The potential application of AI in categorizing and facilitating Buddhist practice

Abstract

Artificial intelligence (AI) becomes more broadly used in modern society. The development of AI has enabled it to accurately recognize images of ancient Buddhism scripts. In fact, AI techniques such as machine learning has widely been applied in neuroimaging data processing, and there is rapid advance of neuroscience in the past decades, supported by major countries like USA, EU, Japan, and China. Buddhism has enormous scriptures of Sutra Pitaka, Vinaya Pitaka and Abhidharma Pitaka. However, what makes Buddhism distinct from secular philosophies is Buddhism practice, only through which the practitioner can understand the ultimate truth illustrated by all the Buddhism scriptures. Nowadays, neuroscientists have begun the research of Buddhism practice. The purpose is to understand the brain mechanism of Buddhism practice and eventually, to facilitate Buddhism practice. Converging evidence has demonstrated significant changes in brain activities during meditation. A most prominent finding is an increased alpha wave during meditation, although different meditation methods may result in different brain activities. Given the advantage of data analysis by AI, it can be a potential tool to analyze the practitioners’ neurophysiological data and help them in practice. If AI could recognize his/her meditation state, for example, whether there is a wandering mind, how is the meditation improved in a specific time period. So that the practitioner can be more aware of their current mental states, and this may improve the individual’s practice by continuous and accurate feedback. In the Lab of Buddhism and Science, Centre of Buddhist Studies, we used portal electroencephalogram (EEG) device to measure brain electronic activities during meditation in daily living. Specifically, we focused on the difference between left and right frontal activities, that is, the frontal asymmetry during meditation, as compared to the normal resting-state. It has been found that the prefrontal alpha-asymmetry in resting EEG can be a biological indicator of mind state, including the emotional state. We found that meditation of mindfulness breathing may reduce the frontal asymmetry. With big data collected further, the pattern may be learned by the AI, and it becomes possible that the pattern recognized by AI can accurately discriminate whether the individual is meditating properly. We suggest that AI in the long run will be able to facilitate the individual’s meditation practice by monitoring the neurophysiological data of the practitioner.