Program, information processing method, and information processing apparatuses
The program and device use brain activity measurement and neurofeedback to train users to make rational judgments on expected values, addressing the challenge of personal biases in decision-making systems.
Patent Information
- Application Number
- JP2023078378
- Authority / Receiving Office
- JP · JP
- Patent Type
- Applications
- Current Assignee / Owner
- Filing Date
- 2023-05-11
- Publication Date
- 2024-11-21
- Estimated Expiration
- 2043-05-11
AI Technical Summary
Existing decision-making support systems struggle to appropriately determine expected values involving probability theory, as humans tend to avoid risks and losses based on personal biases, making rational judgments difficult.
A program and information processing device that utilize an electroencephalogram measurement device to measure brain activity during decision-making tasks, generate a learning model to estimate brain activity, and perform neurofeedback processing to train users to make rational judgments on expected values.
Enables users to understand and train their tendencies toward risk avoidance and loss aversion, improving rational judgment on expected values.
Smart Images

Figure 2024162653000001_ABST
Abstract
Description
[Technical field]
[0001] The present invention relates to a program, an information processing method, and an information processing device. [Background technology]
[0002] Conventionally, a decision support system is known that quantitatively evaluates risk and profitability, proposes multiple candidate operational strategies, and evaluates and presents the effects on risk and profitability when a candidate strategy is selected (see, for example, Patent Document 1). [Prior art documents] [Patent documents]
[0003] [Patent Document 1] Patent Publication No. 2021-117781 Summary of the Invention [Problem to be solved by the invention]
[0004] The above-mentioned Patent Document 1 supports decision-making on operational strategies for a group of facilities, which is a plurality of facilities, but it is difficult to appropriately judge expected values involving probability theory. For example, humans have a subconscious tendency to avoid risks and to avoid or avoid losses, and it is said that it is difficult to rationally judge expected values while taking these risks and losses into account. In addition, risk aversion and loss aversion depend on a person's own way of thinking.
[0005] Therefore, one aspect of the disclosed technology aims to provide a program, an information processing method, and an information processing device that enable a user to appropriately grasp a user's tendency toward risk aversion and loss avoidance, and to train the user to make rational judgments about expected values. [Means for solving the problem]
[0006] A program in one embodiment of the disclosed technology causes a processor included in an information processing device to perform a process of having a user select at least one option from a plurality of options for identifying a judgment tendency with respect to an expected value; acquire electroencephalogram information measured from an electroencephalogram measuring device worn by the user; learn a first brain activity when risk aversion or loss aversion with respect to the expected value is not performed and a second brain activity when the risk aversion or loss aversion is performed based on the electroencephalogram information acquired during the option selection process, and generate a learning model that estimates the user's brain activity; input the electroencephalogram information acquired during a process of training the user regarding expected value judgment into the learning model, and perform a neurofeedback process to perform the training based on the estimated result of the brain activity. Effect of the Invention
[0007] According to one aspect of the disclosed technology, it is possible to appropriately grasp a user's tendency toward risk aversion and loss avoidance, and train them to make rational judgments about expected values. [Brief description of the drawings]
[0008] [Figure 1] FIG. 1 is a diagram for explaining the premise of the disclosed technology. [Diagram 2] FIG. 1 is a diagram illustrating an example of a coin toss. [Diagram 3] FIG. 1 is a diagram showing an example of tasks A and B. [Figure 4] FIG. 1 is a diagram for explaining an experimental outline of the disclosed technique. [Diagram 5] FIG. 1 is a diagram showing an example of neurofeedback training in this experiment. [Figure 6] FIG. 13 shows a comparison of psychological scales before and after neurofeedback training. [Figure 7] FIG. 13 shows the accuracy of the decoder in the coin toss task. [Figure 8] FIG. 13 is a diagram comparing the results of a coin toss task. [Figure 9]FIG. 13 shows average brain activity and coin toss selection rate in NF sessions. [Figure 10] FIG. 1 shows the accuracy of the decoder in the probability task. [Figure 11] FIG. 13 is a diagram comparing the selection rates for the probability task. [Figure 12] This is a diagram comparing the amount of winnings in probability tasks. [Figure 13] FIG. 1 is a diagram illustrating an example of an outline of a system 1 according to an embodiment. [Figure 14] FIG. 1 is a diagram illustrating an example of an earphone set according to an embodiment. [Figure 15] 1 is a diagram showing an example of a schematic cross section of an earphone according to an embodiment. FIG. [Figure 16] 1 is a block diagram showing an example of an information processing apparatus according to an embodiment; [Figure 17] 1 is a block diagram showing an example of an information processing apparatus according to an embodiment; [Figure 18] 11 is a flowchart illustrating an example of a training process of the processing terminal according to the embodiment. [Figure 19] FIG. 4 is a diagram illustrating an example of a screen transition according to the embodiment. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS
[0009] Hereinafter, an embodiment of the present invention will be described with reference to the drawings. However, the embodiment described below is merely an example, and there is no intention to exclude various modifications and application of techniques not specified below. In other words, the present invention can be implemented with various modifications within the scope of the gist of the present invention. In addition, in the following description of the drawings, the same or similar parts are represented by the same or similar reference numerals. The drawings are schematic and do not necessarily correspond to actual dimensions, ratios, etc. The drawings may also include parts with different dimensional relationships and ratios.
[0010] <Premise> First, before describing the disclosed embodiment, the premise of the disclosed technology will be described. FIG. 1 is a diagram for explaining the premise of the disclosed technology. As shown in FIG. 1, people perceive the value of gain and loss differently. For example, even if the amount is the same, if there is a gain of 1 million yen and a loss of 1 million yen, the value of the loss is perceived to be greater than the value of the gain. Therefore, in a game in which there is a 55% probability of a gain of 1 million yen and a 45% probability of a loss of 1 million yen, the expected value is greater than 0, but there is a strong tendency to want to avoid the loss of 1 million yen, and more people choose not to play this game.
[0011] Therefore, the disclosed technology uses neurofeedback, which will be described later, to help train the brain to be able to rationally judge expected values. For example, the disclosed technology provides investment challenges to induce brain activity that allows the brain to appropriately judge expected values, and prevents excessive risk aversion.
[0012] <Related Paper 1> Next, I will introduce a paper that evaluates brain functions related to investment, for example, regarding expected value judgments (Samanez Larkin, GR, Kuhnen, CM, Yoo, DJ, & Knutson, B. (2010). Variability in nucleus accumbens activity mediates age related suboptimal financial risk taking. Journal of Neuroscience 30 (4), 1426 1434.).
