© Serhii Radachynskyi
Junya Morita, an associate professor at Shizuoka, analyzes the mechanisms of negative collective behaviors in the online society and the technology that allows them to combat them.
While web technology has drastically changed our lives and our society, the emotional problems that were once rare have increased in number and severity. These problems are becoming more serious, sometimes inducing large-scale negative collective behaviors such as flame, cyberbullying, and cyberattack. Although the social and legal systems governing these issues have been actively debated, interventions based on understanding cognitive and emotional mechanisms continue to be severely underdeveloped.
With regard to these mechanisms, researchers have noted that the negative side effects of technology are emphasized and that they are metaphorically as severe as drug addiction, creating repetitive overuse. This behavioral addiction can be considered especially problematic if users have symptoms of mental illness. Ruminating, for example, is a psychological state related to depressive mood that involves repetitive negative thinking about a particular topic. Technologically enhanced and socially accumulated ruminant thinking could cause a serious disaster to our society, with synergistic effects through the echo chamber mechanism.
From a more microscopic point of view, the combination of such a negative mental state with information technology causes severe negative symptoms through a feedback loop. Information technology removes the boundaries of cognitive boundaries that had evolved throughout human history. Humans naturally forget information that is not relevant to their current task and, because of this mental function, can focus on the situations they encounter. However, the web environment provides decadent information desired by users without time constraints. This comfort induces addictive behavior and people cannot resist using it even when they are aware of the irrationality of their own behavior.
Interventions based on the simulation of mental processes
Considering that the emotional problems caused by the network are incompatible with the nature of human memory, Morita and her colleagues participated in a research project within the framework of the Program for establishing topics to advance the search for avant-garde humanities and social sciences that would respond to the real society administered by JSPS. In their study, they constructed a system that combines cognitive modeling of human memory with information indications regulated by physiological detection.
Image request based on behavior log
Web ads such as behavioral targeting can be used to intervene naturally in ruminant behavior. This type of online intervention is supposed to cause behavioral changes that are conducive to psychological health. Thus, the system collects user experiences in natural web browsing activities. When the user visits shopping sites, the system automatically collects the product images that are displayed on the browser screen. The collected images become a knowledge structure that leverages images with semantic attributes that are extracted based on deep learning technology. The constructed knowledge structure is assumed as a semantic network that allows the system to retrieve images successively according to the semantic association. The browser extension program embedded in the web environment of users presents recovery sequences on the browser screen as forms of web advertising.
Model of human memory and emotion
The browser image sequence is regulated with a model that simulates human memory, which in turn is regulated by the mechanism of activation of memory elements, which consist mainly of recent and frequency effects. Therefore, recently corrected and frequently viewed images are more likely to be displayed on the screen. Only with these effects does the model naturally converge on a specific memory element, causing repetitive visualizations on the browser screen. To escape this repetitive memory recall, probabilistic noise is added to the activation. If the noise fluctuates widely, it increases the chance of escaping this feedback loop. As a mechanism for modulating noise fluctuation, noise variance is mapped with norepinephrine activation. That is, when the model is in a stressful situation, the output of the model converges; meanwhile, in a relaxed situation, the model produces various behaviors.
Mirror / counterweight of heart rate variability
Based on the previous model, Morita and colleagues developed two types of real-time parameter modulations: replica models and counterweights. In the replica model, noise variance is replaced by the physiological state of the user, which is obtained directly from a heart rate sensor. Memory noise based on the heart rate variability (HRV) of the current user leads to a large fluctuation in image recovery when the user is relaxed, because HRV increases in the state of parasympathetic dominance . In other words, when the user feels stressed and anxious, the activation value always generates the same images as with high frequency and frequency effects. Therefore, this replica model where the noise size corresponds to HRV synchronizes the ruminant behavior with the participants. In contrast, the counterweight model modulates the size of the noise as the inverse of the user’s HRV, thus behaving as a correction of ruminant behavior. This counterweight model plays a role in a technologically enhanced homeostasis system that regulates user memory through web advertising.
Evidence and implications
The developed system was tested with a laboratory experiment where participants were first induced into a negative mood. By observing the recovery process in the subsequent web search task, the effect of the counterweight approach on users coming out of the negative mood was observed. Users who were presented with a sequence of counterweight images reported that their moods were less negative after the task than those presented with the replica sequence. From this result, the approach taken here can be seen as a possible candidate for achieving harmony between natural and artificial cognitive systems, thus balancing emotional issues in this digital age.
This work has been carried out in collaboration with Mr. Thanakit Pitakchokchai, el Sr. Giri Basanta Raj, Dr. Yusuke Yamamoto, Prof. Hiroyasu Yuhashi and Prof. Teppei Koguchi of Shizuoka University as part of the Theme Allocation Program to advance humanities and cutting-edge humanities. Social science research administered by JSPS.
Please note that this is a business profile
© 2019. This work is licensed under a CC BY 4.0 license.