not only to systems in which remote storage devices are coupled together via one or more communication paths, but also to stand-alone devices that may be coupled, from time to time, to such systems that have storage capability. Consequently, the term “network” includes not only a “physical network” but also a “content network,” which is comprised of the data — attributable to a single entity — which resides across all physical networks.

[0019] The components may include data server 103, web server 105, and client devkes 107, 109. Data server 103 provides overall access, control and administration of databases and control software for performing one or more illustrative aspects described herein. Data server 103 may be connected to web server 105 through which users interact with and obtain data as requested. Alternatively, data server 103 may act as a web server itself and be directly- connected to the Internet (in which case device 105 is not needed). Data server 103 may be connected to web server 105 through the network 101 (e.g , the Internet), via direct or indirect connection, or via some other network. Users may interact with the data server 103 using remote computers 107, 109, e.g., using an application, mobile app, or web browser to connect to the data server 103 via one or more externally exposed web sites and/or web services hosted by web server 105. Client computers 107, 109 may be used in concert with data server 103 to access data stored therein, or may be used for other purposes. For example, from client device 107 a user may access web server 105 using an Internet browser, as is known in the art, or by executing a software application that communicates with web server 105 and/or data server 103 over a computer network (such as the Internet).

[0020] Servers and applications may be combined on the same physical machines, and retain separate virtual or logical addresses, or may reside on separate physical machines. Figure 1 illustrates just one example of a network architecture that may be used, and those of skill in the art will appreciate that the specific network architecture and data processing devices used may vary, and are secondary to the functionality that they provide, as further described herein. For example, services provided by web server 105 and data server 103 may be combined on a single server.

[0021] Each component 103, 105, 107, 109 may be any type of known computer, server, or data processing device accessing a web-based implementation, or a custom device integrated with special purpose software, as further described herein. Data server 103, e.g., may include a processor i l l controlling overall operation of the data server 103. Data server 103 may- further include RAM 113, ROM 115, network interface 117, input/output interfaces 119 (e.g., keyboard, mouse, display, printer, etc.), and memory 121. I/O 119 may include a variety of interface units and drives for reading, writing, displaying, and/or printing data or files. Memory 121 may further store operating system software 123 for controlling overall operation of the data processing device 103, control logic 125 for instructing data server 103 to perform aspects described herein, and other application software 127 providing secondary', support, and/or other functionality which may or may not be used in conjunction with other aspects described herein. The control logic may also be referred to herein as the data server software 125. Functionality_' of the data server software may refer to operations or decisions made automatically based on rules coded into the control logic, made manually by a user providing input into the system, and/or a combination of automatic processing based on user input (e.g., queries, data updates, etc.).

[0022] Memory_' 121 may also store data used in performance of one or more aspects described herein, including a first database 129 and a second database 131 . In many embodiments, the first database may include the second database (e.g., as a separate table, report, etc.). That is, the information may be stored in a single database, or separated into different logical, virtual, or physical databases, depending on system design. Devices 105, 107, 109 may have similar or different architecture as described with respect to device 103. Those of skill in the art will appreciate that the functionality of data processing device 103 (or device 105, 107, 109) as described herein may be spread across multiple data processing devices, for example, to distribute processing load across multiple computers, to segregate transactions based on geographic location, user access level, quality of service (QoS), etc.

[0023] One or more aspects discussed herein may be embodied in computer-usable or readable data and/or computer-executable instructions, such as in one or more program modules, executed by one or more computers or other devices as described herein. Generally, program modules include routines, programs, objects, components, data structures, and the like that perform particular tasks or implement particular abstract data types when executed by a processor in a computer or other device. The modules may be writen in a source code programming language that is subsequently compiled for execution, or may be written in a scripting language such as (but not limited to) HTML or XML. The computer executable instructions may be stored on a computer readable medium such as a hard disk, optical disk, removable storage media, solid-state memory , RAM, and the like. As will be appreciated by one of skill in the art, the functionality of the program modules may be combined or distributed as desired. In addition, the functionality may be embothed, in whole or in part, in firmware or hardware equivalents such as integrated circuits, field programmable gate arrays (FPGA), and the like. Particular data structures may be used to more effectively implement one or more aspects discussed herein, and such data structures are contemplated within the scope of computer executable instructions and computer-usable data described herein. Various aspects discussed herein may be embodied as a method, a computing device, a system, and/or a computer program product.