[0013] In the above paper1, because the increase in the elderly investor population could have a significant impact on the global economy, an experiment was conducted to clarify the decision-making tendencies of the elderly using fMRI and an investment task. The contents of the experiment are as follows. (Experimental content) Elderly people (65-81 years old) and young people (19-30 years old) were presented with a low-risk option (bonds) and two high-risk options (two types of stocks) and asked to choose. An example of the options is shown below. 50% chance of +$10, 25% chance of -$10, 25% chance of 0 (expected value +$2.5) 25% chance of +$10, 50% chance of -$10, 20% chance of 0 (expected value -$2.5) 100% chance of $1 The task of choosing one of the three options above is repeated 100 times. The investment ability of each subject was evaluated based on the Bayesian update (the accumulation of information on investments and their results each time) for the subjects' optimal choices. For example, the investment ability of each subject was divided into "risk-seeking mistakes" and "risk-averse mistakes" based on the Bayesian update rules, and was further evaluated based on the amount of "confusion mistakes" in which the subject chooses the option with the lower expected value among high-risk options.
[0014] (Experimental Results) In the above experiment, it was shown that older adults were less likely to choose the optimal option than younger adults, and made more risk-oriented mistakes. It was also reported that there is a strong relationship between risk-oriented mistakes and variability in activity in the Nucleus accumbens (NAcc). In other words, according to related paper 1, it has been shown that there is a relationship between brain activity and investment decisions (an example of expected value judgment).
[0015] <Related Paper 2> We also explain a paper on loss aversion that uses event-related potentials to examine individuals' loss aversion tendencies in a gambling task (Zeng, J., Wang, Y., Zeng, J., Cao, Z., Chen, H., Liu, Y., ... & Su, L. (2019). Predicting the behavioural tendency of loss aversion. Scientific reports, 9(1), 1 7.).
[0016] In the above paper 2, the following experiment was conducted. (Experimental content) Twenty subjects were asked to play a coin toss game in which if the coin lands on heads, they receive X amount of money, and if the coin lands on tails, they lose Y amount of money. The probability of heads or tails is 50%. The amount of loss is determined by the formula Y=LX, where L varies depending on the experimental conditions. For high loss: 0.9≦L≦1.0 For medium loss: 0.5≦L≦0.6 For low loss: 0.1≦L≦0.2
[0017] FIG. 2 is a diagram showing an example of a coin toss. In the example shown in FIG. 2, heads indicates +100 yuan, and tails indicates -90 yuan. L is 0.9, which indicates a high loss condition. In addition, the subjects' brain waves were measured during the coin toss game, and the increase or decrease in LPC (late positive complex) was measured. LPC is an event-related potential observed in tasks that involve high-level processing such as memory, emotion, and language processing, and is said to be large in response to stimuli that include emotional valence.
[0018] (Experimental Results) Many of the subjects chose to bet under the low loss condition, but not under the high loss condition. It has also been reported that when subjects bet under the low loss condition, i.e., when they accepted a certain amount of loss, LPC increased more than when they did not bet under the high loss condition, i.e., when neither profit nor loss occurred. This is consistent with the fact that emotional stimuli increase LPC. Related paper 2 shows that there is a relationship between brain activity and stimuli related to expectation judgment.
[0019] <Related Paper 3> This is a paper on risk aversion, reporting that biofeedback training in stress regulation leads people to choose options with less certainty and higher rewards (Iodice, P., Cannito, L., Chaigneau, et al. Learned self regulation in top level managers through neurobiofeedback training improves decision making under stress. Sci Rep 12, 6127 (2022). doi.org / 10.1038 / s41598 022 10142 x).
[0020] The following experiment was conducted in the above paper 3. (Experimental content) The subjects were 23 male managers working at the headquarters of a multinational corporation. Prior to training, four baseline psychometric measures were administered. Decision-making style: how an individual approaches a decision-making situation (rational, avoidant, dependent, intuitive, spontaneous) Impulsivity: the higher the total score, the higher the level of impulsivity. Emotional state: positive or negative Ability to perceive internal physiological states: Measures awareness of one's own internal state, including distractibility and emotional awareness
[0021] During the training, stress tolerance training was performed for 25 minutes. Experimental group: The temperature of an on-screen thermometer, which reflected skin conductance levels, was alternately raised and lowered, and subjects were helped to relax by listening to binaural beats and guided meditation. Control group: Subjects were asked to watch a video that was unrelated to the purpose of the experiment.
[0022] After stress induction, one of the following tasks is performed: ·Assignment A Choose from two options that combine three monetary values and two immediate rewards (e.g., 25 euros today vs. 35 euros in 25 days) ·Assignment B Choose between a small guaranteed reward and a larger amount offered based on probability (e.g. 20€ guaranteed vs. 30% chance of winning 80 users) Fig. 3 shows examples of tasks A and B. As shown in Fig. 3, in paper 3, different types of tasks were presented to the subjects.
[0023] (Experimental Results) It was found that the experimental group that underwent stress adjustment training showed a longer response time and a reduced instinctive reaction after the training. In addition, the analysis results of tracking the subjects' cursor movements showed the following: The control group showed more instinctive reactions (= shorter response time), but the experimental group showed a lower MD value, despite taking longer to react, which shows that the subjects selected the option with more confidence. MD value: A unique geometric calculation of the trajectory of the cursor from the start button to the selection of an option. Higher values indicate greater attraction to non-selected options during the trajectory toward the selected option (= more indecision).
[0024] Related paper 3 suggests that biofeedback training regulates an individual's stress and enhances their self-regulation ability, which in turn improves their ability to manage stress, leading to them choosing options that are less certain but have a higher return.
[0025] Here, the inventor of the disclosed technology furthered the study of the findings of the above-mentioned papers 1 to 3, etc., and conducted the following experiment in the hope that the rationality of an individual's expectation judgment can be improved by neurofeedback using electroencephalogram information. An experiment was conducted using an investment task in which monetary gains and losses occur as the actual task for expectation judgment.
[0026] <Experiment Overview> 4 is a diagram for explaining an outline of an experiment of the present disclosure. The outline of the experiment shown in FIG. 4 is performed according to the following steps (1) to (4). (1) Subjects were asked to perform an investment task while their brainwave information was being measured. The brainwave measuring device used was an earphone-type brainwave measuring device as shown in Figs. 14 and 15, which will be described later. In this experiment, each subject performed 12 practice trials, and then performed a total of 90 tasks (30 trials x 3 sessions).
[0027] The challenges are as follows: The probability of the coin landing on heads or tails is 50% If it lands on heads, you win X yen, if it lands on tails, you lose Y yen. The value of X is randomly set within the range of 100 to 1500 yen. Loss amount is calculated as Y=LX The value of L is set under the following three conditions: For high loss: 0.9≦L≦1.0 For medium loss: 0.5≦L≦0.6 For low loss: 0.1≦L≦0.2 Within one session (30 trials), each of the High, Medium, and Low conditions was presented ten times in a random order.
