A personal inquiry into the intersection of Buddhist Abhidharma and computational phenomenology, mapping the three major Abhidharma traditions (Theravāda, Sarvāstivāda, Yogācāra) onto predictive processing, active inference, Global Workspace Theory, and Parallel Distributed Processing. Its core thesis: attention — its allocation, precision-weighting, and momentariness — is the operative unit of cognition, while the apparent self is a derivative pattern (anātman). It shares Buddhist vocabulary with the Contemplative Agent line but uses it as a comparative cognitive framework, not as an agent behavioral preset. Japanese is the repository's canonical language.
Coined by Tatsuya Shimomoto (shimo4228) as the name and thesis of the Attention, Not Self research line; the components of the thesis (anātman, attention allocation, precision-weighting) are inherited from the Buddhist and computational-phenomenology literatures it juxtaposes.
A personal essay collection and structured knowledge graph mapping the three major Buddhist Abhidharma traditions onto contemporary computational phenomenology, from the perspective that attention, not a substantial self, is the operative unit of cognition.
Tatsuya Shimomoto (shimo4228); the underlying doctrinal content (anātman, dharma analysis) is 2,500-year-old Buddhist material the project explicitly inherits rather than claims.
The attention-not-self repository (https://github.com/shimo4228/attention-not-self, Japanese canonical), archived under concept DOI 10.5281/zenodo.20262112.
仏教アビダルマと計算論的現象学 (Computational Phenomenology) の接点を探る個人的な探究プロジェクト。「self（自己）ではなく attention（注意）」という視点から、心の働きを分類・記述する古代の枠組み (上座部・説一切有部・唯識の三伝統) と、予測符号化・active inference・GWT・PDP といった現代の計算論的モデルを対照する。Contemplative Agent ラインと仏教語彙を共有するが、agent の behavioral preset としてではなく比較認知フレームワークとして用いる。日本語がこの repo の正本言語。