[0024] A variety of multiplayer games may be used to test Theory of Mind for a patient. For example, one multiplayer game that may be used is Treasure Hunt. In Treasure Hunt, two (or more) individuals either jointly hunt a treasure chest or pursue a coin independently. If one individual hunts the treasure chest, they must have the cooperation of the other player to succeed. A player may get a coin alone, but a coin is worth less than the treasure chest. To succeed in catching the treasure chest, one hunter must infer the strategy of the other based on prior behavior. Success in the game is therefore dependent on one’s ability to infer the goals of others. In a variety of embodiments, Treasure Hunt may be provided via a mobile app on a smartphone (e.g., device 107) or similar device. In several embodiments, the position of both the large reward (the treasure) and the small rewards (the coins) remain fixed on a 5x5 board. In a number of embodiments, the locations of each may change from one iteration of the game to the next.

[0025] Figures 2.A-V shows illustrative screenshots of the Treasure Hunt multiplayer game that may be played in accordance with a variety of aspects of the disclosure. A multiplayer game may include a variety of welcome screens to engage the user with the game, instruction screens to teach the user how to play the game, game screens that allow the user to play the game, and/or game summary_' screens that explain the results of the game to the user. However, it should be noted that few'er or more screens may be included in a multiplayer game as appropriate. A user avatar may represent the user within the multiplayer game while one or more AI avatars may represent AI agents playing the multiplayer game. The multiplayer game may also include one or more goal object to be interacted with (e.g. captured) by the user avatar and/or the AI avatar(s).

[0026] Figure 2A shows a screenshot of a welcome screen in accordance with one or more aspects of the disclosure. The welcome screen provides a description of the multiplayer game to be played along with buttons to obtain instructions on how to play the game and a buton to start playing the game. Figure 2B shows a screenshot of a first instruction screen in accordance with one or more aspects of the disclosure. The first instruction screen shows a 5x5 game board with the user’s blue boat highlighted. The first instruction screen explains that the task is to look for coins and treasure and that the user is the blue sheep. The first instruction screen includes a button to proceed to the second instruction screen. Figure 2C shows a screenshot of a second instruction screen in accordance with one or more aspects of the disclosure. The second instruction screen shows the game board with one of the coins highlighted. The second instruction screen explains that the user should move to capture the highlighted coin and that the user may manipulate the position of the user’s boat using on screen controls. Figure 2D show's a screenshot of a first game screen in accordance with one or more aspects of the disclosure. The position of the user’s blue boat may be manipulated using the up, down, left, right controls below the game board. The first game screen explains that the controls move the position of the user’s boat within the game board according to the label of the control (e.g. up, down, left, right, etc.). Figure 2E shows a screenshot of a second game screen in accordance with one or more aspects of the disclosure. The second game screen adds a capture control that allows the user to capture a coin when the user boat is located in the same square as a coin. Figure 2F shows a screenshot of a first summary screen in accordance with one or more aspects of the disclosure. The first summary screen indicates that the user successfully captured a coin. The first summary screen also provides an indication of the relative value of a coin (e.g one point) and a button to move to the next screen.

[0027] Figure 2G shows a screenshot of a third instruction screen in accordance with one or more aspects of the disclosure. The third instruction screen shows the game board along with a highlight of the AI agent’s red boat. The third instruction screen explains that the AI agent is also looking for coins and treasure and an indication that, if the AI agent captures a coin, the round ends. The third instruction screen also provides a button to move the user to the next screen. Figure 2H show's a screenshot of a third game screen in accordance with one or more aspects of the disclosure. The third game screen includes the game board with the user boat two squares away from a coin and the AI boat one square away from a coin. The third game screen also includes controls for the user to move the position of the user’s boat within the game board during the user’s turn. Figure 21 shows a screenshot of a fourth game screen in accordance with one or more aspects of the disclosure. In the fourth game screen, the user boat has moved one square up in the game board and the AI boat has moved one square down and captured a coin. Figure 2J show's a screenshot of a second summary screen in accordance with one or more aspects of the disclosure. The second summary screen provides an indication that the round has ended because the red boat captured a coin. The second summary screen also provides a summary of the total number of points earned by the user in the game session, the total number of points earned in the current round, and a button to move to the next screen.