[0028] (2) The investment tendency of each subject is analyzed using the EEG information. The investment tendency may be, for example, a tendency to avoid risk or loss, making it difficult to obtain a large gain, or a tendency to take risks, making it possible to obtain a large gain. A learning model (decoder) is generated that classifies the brain activity of each subject based on the EEG information during task processing. For example, the first brain activity when a high-risk option with a high expected value (high-gain option) is selected, and the second brain activity when a low-risk option with a low gain is selected, and the EEG information can be made visible as to which brain activity it corresponds to.
[0029] In this experiment, a learning model (decoder) is created that classifies the first brain activity based on the EEG information when the coin toss is selected, and the second brain activity based on the EEG information when the coin toss is not selected. In addition, for each L that varies the expected value, the total value of the results when all coin tosses are selected (expected acquisition value) and the acquisition value that each subject actually acquired are presented to each subject.
[0030] A preliminary questionnaire was then administered to each subject, consisting of 16 questions. A. Investing seems fun. B. Investing is an efficient way to increase assets. C1. I want to try investing (if you are already doing it, do you want to continue doing it) C2. I would like to invest a portion of my current savings. D. I don't want to invest because I have a strong image of losing money. E. I don't want to invest because there is a lot of uncertainty. F. Investing is something that professionals do, not something that you should actively do. G. I don't have enough knowledge so I shouldn't invest. H. I'm not confident in my investment skills, so I shouldn't invest. I. Collecting information and studying for investing is a hassle J. If I have any surplus money, I will save it as much as possible and not consider increasing it through investment. K. If I had any surplus funds, I would build a balanced portfolio by not only saving but also investing in risk assets. L. I am attracted to short-term investments (such as stock investments) that have high liquidity and quick returns. M. I am attracted to long-term investments (such as bond investments) that have low liquidity and can provide stable returns. N. If I were to invest, I would like to prioritize profitability (high yields and the expectation of price increases) O. If I were to invest, I would like to prioritize stability (I can't expect a big price increase, but the possibility of a big price drop is also low)
[0031] (3) Neurofeedback processing is performed. In the neurofeedback processing, for example, processing is performed to induce the subject to enter a first brain activity state or a second brain activity state according to the subject's electroencephalogram state.
[0032] (4) After the neurofeedback process, a task is performed. This makes it possible to measure the effect of the neurofeedback process. For example, if the subject is trained by the neurofeedback process to select the option with the higher expected value, a subsequent investment task can be used to verify whether there is a difference in the amount of money earned before and after the experiment. Note that the repeated execution of (3) and (4) is referred to as neurofeedback training (hereafter also referred to as "NT").
[0033] Two types of neurofeedback training (NTA and NTB) were conducted. (NTA) Neurofeedback duration: 30 seconds An object (e.g., a banknote) is displayed that changes size according to an individual's brain activity, and the subject is instructed to try and make the object larger or smaller through trial and error. For example, neurofeedback processing is an instruction to make the amount of money larger under conditions where the amount would be larger if it were closer to the brain activity experienced when tossing a coin, while invNF is an instruction to make the amount of money smaller. After neurofeedback processing, the subjects were asked to select a coin toss task option. - Five trials were conducted per task (toss a coin or not) x one neurofeedback session (hereafter referred to as "NF"). Each block: The selection rate of the coin toss is evaluated for 4 blocks of alternating invNF and NF, for a total of 20 trials.
[0034] (NTB) Neurofeedback duration: 30 seconds An object (e.g., a banknote) is displayed that changes size according to the individual's brain activity, and the subject is instructed to try and make the object larger or smaller through trial and error. For example, NF is an instruction to increase the amount of money under conditions where the amount would be larger if the brain activity were to approach that observed when a risky option with a high expected value was selected, while invNF is an instruction to decrease the amount of money. After neurofeedback processing, the subjects were presented with options for the probability task and asked to select one. The probability task is as follows: The following two options are displayed on the screen, and the subject must select one. Option 1: You will definitely receive the amount shown. Option 2: You will receive the displayed amount with a specified probability (there is a chance that you will not receive it). Example: 4,000 yen is 100% or 10,000 yen is 50% Condition: An object (e.g., a banknote) is presented that changes size depending on the individual's brain activity. NF: Instructed to increase the amount of money under conditions where risk is present and the amount of money will increase if brain activity approaches that of when choosing the larger expected value. invNF: Instruct to make the amount smaller Task (choose one of the options) x 5 trials per NF (neurofeedback) Each block: The selection rate of the coin toss is evaluated for 4 blocks of alternating invNF and NF, for a total of 20 trials.
[0035] FIG. 5 is a diagram showing an example of neurofeedback training in this experiment. FIG. 5(A) shows neurofeedback training A (NTA), in which the task includes a coin toss task. In the example shown in FIG. 5(A), an object (1000 yen bill) showing brain activity is displayed on the screen, and the instruction "Please make it bigger" is displayed on the screen, and the subject tries to make the object bigger or smaller by trial and error. At this time, the size of the object is determined so that the object becomes bigger as it approaches brain wave information that indicates taking a risk in the task and selecting an option with a high expected value.
[0036] After training using the size of the object through neurofeedback processing, the coin tossing task is performed a predetermined number of times. One session is composed of the neurofeedback processing and the coin tossing task, and a predetermined number of sessions are performed.
[0037] FIG. 5(B) shows neurofeedback training B (NTB), in which the task includes a probability task. In the example shown in FIG. 5(B), objects showing brain activity (1000 yen bill and 10,000 yen bill) are displayed on the screen, and the instruction "make it bigger" is displayed on the screen for one of the bills, and the subject tries to make the object bigger by trial and error. At this time, the size of the object is determined so that the 10,000 yen bill object becomes larger if it approaches brain wave information in which a risk-taking option with a high expected value is selected in the task, or the 1000 yen bill object becomes larger if it approaches brain wave information in which a risk-avoiding or loss-avoiding option is selected.
[0038] After training is performed using the size of the object through neurofeedback processing, a probability task is performed a predetermined number of times. A session is composed of the neurofeedback processing and the probability task, and a predetermined number of sessions are performed.
[0039] After neurofeedback training (NT), the 16 questions described above were asked again to each subject.
