In March of this year, Risk Measurement Technologies Co., Ltd. (hereinafter referred to as “Company”) released a “mind health measurement system” that can visualize mental health easily with only voice. In this column, while Covid-19 spreads, we will examine how the actual data obtained through actual life and consider the factors of the fluctuation.
What can we recognize from our voice?
Chart 1: Mind Activity
The results from April 1, 2020 to August 18, 2020 are shown for the mind activity that shows the state of mind. The higher the number, the better the state of mind. Mind Activity (unit: points) is on the vertical axis and time is on the horizontal axis. The index is a moving average for the last three days of the daily average index. The data used to calculate the numbers are over 4000 observations consisting of over 100 users (*1).
*1 Various users are mixed, such as users who measure every day and users who measure only workdays, making it more practical data.
Chart 2: Number of new positives for Covid-19 in Japan
It shows the number of new positive cases of Covid-19 daily. The data source is “Current situation of new type coronavirus infection and correspondence of Ministry of Health, Labor and Welfare (Ministry of Health, Labor and Welfare)”, and we are totaling (*2)
In response to the spread of Covid-19, a state of emergency was announced to seven prefectures based on the provisions of Article 32, Paragraph 1 of the Special Measures Law. The decrease in the index is considered to reflect the anxiety based on the declaration.
It is considered that the index increased because the number of new positive every day decreased, and the sense of security expanded. On the other hand, the figures have fallen before the state of emergency was lifted. This may be due to a mixed feeling of anxiety about lifting it.
The index has risen, reflecting expectations for the relaxation of travel restrictions across prefectures. However, the index temporarily dropped, probably due to the mixed feeling of anxiety about the relax. After that, the index has remained stable for some time.
In July, the number of new positives gradually increased. In response to this, the optimism and pessimism of the future were mixed up, and the index showed repeated fluctuations. After that, the index decreased as the number of new positives (whole of country) increased to over 1000. On the other hand, when the number of new positives hit the ceiling at the beginning of August, the index turned upward.
What is a relationship with other events?
When calculating this index, we performed a correlation analysis with economic indicators that can be referenced daily or monthly. As a result, within a certain period, a positive (or negative) correlation was confirmed, but the correlation was not continuously confirmed for a long period of time.
It is thought that the reason for this result is that users with various backgrounds measure in various environments. In other words, it can be pointed out that the indicators of each user do not fluctuate in the same direction unless there is a big event that everyone pays attention to (the event that is common to all) because the measurement variations vary among users.
What are the factors that change Mind Activity?
We will not focus on individual reasons for changes in Mind Activity, but in this column, we will focus on the structure of change factors. Based on the above, it can be assumed that the fluctuation factors of mind activity can be separated into two factors: common factors and individual factors. This is formulated as follows. MA means the indicator of Mind Activity.
X represents an event specific to an individual (individual event), and Y represents an event common to all people (common event). α means sensitivity of individual factors, and β means sensitivity of common factors.
- If Mind Activity fluctuates with a common event, we can recognize a big shock to Y. For example, events related to Covid-19.
- The magnitude of sensitivity (α, β) is different for each user. However, for example, users belonging to the same group can think that the sensitivity β of the common factor is relatively larger than α. In other words, users belonging to the same group can be more affected by common events and point out that their mind activity behaves similarly.
- If someone is in the same group but deviate from the trend of the group, we can consider that someone may be more affected by individual factors.
In this column, we used the average index of Mind Activity to analyze the relationship with the events related to Covid-19. Fortunately, this analysis can be done because the events of Covid-19 had a large impact as a common factor. However, it is undeniable that the relationship between the two was a pseudo correlation, and it is necessary to continue observing for a longer period in the future.
It can be considered that the fluctuation factors of Mind Activity can be classified into two factors: individual factors and common factors. In that case, for example, it is possible to consider whether the fluctuation of mind activity of individual users is strongly influenced by external common factors or internal individual factors. The comparative analysis with our data will be useful as a benchmark for further consideration.
Finally, please keep in mind that the analysis in this article is still evolving and content may change in the future. If you have any mistakes in our consideration, we would appreciate it if you could contact us.
Please contact us from the following URL.