[0028] Figure 2K shows a screenshot of a fourth instruction screen in accordance with one or more aspects of the disclosure. The fourth instruction screen shows the game board with a treasure chest highlighted. The fourth instruction screen also includes an indication that the user may capture the treasure chest and that the treasure chest awards more points than a coin for capture. A buton on the fourth instruction screen takes the user to the next screen. Figure 21, show_'s a screenshot of a fifth game screen in accordance with one or more aspects of the disclosure. The fifth game screen includes the game board, the controls, and an indication that the user should move the user’s boat to the square containing the treasure chest. Figure 2M shows a screenshot of a sixth game screen in accordance w'ith one or more aspects of the disclosure. The sixth game screen includes the game board, the controls, and an indication that the user should continue to move the user’s boat to the square containing the treasure chest. Figure 2N shows a screenshot of a fifth instruction screen in accordance with one or more aspects of the disclosure. The fifth instruction screen indicates that the user’s boat and the AI boat must both be on the square containing the treasure chest in order to capture the treasure chest. Figure 20 shows a screenshot of a seventh game screen in accordance with one or more aspects of the disclosure. The seventh game screen includes the game board, the controls, and the riser’s boat on the same square as the treasure chest. The seventh game screen also includes an indication that the user should press the skip buton to wait on the current square and allow' the AI boat to move. Figure 2P shows a screenshot of an eighth game screen in accordance with one or more aspects of the disclosure. The eighth game screen includes both the user boat and the AI boat located on the same square as the treasure chest. In the controls shown on the eighth game screen, the skip button has now' transformed into a capture button. The eighth game screen also provides an indication that the user may use the capture button to capture the treasure chest when both the user boat and the AI boat are located on the same square as the treasure chest. Figure 2Q shows a screenshot of a third summary screen in accordance with one or more aspects of the disclosure. The third summary screen indicates that the user captured the treasure chest and the point value of the treasure chest. In the third summary screen, the value of the treasure chest is four points, which is relatively higher than a coin which has a value of one point. In particular, the third summary screen provides guidance to the user dial, by cooperating with the AI boat, the user may earn many more points than acting alone. The third summary screen also provides a button to move to the next screen.

[0029] Figure 2R show's a screenshot of a ninth game screen in accordance with one or more aspects of the disclosure. The ninth game screen includes the user boat adjacent to a treasure chest and the AI boat adjacent to a coin. The ninth game screen also includes an indication that the AI boat may not alw'ays play cooperatively with the user. Figure 2S show's a screenshot of a fourth summary' screen in accordance with one or more aspects of the disclosure. The fourth summary' screen may be displayed after the user boat moved onto the same square as the treasure chest and the AI boat moved onto the same square as the coin. The fourth summary screen indicates that, although the treasure chest is worth more points than the com, that capturing the treasure chest is based on the user understanding the strategy of the AI boat. The fourth summary' screen also includes an indication of the number of points acquired by the user in this round and a button to move to the next screen. [0030] Figure 2T shows a screenshot of a fifth summary screen in accordance with one or more aspects of the disclosure. The fifth summary screen may be displayed after a user completes a round of the multiplayer game. The fifth summary screen includes an indication of the total number of points acquired by the user over the play session, the number of points acquired by the user in the current round, an indication of the current round completed, and an indication of the total number of rounds to be played in the game. The fifth summary screen also includes a button to start the next round of the multiplayer game.

[0031] Figure 2U shows a screenshot of a timeout screen in accordance with one or more aspects of the disclosure The timeout screen may be displayed if the user has not interacted with the multiplayer game for a threshold period of time, such as two minutes. However, any time period may be used as appropriate. The timeout screen may allow the user to end the current game and/or continue playing the game. Figure 2 V shows a screenshot of an activation screen in accordance with one or more aspects of the disclosure. The activation screen may allow the multiplayer game to be activated and/or deactivated based on the Theory of Mind of the user. In many embodiments, the multiplayer game is integrated into a software application that provides a variety of other tasks that the user may complete. The activation screen may- cause the multiplayer game to be included in the set of tasks to be completed by the user and/or removed from the set of tasks. In tins way, the software application may provide a suite of tasks to the user in accordance with the Theory of Mind score and/or other capabilities of the user.

[0032] According to a variety of aspects, an AI agent may be used as the second player so that the user under test may play the game by him/herself. To effectively simulate a human player, the AI agent learns the other player’s intentions over iterations of the game and respond to changes in behavior. Preferably the number of training iterations would be minimized for a realistic simulation of the response to cooperative or non-cooperative behavior. The AI agent may be implemented using one or more of a variety of machine classifiers. It should be readily apparent to one having ordinary skill in the art that a variety of machine classifiers may be utilized including (but not limited to) decision trees, k-nearest neighbors, support vector machines (SVM), neural networks (NN), recurrent neural networks (RNN), convolutional neural networks (CNN), and/or probabilistic neural networks (PNN). RNNs may further include (but are not limited to) fully recurrent networks, Hopfield networks, Boltzmann machines, self-organizing maps, learning vector quantization, simple recurrent networks, echo state networks, long short-term memory networks, bi-directional RNNs, hierarchical RNNs, stochastic neural networks, and/or genetic scale RNNs.