[0040] <Experimental Results> Next, the results of the above-mentioned experiment will be described. Figures 6 to 12 are diagrams showing the evaluations of the above-mentioned experiment. Figure 6 is a diagram showing a comparison of psychological scales before and after neurofeedback training. The psychological scales shown in Figure 6 are psychological scales for a 16-item questionnaire, and the questionnaire for each alphabet shows the following psychological scales. A. Interest in investing, B. Recognition of benefits of investing, C (C1+C2). Intention to invest, D. Negative image of investing (1) Aversion to loss, E. Negative image of investing (2) Aversion to uncertainty, F. Psychological distance from investing, G. Perceived lack of knowledge about investing, H. Perceived lack of skills in investing, I. Psychological cost of starting to invest, J. Asset formation: Preference for saving, K. Asset formation: Orientation towards balance, L. Attraction to highly liquid assets, M. Attraction to less liquid assets, N. Attraction to profitable products, O. Attraction to stable assets.
[0041] In the example shown in Figure 6, a significant difference can be seen between the items "H. Perception of insufficient investment skills" and "L. Attraction of highly liquid assets" (P<0.05). In other words, after the neurofeedback training, the subjects became more confident in their investment skills in terms of the perception of insufficient investment, and in terms of the attraction of highly liquid assets, the subjects felt that investments with high liquidity were more attractive.
[0042] FIG. 7 is a diagram showing the accuracy of the decoder in the coin tossing task. In the example shown in FIG. 7, the average accuracy of the decoder for the 14 subjects is 74.95%. The accuracy of the decoder indicates the accuracy of appropriately classifying brain activity during coin tossing (first brain activity) and brain activity during non-coin tossing (second brain activity) from EEG information. The results shown in FIG. 7 show that it is possible to distinguish between coin tossing and non-coin tossing from EEG information.
[0043] The average percentage of participants who chose the coin toss before neurofeedback training was about 20% under the high loss condition, about 60% under the medium loss condition, and about 100% under the low loss condition. In addition, the amount actually won before neurofeedback training was lower than the expected value when a coin was tossed in both the high loss and medium loss conditions. This is thought to be because the rate at which participants actually chose the coin toss was low, and therefore the amount won was also low.
[0044] Figure 8 is a diagram comparing the results of the coin toss task. In the example shown in Figure 8, the average amount earned in each condition before neurofeedback training was just under 120 yen, after neurofeedback training aimed at liquid investment (after NF) the average amount earned in each condition was just over 120 yen, and after neurofeedback training aimed at stable investment (after invNF) the average amount earned in each condition was just under 60 yen. Here, there was a significant difference between after NF and after inverse NF (invNF) (P<0.05). This result shows that the amount earned differs depending on the method of neurofeedback training.
[0045] Figure 9 shows the average brain activity and the coin toss selection rate in the NF session. Note that the NF session also includes the inverse NF. In the example shown in Figure 9, a slight positive correlation (slope 0.27) was observed between the decoder's decoded result and the coin toss selection rate. In other words, the closer the decoder's decoded result is to 1 (closer to the brain activity when selecting the coin toss), the higher the coin toss selection rate is.
[0046] Fig. 10 is a diagram showing the accuracy of the decoder in a probability task. In the example shown in Fig. 10, the average accuracy of the decoder for the 16 subjects is 73.79%. The accuracy of the decoder indicates the accuracy of appropriately classifying brain activity (first brain activity) when fluidity and high expectation are selected in a probability task from brain wave information, and brain activity (second brain activity) when stability and low expectation are selected. The results shown in Fig. 10 show that it is possible to distinguish between when fluidity and high expectation are selected and when stability and low expectation are selected from brain wave information.
[0047] Here, the amount actually won before the neurofeedback training (just under 5,000 yen) was lower than the maximum expected value (approximately 5,500 yen) if the higher expected value was chosen. This suggests that each subject had a tendency to choose the low expected value option due to risk aversion.
[0048] Figure 11 is a diagram comparing the selection rates of probability tasks. In the example shown in Figure 10, the probability of choosing a risky option before neurofeedback training was about 50%, the probability of choosing a risky option (high risk but high return option) after neurofeedback training (after NF) with a preference for liquid investment was about 70%, and the probability of choosing a risky option after neurofeedback training (after invNF) with a preference for stable investment was about 60%. In particular, there was a significant difference between after NF and before neurofeedback training (P<0.05). This indicates that the effect of NF neurofeedback training caused the subjects to choose the risky option.
[0049] Figure 12 is a diagram comparing the amount of money earned in the probability task. In the example shown in Figure 12, the average amount earned in each condition before neurofeedback training was about 4,800 yen, and after neurofeedback training (after NF), which is oriented toward fluidity and high expected value, the average amount earned was just over 5,000 yen, and these differences were significant (P<0.05). This shows that the amount earned increased significantly as a result of neurofeedback training.
[0050] <Summary of this experiment> Before training, participants in the coin toss task would not make a choice even if the expected value was greater than 0. This indicates a tendency to make an opportunity loss as a result of loss aversion. Furthermore, during neurofeedback training, participants in the coin toss task were slightly more likely to choose the coin toss, which could be said to weaken the tendency to lose loss aversion.
[0051] In the probability task, before training, participants were observed to have a tendency to take risk-averse choices and lose opportunities. In the probability task during neurofeedback training, participants were significantly more likely to take risky options, and the amount of money won increased significantly.
[0052] The subjects performed a coin toss task and a probability task, and by experiencing neurofeedback training, changes in their financial attitudes were observed. For example, their aversion to uncertainty was slightly reduced, and they became more confident in their investment skills. In addition, their preference for certainty in terms of savings was softened, and their attitude toward liquid assets became neutral. This can be said to be because the subjects became aware of their own decision-making tendencies through the financial decision-making task, and were trained using neurofeedback processing, etc. Therefore, grasping the user's expected value judgment tendency and conducting neurofeedback training to bring about the user's desired inclination is thought to be useful, for example, as content for users who make investments. For example, it can be said to be promising as introductory investment content for beginners and users with little experience.
[0053] As described above, it was found that performing neurofeedback processing on expectation judgment increases the possibility that the user will approach the desired inclination. Therefore, the inventors have constructed the following system to generalize and realize this experiment.
[0054] [Embodiment] Hereinafter, an overview of the system according to the embodiment will be described with reference to the drawings. <System Overview> 13 is a diagram illustrating an example of a schematic of a system 1 according to an embodiment. In the system 1, a user who is to measure an electroencephalogram wears an earphone set 10 having bioelectrodes provided in the ear canal and around the back of the head as an electroencephalogram measuring device.
[0055] 13 shows an example of a neck-hanging earphone set 10, but any electroencephalogram measuring device may be used as long as it can sense an electroencephalogram signal from the ear canal and can sense the electroencephalogram signal from the back of the head. For example, an earphone set that obtains a reference signal from the earlobe, an earphone that obtains a reference signal or a ground signal from another position (another position in the ear canal), or a completely wireless earphone may be used.
