Aiming to alleviate anxieties felt in the school setting

The outbreak of COVID-19 significantly disrupted the education field. Amidst demand for extended school closures and other previously unheard-of measures,*¹ the question of how to provide ongoing education for children while ensuring their safety became a major issue for schools everywhere.

^{*1 The Ministry of Education, Culture, Sports, Science and Technology; “Regarding the Temporary Closing of All Elementary, Middle, High and Special Needs Schools in an Effort to Combat the Novel Coronavirus, February 28, 2020”}

Addressing risk-related worries associated with the school setting, the project team chose elementary schools as their next research target as they are host to many active children and infection transmission is high.

“The risk of coming in contact with pathogens is much higher in high-traffic public spaces. This was our thinking when shifting our research from the home to public spaces. Research on how infectious diseases spread within a school setting was fairly limited worldwide, which is why we considered this an important area to look into.” ―Project Leader Mr. Takizawa

“When we were conducting our research, we visited an elementary school*² a number of times. I remember thinking how sad I felt and how badly I wanted the threat of the virus to be gone each time I watched the children silently eating their lunch while facing the front of the classroom. What is an appropriate infection control policy for school settings? I aimed to find an answer that would not reduce students’ ability to socialize with friends or hamper their ability to learn, while ensuring their safety at the same time. I truly believed that pursuing this research would be very important.” ―Researcher Mr. Nakajima.

^{*2 During the course of research into children’s behavioral data in February 2022}

We constructed a simulation model in an elementary school setting and analyzed how the virus spread through contact between children. This will allow us to share accurate information on hygiene habits that effectively control the transmission of infectious diseases. Furthermore, insights gained from this simulation model will open the door to similar research in other public spaces.

Information obtained from the behavior of children

The creation of a simulation model that involves social group behavior differs from similar research done for the home as it requires considering group dynamics and how multiple children can simultaneously spread the virus. In short, in addition to data on the children’s behavior, we needed an understanding of the group’s social dynamics. However, it is difficult for children to report on their relationships with friends, and even harder to record their own actions and things they have touched. Therefore, theorizing that the risk of infection was highest during break times, researchers decided to let the children move about freely while setting up cameras in the classrooms and hallways to record behavior during those times. This is how the simulation was set up to produce authentic data.*³

^{*3 We received the cooperation of a school in Tokyo to film classrooms and hallways frequented by fourth-grade students. Prior to carrying out the research, the details of the process and assurances as to how personal information would be handled were presented to the Board of Education and the teachers as well as to the children themselves and their guardians, and permission was obtained from all parties concerned. We made sure to modify the footage so that individuals could not be identified and to delete all footage involving children whose guardians did not consent to their participation so that it could not be used for research.}

The information from the footage falls into two key categories: objects the children touched and their interactions with each other. To incorporate this data into the simulation, we converted it into quantifiable values. The diagram below shows the objects that the 30 children touched and how many times these objects were touched during a five-minute break. Children touched not only their own belongings but also those of others, including their classmates’ desks and clothing as well as shared items.*⁴

^{*4 Keisuke Nakajima, et al. “Proposal for a Simulation Model to Visualize the Risk of Contact Virus for Students in Elementary Schools,” The 37th Annual Conference of the Japanese Society for Artificial Intelligence, 2023.}

We additionally totaled and quantified the communication between children in terms of who was interacting with whom as well as the nature, frequency and timing of such interaction. The illustration below shows examples of interactions. We also found that communication styles vary, and while some children communicate with many other children, there are also those who communicate with only a limited number.

“Our study predicted that when 30 children are allowed to move around freely, the virus will spread more than originally expected. We also realized that the virus could even spread to students who didn’t interact with their infected classmates as long as they were in the same room together.” ―Researcher Mr. Kikuchi

We collaborated closely with Professor Setsuya Kurahashi*⁵ in the examination of this data. Over a short period of time, we repeatedly analyzed and quantified the behavioral data, formulated hypotheses, analyzed and quantified additional data, and incorporated it all into the simulation for verification. By doing so, we continued to refine the social group simulation model.

^{*5 Affiliation at time of development: Professor of Business Sciences, University of Tsukuba}

“We aimed to publish the results of our research during the period when the virus was still continuing to spread. Although it was difficult to refine the data and construct the simulation model within this limited time frame, it was very interesting and rewarding to see characteristics and tendencies inherent to children become apparent through the analysis of this important data.” ―Researcher Mr. Kikuchi

Using the constructed simulation model in an elementary school, we evaluated how effective hygiene habits such as handwashing were at limiting the spread of the virus. Specifically, by evaluating the number of children who came into contact with the hand of an infected classmate during one of their five-minute break times, we verified whether hygiene habits could affect viral spread.

Our results found that the total number of children who came into contact with the virus decreased when the infected children washed or disinfected their hands immediately after virus exposure. 

However, children touching the classroom door when they go to wash their hands conversely led to the virus spreading more. To summarize, we believe a strategy where the children don’t need to touch the classroom door and are able to quickly wash or disinfect their hands after exposure to the virus would be most effective in limiting its spread.

Connecting this research to effective preventive measures that do not negatively impact educational opportunities

Researcher Mr. Kato conducted a handwashing class and showed the results of our research to the children who participated in the study. His words regarding the impact this research will have on children are as follows.

“Everyone listened with great interest. I think that they probably viewed it as directly related to themselves since the data was collected from them. With the help of the information we acquired through the simulation model we constructed using the children’s behavioral data, we were able to shine a light on hygiene habits. I feel that this research was very valuable as it allowed children to reflect on their actions in everyday lives.”

Stopping the spread of infectious diseases through handwashing alone is difficult. Therefore, it is important to take a variety of different preventive measures such as the frequent sanitization of high-touch items. This research will act as an aid in proposing such measures.

“Using this simulation model and visualizing behaviors we previously only vaguely understood, we can now answer the question of how to act. When we use this simulation model in an elementary school, we are able to visualize not only individual risk of infection, but also community risk. Going forward, we plan to share our results with many more elementary schools. We are also further considering how this discovery of the importance of hygienic habits can be applied to other public spaces and promoted overseas.” ―Researcher Mr. Takizawa

By highlighting invisible risks in this way, we are able to create policies and hygiene havits that properly combat infection. Even before the COVID-19 pandemic, infectious diseases frequently spread in elementary schools. To minimize the spread of infection without reducing educational and social interactions, the accumulation of research like this will become increasingly more important.

Contents, and affiliations are as of the time of the interview. (Interviewed Jan 2024)

【PICK UP】 Comments from Dr. Setsuya Kurahashi^*6, Collaborative Researcher and Professor in the Department of Business Sciences, University of Tsukuba

Through this research, we have developed a technique for predicting the risk of virus transmission using the simulation model while investigating and understanding the subconscious behaviors of consumers. I believe this will be very important in preparing against future infectious diseases. We hope to be steadfast in continuing our research and proposing hygiene habits that will prevent the spread of infection across all of society.
*6 Affiliations as of the time of development

＜Professor Kurahashi’s Background＞
Professor of Business Sciences. Engaged in research in the following: social communication, business information analysis, gaming, evolutionary machine learning and abnormality diagnosis, etc.