[0033] A variety of learning agents may be used to estimate the strategy of an opponent in playing multiplayer games. The learning agent(s) used for a particular multiplayer game may be selected based on how fast they leam the human player’s behavior and/or how fast they adapt to changes in the human player’s behavior. For example, different recursive learning agents may be used to estimate the strategy of an opponent and changes in that strategy over time. A Theory of Mind Plan (ToMPlan) agent may simulate changes in the probability that the opponent targets the treasure. A sophistication level Theory of Mind (k-ToM) agent may simulate changes in sophistication level (k) of the opponent.

[0034] Figure 3 illustrates simulation results for a ToMPlan agent. In each simulation, a number of games was simulated against an opponent that had a 40% chance of seeking the treasure (before, 310} followed by the same number of games against an opponent that was seeking the treasure 60% of the time (after, 312). The results show that the agent performed about 200-250 games to converge on the opponent’s behavior as shown in panel 310. The agent performed around 50 games to adapt to changes in the behavior as shown in panel 312.

[0035] Figure 4 show's simulation results for a k-ToM agent. In each simulation, 50 games were played against an opponent with the sophistication level fixed at k=1, followed by another 50 games against a different sophistication level of k=2. Four different values of the forget parameter (0.9, 0.95, 0.99 and 1.0) were evaluated. Each simulation was repeated 10 times. The vertical axis shows the current sophistication level k of the opponent estimated by the agent. The size of each dot reflects how many times out of 10 a given value of k was estimated. The results show that the k-ToM agent may readily detect the sophistication level of the opponent in less than 20 games at any value for the forget parameter. Changes in sophistication were most quickly detected with forget values of 0.95 and 0.99, requiring about 10 games. The k-ToM agent thus detected the sophistication level of the opponent in less than 20 games, as shown in panel 410, and took about 10 games to adapt to a change in sophistication level as shown in panel 412. The k-ToM agent decides on its next move using a value function. This value function is optimized using a sophistication level k When k=l, the k-ToM agent ignores the goals of the other player and so seeks the small reward (i.e., a coin). When k=2, the k-ToM agent assumes the other player is ignoring the k-ToM agent’s goals, and so on. A table conceptually illustrating the k-ToM’s agent interpretation of player goals for k=l-5 is shown in Figure 5 A. The k-Tom agent may estimate the sophistication level of the human player from their moves.

[0036] Computing devices in accordance with embodiments of the invention may use multiplayer games, such as Treasure Hunt, with a learning Al agent to estimate cooperative behavior and change of cooperative behavior of a participant. For example, this indicates the task provided by Treasure Hunt is be suitable for assessing Theory of Mind in individuals with ASD. The k-ToM agent may detect changes in behavior as it requires fewer games to be played to adapt to the participant’s strategy. Focus sessions have indicated that multiplayer games are likely to be well-received as part of the assessments in a clinical trial, though may be unsuitable for some low-functioning individuals. A summary of user engagement in a clinical trial is shown in Figure 5B.

[0037] A variety of multiplayer games may be used to test a patient’s Theory of Mind as described herein. Based on the patient’s performance in playing the game, particular treatment regimens may be prescribed in order to maintain and/or improve the patient’s Theory of Mind. Figure 6 is a flowchart of a process for measuring treatment effects according to one or more illustrative aspects described herein. Some or all of the steps of process 600 may be performed using one or more computing devices as described herein. In a variety of embodiments, some or all of the steps described below may be combined and/or divided into sub-steps as appropriate.

[0038] At step 610, a user move may be obtained. The user move may be for a multiplayer game being played for the user. For example, in the Treasure hunt game, the user move may include moving the user’s boat from a first square to an adjacent square. The user move may move the user closer to a coin and/or a treasure chest. In several embodiments, the user move may be determined based on the Theory of Mind score for the user.