[0056] The user may also wear, for example, a headgear that measures brain waves using the International 10 / 20 system as an electroencephalogram measuring device. The electroencephalogram measuring device includes devices capable of measuring brain waves, such as a measuring device that uses scalp electrodes, a measuring device that measures brain activity using intracranial electrodes, a measuring device that measures brain activity using functional magnetic resonance imaging (fMRI), and a measuring device that measures brain activity using near-infrared spectroscopy (NIRS).
[0057] 13, the earphone set 10 acquires electroencephalogram information from the ear canal or near the back of the head, and transmits the electroencephalogram information to the information processing device 30 or the information processing device 50 via the network N. The earphone set 10 may also perform predetermined processing on the electroencephalogram information and transmit it to the information processing device 30 or the information processing device 50. The predetermined processing includes amplification, sampling, filtering, difference calculation, and the like.
[0058] The information processing device 30 is, for example, a server, and generates a learning model that classifies brain activity using the user's brain wave information measured by an electroencephalogram measuring device while processing a task related to expectation value judgment. The information processing device 30 inputs the user's brain wave signal into the learning model, and performs neurofeedback processing by displaying an object based on the estimated brain activity on a screen. The information processing device 30 may also combine the neurofeedback processing with the task processing to output the result of the task processing to the user and inform the user of the effect of the neurofeedback processing.
[0059] The above-described series of processes may be implemented as an application executed on a web browser or a native application installed and executed on a smartphone, etc. The information processing device 30 may transmit the generated application (hereinafter also referred to as "app") to the information processing device 50 of each user.
[0060] The information processing device 50 is, for example, a processing terminal such as a mobile terminal held by a user, and sequentially acquires brainwave information from the earphone set 10. The information processing device 50 inputs the sequentially acquired brainwave information into a training application, and uses neurofeedback to bring brain activity into a desired state, thereby training the user to be able to make expected value judgments.
[0061] <Earphone set configuration> 14 to 15 provide an overview of the earphone set 10 according to the embodiment. Note that the earphone set 10 is not limited to the examples shown in Figs. 14 to 15, and any earphones can be applied to the technology of the present disclosure as long as they are capable of sensing brain waves from the ear canal or around the back of the head and capable of outputting to an external device.
[0062] Fig. 14 is a diagram showing an example of an earphone set 10 according to an embodiment. The earphone set 10 shown in Fig. 14 has a pair of earphones 100R, 100L and a neck hanging part 110. Each of the earphones 100R, 100L is connected to the neck hanging part 110 using a cable capable of signal communication, but may also be connected using wireless communication. Hereinafter, RL will be omitted when there is no need to distinguish between left and right.
[0063] The neck hanging part 110 has a central member that fits along the back of the neck, and rod-shaped members (arms) 112R, 112L that are curved along both sides of the neck. Electrodes 122, 124 for sensing brainwave information are provided on the surface of the central member that comes into contact with the neck on the back side. For example, one of the electrodes 122, 124 is an electrode that senses brainwave signals from the back of the head Oz, and the other is an electrode that is connected to earth or a reference electrode. The neck hanging part 110 may also have a processing unit that processes brainwave information and a communication device that communicates with the outside, but these processing units and communication units may be provided in the earphones 100.
[0064] In addition, the tip sides of the rod-shaped members 112R, 112L on both sides of the neck hanging part 110 are heavier than the base sides (center member sides), so that the electrodes 122, 124 are appropriately pressed against the neck of the wearer. For example, weights are provided on the tip sides of the rod-shaped members 112R, 112L. Note that the positions of the electrodes 122, 124 are not limited to these positions.
[0065] Fig. 15 is a diagram showing an example of a schematic cross section of an earphone 100R according to an embodiment. The earphone 100R shown in Fig. 15 may have an elastic member (e.g., urethane) 108 provided between the speaker 102 and the nozzle 104. Providing this elastic member 108 makes it difficult for vibrations from the speaker 102 to be transmitted to the elastic electrode of the ear tip 106, preventing sound interference between the elastic electrode of the ear tip 106 and the speaker 102.
[0066] Furthermore, ear tip 106 including an elastic electrode is located at the sound guide port, but the elasticity of the elastic electrode itself makes it possible to prevent interference from sound vibrations. Also, by using an elastic material for the housing, this elastic material makes it difficult for sound vibrations to be transmitted to the elastic electrode of ear tip 106, making it possible to prevent interference from sound vibrations.
[0067] Furthermore, the ear tip 106 conducts brainwave information sensed from the ear canal to the contact of an electrode provided in the nozzle 104. The brainwave signal is transmitted from the ear tip 106 to a biosensor (not shown) inside the earphone 100 via the contact. The biosensor outputs the sequentially acquired brainwave information via a cable to a processing device provided in the neck hanging part 110, or transmits it to an external device. Furthermore, the ear tip 106 may be insulated from the housing including the biosensor and the audio sound processor.
[0068] <Server configuration example> FIG. 16 is a block diagram showing an example of an information processing device 30 according to an embodiment. The information processing device 30 is, for example, a server, and may be composed of one or more devices. The information processing device 30 acquires and processes electroencephalogram information (including electroencephalogram signals) from each user terminal, and generates or supports, for example, neurofeedback related to expectation value judgment. The server 10 may generate or support training processing executed on the user terminal side. The information processing device 30 is also referred to as server 30. The information processing device 30 does not necessarily have to be a server, and may be a general-purpose computer.
[0069] The server 30 includes one or more processors (CPUs: Central Processing Units) 310, one or more network communication interfaces 320, memory 330, a user interface 350, and one or more communication buses 370 for interconnecting these components.
[0070] Server 30, for example, may optionally include a user interface 350. User interface 350 may include a display device (not shown), and a keyboard and / or a mouse (or some other input device, such as a pointing device, not shown).
[0071] The memory 330 may be, for example, a high-speed random access memory such as an SSD, DRAM, SRAM, DDR RAM, or other random access solid-state storage device, or may be a non-volatile memory such as one or more magnetic disk storage devices, optical disk storage devices, flash memory devices, or other non-volatile solid-state storage devices. The memory 330 may be a non-transitory recording medium readable by a computer.
[0072] As another example, memory 330 may be one or more storage devices located remotely from processor 310. In one embodiment, memory 330 stores the following programs, modules and data structures, or a subset thereof:
[0073] The one or more processors 310 read and execute a program from the memory 330 as necessary. For example, the one or more processors 310 may configure a training control unit 312, an acquisition unit 313, a task processing unit 314, a learning unit 315, a training unit 316, and an output unit 317 by executing a program stored in the memory 330. The training control unit 312 controls and processes the electroencephalogram information acquired in sequence, and controls each of the following processes.
[0074] The acquiring unit 313 sequentially acquires brain wave information, for example, brain wave signals, measured by a brain wave measuring device worn by a user, for example, a bioelectrode included in the earphone set 10. The brain wave measuring device is not limited to the earphone set 10. The acquiring unit 317 may also acquire brain wave information from a user terminal that has acquired the brain wave information from the brain wave measuring device.
[0075] The task processing unit 314 performs a process of having the user select at least one option from among multiple options related to the specification of expected value judgment. For example, the task processing unit 314 performs a process of presenting multiple options using expected value to the user regarding risk aversion or loss aversion, and having the user select at least one option. The options may also include not selecting an option as one option. The task processing unit 314 executes a process such as a questionnaire and / or a game to specify the user's judgment inclination regarding expected value. As a specific example, the task processing unit 314 executes the above-mentioned coin toss task and probability task, but is not limited to these examples.
[0076] The learning unit 315 learns the first brain activity when risk aversion or loss aversion is not performed with respect to the expected value and the second brain activity when risk aversion or loss aversion is performed based on the electroencephalogram information acquired during the selection process of the option included in the task, and generates a learning model for estimating the user's brain activity. For example, the learning model uses a neural network.
[0077] For example, the learning unit 315 inputs the electroencephalogram information when the first option with fluidity and high expectation value is selected and the electroencephalogram information when the second option with certainty and low expectation value is selected as learning data into the learning model, and performs machine learning. The learning unit 315 performs machine learning using the learning data to generate a trained learning model that solves a classification problem in which either the first brain activity or the second brain activity is an estimated result, based on the electroencephalogram information. In addition, as in a coin toss task, the time when an option is selected may be the time when the first option is selected, and the time when no option is selected may be the time when the second option is selected.
[0078] The training unit 316 inputs electroencephalogram information acquired during the training process into the learning model, and performs neurofeedback processing to train the user on expected value judgment based on the estimated brain activity. For example, the training unit 316 controls the display of an object that changes based on the estimated brain activity on the screen, outputs instruction information for the user, and performs processing to guide the user to achieve a desired brain activity.
[0079] As a specific example, when using an image of a banknote as an object, the training unit 316 outputs instruction information to make the banknote larger when inducing a first brain activity, and outputs instruction information to make the banknote smaller when inducing a second brain activity.
[0080] The training unit 316 may increase the size of the object as the electroencephalogram information approaches the electroencephalogram information during the first brain activity, and may decrease the size of the object as the electroencephalogram information approaches the second brain activity. In addition, the color, shape, and the like may be used as a change target for the object other than the size.
[0081] The above processing makes it possible to appropriately grasp the user's tendency toward risk aversion and loss avoidance, and train the user to be able to rationally judge expected values. The learning model may also be generated using the user's own electroencephalogram information, in which case it is possible to grasp the user's unique electroencephalogram information and improve the rationality of expected value judgment in accordance with the user's inclinations. The learning model does not necessarily have to be customized for each individual user, and may be learned using electroencephalogram information from multiple people.
[0082] The output unit 317 outputs training result information in the neurofeedback processing. For example, the output unit 317 may include, as the result information, whether or not the brain activity has become appropriately controlled, specifically, whether or not the brain has been trained to be in the first or second brain activity state.
[0083] The above process allows the user to check the result of the neurofeedback process in a timely manner and to understand whether the brain training is going well or not. This makes it easier for the user to get a sense of induction to the first brain activity or the second brain activity.
[0084] The learning unit 315 may also identify the user's tendency based on the options selected by the user. For example, the learning unit 315 may identify the user's selection tendency or selection inclination, such as whether the user is more likely to select an option with liquidity and high expected value, or more likely to select an option with certainty and low expected value, or whether there is no difference between the two, based on the selected options. The selection tendency may also include objective figures such as a selection rate.
[0085] The output unit 317 may also output trend information regarding the user's tendency to the user. For example, the output unit 317 may output the user's selection tendency after the task processing for generating the learning model is performed. The output method may be either display on the screen of the user terminal or output by voice.
[0086] By the above process, it is possible to inform the user of the judgment tendency of the user with respect to the expected value. In addition, by notifying the tendency using objective figures, it is possible to inform the user of the tendency of the expected value judgment that the user himself is not aware of.
[0087] The acquiring unit 313 may also acquire user instruction information related to the user's selection tendency. For example, the acquiring unit 313 may acquire instruction information input by the user operating the user terminal. The instruction information may include an orientation related to expected value judgment, and may include a judgment tendency desired by the user. For example, the instruction information may include a judgment tendency such as a desire to select a selection with fluidity and high expected value.
[0088] The training unit 316 may also change the training content of the neurofeedback processing according to the instruction information acquired in the neurofeedback processing. For example, when the instruction information includes a tendency toward fluidity and high expectation, the training unit 316 may change the expectation value of the tasks in training to a higher value, change the tasks themselves, or change the number of tasks to be presented. Through the above processing, it becomes possible to carry out more appropriate training according to the instruction information.
[0089] Furthermore, when the instruction information includes an instruction to weaken the tendency to avoid risk or loss, the training unit 316 may perform a process to induce the electroencephalogram information during the training process to approach the electroencephalogram information during the first brain activity. For example, in the case of a coin toss task, the training unit 316 may instruct to increase the size of an object (e.g., a banknote) displayed on the screen. The above process may lead to more selection of options with high liquidity and expectation, thereby increasing the expected value of acquisition.
[0090] The training unit 316 may also include performing a process of having the user select an option using the expected value after the neurofeedback process. For example, the training unit 316 may instruct the task processing unit 314 to present a coin toss task and / or a probability task to the user after the neurofeedback process, and the task processing unit 314 may present the coin toss task and / or the probability task to the user according to the instruction of the training unit 316.
[0091] By carrying out the above-described process, the effect of neurofeedback can be immediately confirmed by carrying out the task after the neurofeedback process.
[0092] The training control unit 312 may repeat the neurofeedback processing by the training unit 316 and the selection after the neurofeedback processing by the task processing unit 314 until a termination condition is satisfied. The termination condition may be, for example, a number of times, or a case where the value of the current electroencephalogram information is included within a predetermined range including a desired brain activity value. For example, the training control unit 312 may repeat the session until a plurality of sessions have been performed, with one session consisting of a neurofeedback processing for a predetermined time and a task performed a predetermined number of times.