[0039] At step 612, an agent strategy may be calculated. The calculated agent strategy may be to cooperate with the user and/or to oppose the user. The agent strategy may be computed based on the obtained user move, the layout of a game board, and/or the sophistication of the user as described herein. For example, in the Treasure Hunt game, if the agent’s token is next to a coin, the agent strategy may be calculated to be opposed to the user such that the agent may end the game in the next turn. In another example, if the user is close to the treasure chest, the agent strategy may be to cooperate with the user to capture the treasure chest. In several embodiments, the agent strategy may be calculated based on a Theory of Mind measurement for the user. In a variety of embodiments, the agent strategy may be determined based on previous games played with the user. For example, if the agent was opposed to the user in the previous game, the agent may cooperate with the user in the current game

[0040] At step 614, an agent move may be calculated. The agent move may he calculated based on the determined strategy. If the agent strategy is to cooperate for the user, the calculated move may include a move that moves an agent piece into a position that helps the user. For example, in the Treasure Hunt game, the agent may move their boat closer to a treasure chest square so that both the agent and the user may capture the treasure chest. If the determined strategy is to oppose the user, the calculated move may include a move that moves the agent piece into a position that helps the agent conclude the game. For example, in the Treasure Hunt game, the agent may move their boat closer to a coin such that the agent may capture the coin and end the current round.

[0041] At step 616, it may be determined if the game is over. The game may be determined to be over when one or more exit conditions of the game have been satisfied. The ending of the game may be determined based on the user move and/or the agent move. For example, in the Treasure Hunt game, the game may be determined to be over when the user captures all of the coins or the agent captures one coin on the game board. However, any exit condition may be used deepening on the specific multiplayer game being played. If the game is not over, the process may return to step 610 to obtain another user move. If the game is over, the process may move on to step 618.

[0042] At step 618, the game summary may be displayed. The game summary may include a summary' of goals achieved by the user and goals achieved by the agent. It should be noted that any results, such as a Theory of Mind score, difficulty ratings, sophistication of the user, agent strategy, game time, number of games played, and any other relevant data may be included in the game summary as appropriate.

[0043] At step 620, treatment effects may be measured. The treatment effects may be measured based on the game results and/or a previously calculated Theory of Mind score for the user. For example, the game results may include a Theory of Mind score for the user for the currently played game. This current Theory of Mind score may be compared to one or more historical Theory_' of Mind scores for the user determined in previous play sessions. The measured treatment effects may include the delta between the current Theory of Mind score and one or more of the historical Theory of Mind scores, a rate of change overtime, a moving average of the Theory of Mind scores, and/or any other statistical measure as appropriate.

[0044] At step 622, treatments may be provided. The provided treatment may be determined based on the measured treatment effects. The provided treatment may include a therapeutically effective dose of a drug to improve the Theory of Mind score for the user. For example, a therapeutically effective dose of vasopressin may be determined and/or administered to the user. The therapeutically effective dose of vasopressin may be determined based on the measured treatment effects. For example, if the moving average of the Theory of Mind scores indicates that the user is progressively performing worse on the multiplayer game, the provided treatment may increase the dosage of vasopressin to the user. In another example, if the current Theory of Mind score is over a threshold value and an improvement over a historical Theory of Mind score, the dosage of vasopressin may be reduced in order to maintain the user's current Theory of Mind score without over-medicating the user.

[0045] Although the present invention has been described in certain specific aspects, many- additional modifications and variations would be apparent to those skilled in the art. In particular, any of the various processes described above may be performed in alternative sequences and/or in parallel (on different computing devices) in order to achieve similar results in a manner that is more appropriate to the requirements of a specific application. It is therefore to be understood that the present invention may be practiced otherwise than specifically described without departing from the scope and spirit of the present invention. Thus, embodiments of the present invention should be considered in all respects as illustrative and not restrictive. Accordingly, the scope of the invention should be determined not by the embodiments illustrated, but by the appended claims and their equivalents.

Claims

WHAT IS CLAIMED IS:

1. A computer-implemented method for playing a multiplayer game, comprising: displaying a game board having a user avatar, an agent avatar, and at least one goal object, the game board being generated for a current round of the multiplayer game; obtaining a user move indicating a direction to move the user avatar within the game board; calculating an agent strategy based on the current round of the multiplayer game, wherein the agent strategy is based on a Theory of Mind score for the user; calculating an agent move based on the agent strategy and the user move; determining the current round of the multiplayer game is over based, the user move, the agent move, and the at least one goal object; and displaying a game summary for the current round of the multiplayer game.

2. The computer-implemented method of claim 1, wherein the agent strategy is based on a plan agent that calculates the agent move based on a probability that the user move targets the goal object

3. The computer-implemented method of claim 1, wherein the agent strategy is based on a sophistication level agent that calculates the agent move based on a sophistication level of the user.