[0093] By repeating the above process, the user can repeatedly perform the neurofeedback and the tasks, and can repeatedly perform training and check the results. The user can gradually train to achieve the desired brain activity state, and the training results can be solidified. The training control unit 312 may also control to display an input field on the screen so that the user can input information during or after the neurofeedback training. For example, if the user has acquired a knack or hint for transitioning to the desired brain activity by repeating the training, the user inputs the knack or hint into the input field. As a specific example, if the user thinks of a certain proverb or the like and the desired brain activity is achieved, the user may input the proverb or the like into the input field, and the training control unit 312 may associate the input information (proverbs, etc.) with the desired brain activity and store it in the memory 330. In the neurofeedback training after the information is input into the input field, the training control unit 312 may control to obtain information associated with the desired brain activity from the memory 330 and display it on the screen during the training.
[0094] In addition, the output unit 315 may output an application for performing the above-mentioned neurofeedback training to a specified information processing device 50, or output it from an app store or the like to a specified information processing device 50 via the network communication interface 320.
[0095] As described above, the server 30 can configure or provide an application that performs neurofeedback regarding expectation value judgment, using the electroencephalogram information of an individual user who wears an electroencephalogram measuring device capable of measuring electroencephalogram signals in the ear canal and around the back of the head.
[0096] <Example of processing terminal configuration> 17 is a block diagram showing an example of an information processing device 50 according to an embodiment. The information processing device 50 is a user terminal such as a mobile terminal (such as a smartphone), a computer, or a tablet terminal. The information processing device 50 is also referred to as a processing terminal 50.
[0097] The processing terminal 50 includes one or more processors (eg, CPUs) 510, one or more network communication interfaces 520, a memory 530, a user interface 550, and one or more communication buses 570 for interconnecting these components.
[0098] The user interface 550 includes a display device 551 and an input device (such as a keyboard and / or a mouse or any other pointing device) 552. The user interface 550 may also be a touch panel.
[0099] The memory 530 may be, for example, a high-speed random access memory such as an SSD, DRAM, SRAM, DDR RAM, or other random access solid-state storage device, or may be a non-volatile memory such as one or more magnetic disk storage devices, optical disk storage devices, flash memory devices, or other non-volatile solid-state storage devices. The memory 530 may be a non-transitory recording medium readable by a computer.
[0100] Additionally, another example of memory 530 may include one or more storage devices located remotely from processor 510. In one embodiment, memory 530 stores the following programs, modules and data structures, or a subset thereof:
[0101] The one or more processors 510 read and execute a program from the memory 530 as necessary. For example, the one or more processors 510 may configure an application control unit (hereinafter also referred to as an "application control unit") 512 by executing a program stored in the memory 530. This application control unit 512 includes an application that performs neurofeedback processing related to the above-mentioned expected value judgment. For example, the application control unit 512 has an acquisition unit 513, a task processing unit 514, a learning unit 515, a training unit 516, and an output unit 517.
[0102] The acquisition unit 513 acquires electroencephalogram information measured by an electroencephalogram measuring device worn by a user. For example, it is preferable that the electroencephalogram measuring device can acquire electroencephalogram signals from at least two points, such as the earphone set 10 shown in Fig. 14, and can acquire electroencephalogram signals around the occipital lobe that includes the visual cortex.
[0103] The task processing unit 514, the learning unit 515, the training unit 516, and the output unit 517 perform the same processes as the task processing unit 314, the learning unit 315, the training unit 316, and the output unit 317 of the corresponding server 30. The output unit 517 outputs the training results to a display 551. Selections for tasks are input by an input device 552, and the selection results are input to the task processing unit 514.
[0104] As described above, the processing terminal 50 of the user can also execute the same processing as the server 30, and therefore the effect of the above-mentioned processing can be achieved in the server 30. The processing terminal 50 may also utilize processing provided in another device (e.g., the server 30 on the cloud, etc.). For example, the learning unit 515 may not be included in the processing terminal 50, and the acquisition unit 513 may acquire a learning model (e.g., a decoder) generated by the server 30, and the training unit 516 may perform neurofeedback processing using the acquired learning model.
[0105] <Operation> Next, an operation of the system 1 according to the embodiment will be described. Fig. 18 is a flowchart showing an example of training processing of the processing terminal 50 according to the embodiment.
[0106] In step S102, the task processing unit 514 performs a process of having the user select at least one option from a plurality of options regarding the specification of the judgment orientation for the expected value. For example, the task processing unit 514 executes the process of the coin toss task or the probability task described above.
[0107] In step S104, during the task processing, the acquisition section 513 acquires electroencephalogram information measured by an electroencephalogram measuring device worn by the user.
[0108] In step S106, the task processor 514 determines whether or not a pre-processing termination condition is satisfied. The pre-processing termination condition includes, for example, whether or not a predetermined number of task processes have been completed. If the termination condition is satisfied (step S106-YES), the process proceeds to step S108, and if the termination condition is not satisfied (step S106-NO), the process returns to step S102.
[0109] In step S108, the learning unit 515 learns the first brain activity when risk aversion or loss aversion is not performed with respect to the expected value and the second brain activity when risk aversion or loss aversion is performed based on the electroencephalogram information acquired during the option selection process, and generates a learning model that estimates the user's brain activity.
[0110] In step S110, the training unit 516 inputs the electroencephalogram information sequentially acquired by the acquisition unit 513 into the learning model, and performs a neurofeedback process to train the user on expected value judgment based on the estimated brain activity. For example, the training unit 516 performs training for a predetermined time to induce the user's brain to have a desired brain activity using an object that changes according to the estimated brain activity. As a specific example, the induction process includes instructing the user to change the object, causing a predetermined change in the object, that is, guiding the user to have a desired brain activity corresponding to the change in the object.
[0111] In step S112, the task processing unit 514 performs a process of having the user select at least one option from a plurality of options regarding the specification of the judgment orientation for the expectation value. This task processing may be the same as or different from the task in step S102.
[0112] In step S114, the application control unit 512 determines whether or not a termination condition of the neurofeedback training is satisfied. The termination condition of the neurofeedback training includes, for example, whether or not a predetermined number of sessions (combinations of NF and tasks) have been completed. If the termination condition is satisfied, the process proceeds to step S116, and if the termination condition is not satisfied, the process returns to step S110.
[0113] In step S116, the output unit 517 outputs the training result information in the neurofeedback processing.
[0114] 18, the task processing in step S112 is not necessarily required, and in this case, the end condition in S114 may include whether a predetermined time has elapsed. As described above, the generation of the learning model in step S108 may be executed by the server 30, and the neurofeedback processing may be executed using the learned learning model (e.g., brain activity estimation algorithm) acquired from the server 30.
[0115] <Screen example> Next, a description will be given of an example of a screen of the processing terminal 50 on which an application of the disclosed technology is installed. Fig. 19 is a diagram showing an example of a screen transition according to an embodiment. A screen D10 in Fig. 19 shows an example of a screen for starting a task related to identifying a judgment orientation for an expected value. The task shown in Fig. 19 is a task related to investment, and options for increasing the user's money are displayed and the user is asked to select one.
[0116] Screen D10 is an example of a screen on which a predetermined number of tasks are displayed, controlled by task processing unit 514. For each task, the user selects the task that he or she thinks will increase the amount of money in hand. For example, the probability task described above is displayed, but a coin toss task may also be displayed. At this time, electroencephalogram information is acquired by acquisition unit 513 from an electroencephalogram measuring device worn by the user during task processing.
[0117] Screen D12 is an example of a screen showing a result controlled by the output unit 517 after the task processing before the neurofeedback. Screen D12 displays the amount of money acquired for the option selected by the user and the expected acquisition value when the option with the high expected value is selected, and may display the user's judgment inclination. At this time, the learning unit 514 generates a learning model that estimates brain activity based on the electroencephalogram information acquired in sequence during the task processing.
[0118] Screen D14 is an example of a screen where the user sets a goal, which is controlled by the application control unit 512. In screen D14, the user is prompted to practice increasing the acquired value, and the user selects "YES." The selection result at this time becomes instruction information in the user's selection orientation.
[0119] Screen D16 is an example of a screen of neurofeedback processing controlled by the training unit 516. In screen D16, the training unit 516 instructs the user to enlarge, for example, a 10,000 yen object. This indicates training for bringing the user closer to the first brain activity state. The training unit 516 may also instruct the user to enlarge, for example, a 1,000 yen object. This indicates training for bringing the user closer to the second brain activity state.
[0120] Screen D18 is an example of a screen on which a predetermined number of tasks are displayed, controlled by task processing unit 514. The tasks displayed on screen D18 may be the same type of task as the task before neurofeedback, or may be a different type of task, as long as they are related to investment tasks.
[0121] Here, the neurofeedback processing by the training unit 516 and the task processing by the task processing unit 514 are regarded as one session, and the application control unit 512 controls so that a predetermined number of sessions are performed. When the predetermined number of sessions have been performed, the screen transitions from screen D18 to screen D20.
[0122] Screen D20 is an example of a screen showing a result that is displayed and controlled by the output unit 517 after neurofeedback processing. For example, if the actual acquired value for the task during neurofeedback training increases as a result of training using neurofeedback processing, this result is displayed. Screen D20 may also include a comparison result before and after neurofeedback. For example, the comparison result may include a comparison of the selection rate for the first option, a comparison of selection tendency analyzed from the selection result, and the like.
[0123] <Application Examples> The above-mentioned application is considered to be useful for users who want to rationally judge expected values. For example, the application of the disclosed technology is suitable as an application for training beginners or people with little experience in investing to appropriately judge expected values and avoid opportunity losses.
[0124] <Modification> The above-described embodiments and examples are merely examples for explaining the technology of the present disclosure, and are not intended to limit the technology of the present disclosure to only those embodiments and examples. The technology of the present disclosure can be modified in various ways without departing from the gist of the technology. [Explanation of symbols]
[0125] 1 System 10 earphone sets 30, 50 Information processing device 100 Earphones 104 Nozzle 106 Ear tips (elastic electrodes) 310, 510 processors 312 Training Control Unit 313, 513 Acquisition Department 314, 514 Problem Processing Unit 315, 515 Learning Department 316, 516 Training Department 317, 517 Output section 310, 510 processors 330, 530 Memory 512 Application control unit 550 User Interface
Claims
1. A processor included in the information processing device performing a process of having a user select at least one option from a plurality of options regarding the identification of a judgment orientation with respect to an expectation value; acquiring electroencephalogram information measured by an electroencephalogram measuring device worn by the user; based on the electroencephalogram information acquired during the selection process of the option, learning a first brain activity when risk aversion or loss aversion with respect to the expected value is not performed and a second brain activity when the risk aversion or loss aversion is performed, and generating a learning model for estimating the user's brain activity; inputting the electroencephalogram information acquired during a training process for the user regarding expected value judgment into the learning model, and performing a neurofeedback process for the training based on the estimation result of the brain activity; A program that executes the following.
2. The program according to claim 1 , further causing the processor to execute the step of outputting information on the results of the training in the neurofeedback processing.
3. identifying tendencies of the user based on choices selected by the user; The program of claim 1 , further comprising causing the processor to output trend information regarding the trend to the user.
4. obtaining an indication of the user regarding the trend; The performing of the neurofeedback processing includes:
4. The program according to claim 3, further comprising changing training content of the neurofeedback processing in response to the instruction information.
5. If the instruction information includes an instruction to weaken the risk aversion or loss aversion tendency, The program according to claim 4 , wherein the training includes a process of inducing electroencephalogram information during a training process to become the first brain activity.
6. 6. The program according to claim 1, further comprising causing the processor to execute, after the neurofeedback processing, a process of having the user select an option that uses an expected value.
7. 7. The program according to claim 6, further causing the processor to repeat the neurofeedback processing and the processing of having the user select an option after the neurofeedback processing until a termination condition is satisfied.
8. A processor included in the information processing device performing a process of having a user select at least one option from a plurality of options regarding the identification of a judgment orientation with respect to an expectation value; acquiring electroencephalogram information measured by an electroencephalogram measuring device worn by the user; based on the electroencephalogram information acquired during the selection process of the option, learning a first brain activity when risk aversion or loss aversion with respect to the expected value is not performed and a second brain activity when the risk aversion or loss aversion is performed, and generating a learning model for estimating the user's brain activity; inputting the electroencephalogram information acquired during a training process for the user regarding expected value judgment into the learning model, and performing a neurofeedback process for the training based on the estimation result of the brain activity; An information processing method for performing the above.
9. An information processing device including a processor, The processor, performing a process of having a user select at least one option from a plurality of options regarding the identification of a judgment orientation with respect to an expectation value; acquiring electroencephalogram information measured by an electroencephalogram measuring device worn by the user; based on the electroencephalogram information acquired during the selection process of the option, learning a first brain activity when risk aversion or loss aversion with respect to the expected value is not performed and a second brain activity when the risk aversion or loss aversion is performed, and generating a learning model for estimating the user's brain activity; inputting the electroencephalogram information acquired during a training process for the user regarding expected value judgment into the learning model, and performing a neurofeedback process for the training based on the estimation result of the brain activity; An information processing device that executes the above.
Citation Information
Patent Citations
Decision support system and decision support method
JP2021117781A