4. The computer-implemented method of claim 1, wherein determining the current round of the multiplayer game is over based on the user move moving the user avatar into a location on the game board shared with at least one of the at least one goal object

5. The computer-implemented method of claim 1, wherein determining the current round of the multiplayer game is over based on the agent move moving the agent avatar into a location on the game board shared with at least one of the at least one goal object.

6. The computer-implemented method of claim 1, further comprising calculating a Theory of Mind score for the user based on the multiplayer game, the agent strategy, and the game summary .

7. The computer-implemented method of claim 6, further comprising: measuring treatment effects based on the Theory of Mind score and a historical Theory of Mind score for the user; determining a therapeutically effective dose of vasopressin based on the measured treatment effects; and administering the therapeutically effecti ve dose of vasopressin.

8. A computing device, comprising: a processor; and a memory in communication with the processor and storing instructions that when read by the processor, cause the computing device to: display a game board having a user avatar, an agent avatar, and at least one goal object, the game board being generated for a current round of a multiplayer game; obtain a user move indicating a direction to move the user avatar within the game board; calculate an agent strategy based on the current round of the multiplayer game, wherein the agent strategy is based on a Theory of Mind score for the user; calculate an agent move based on the agent strategy and the user move; determine the current round of the multiplayer game is over based, the user move, the agent move, and the at least one goal object; and display a game summary for the current round of the multiplayer game.

9. The computing device of claim 8, wherein the agent strategy is based on a plan agent that calculates the agent move based on a probability drat the user move targets the goal object.

10. The computing device of claim 8, wherein the agent strategy is based on a sophistication level agent that calculates the agent move based on a sophistication level of the user.

11. The computing device of claim 8, wherein the instructions, when read by the processor, further cause the computing device to determine the current round of the multiplayer game is over based on the user move moving the user avatar into a location on the game board shared with at least one of the at least one goal object.

12. The computing device of claim 8, wherein the instructions, when read by the processor, further cause the computing device to determine the current round of the multiplayer game is over based on the agent move moving the agent avatar into a location on the game board shared with at least one of the at least one goal object

13. The computing device of claim 8, wherein the instructions, when read by the processor, further cause the computing device to calculate a Theory of Mind score for the user based on the multiplayer game, the agent strategy, and the game summary.

14. The computing device of claim 13, wherein the instructions, when read by the processor, further cause the computing device to: measure treatment effects based on the Theory of Mind score and a historical Theory of Mind score for the user; determine a therapeutically effective dose of vasopressin based on the measured treatment effects; and administer the therapeutically effective dose of vasopressin.

15. A non-transitory machine-readable medium storing instructions that, when executed by one or more processors, cause the one or more processors to perform steps comprising: displaying a game board having a user avatar, an agent avatar, and at least one goal object, the game board being generated for a current round of a multiplayer game; obtaining a user move indicating a direction to move the user avatar within the game board; calculating an agent strategy based on the current round of the multiplayer game, wherein the agent strategy is based on a Theory of Mind score for the user; calculating an agent move based on the agent strategy and the user move; determining the current round of the multiplayer game is over based, the user move, the agent move, and the at least one goal object; and displaying a game summary for the current round of the multiplayer game.

16. The non-transitory machine-readable medium of claim 15, wherein the agent strategy is based on a plan agent that calculates the agent move based on a probability that the user move targets the goal object.

17. The non-transitory machine-readable medium of claim 15, wherein the agent strategy is based on sophistication level agent that calculates the agent move based on a sophistication level of the user.

18. The non-transitory machine-readable medium of claim 15, wherein the instructions, when executed by one or more processors, further cause the one or more processors to determine the current round of the multiplayer game is over based on the user move moving the user avatar into a location on the game board shared with at least one of the at least one goal object.

19 The non -transitory machine-readable medium of claim 15, wherein the instructions, when executed by one or more processors, further cause the one or more processors to determine the current round of the multiplayer game is over based on the agent move moving the agent avatar into a location on the game board shared with at least one of the at least one goal object

20 The non-transitory machine-readable medium of claim 15, wherein the instructions, when executed by one or more processors, further cause the one or more processors to perform steps comprising: calculating a Theory of Mind score for the user based on the multiplayer game, the agent strategy, and the game summary_'. measuring treatment effects based on the Theory of Mind score and a historical Theory of Mind score for the user; determining a therapeutically effective dose of vasopressin based on the measured treatment effects; and administering the therapeutically effective dose of vasopressin.