This working paper proposes PPPM (Parasocial Proxy Punishment Mode) v3.0, a fully reconstructed revision of v2.0 (DOI: 10.5281/zenodo.19567733). v3.0 redefines the theoretical core around two central constructs: spontaneous proxy activation (HIS acts as the performer's proxy without explicit command) and desynchronization (the divergence between the performer's actual intentions and HIS's internal action-line). The central thesis is the paradox of identification: paradoxically, the stronger the identification, the more HIS diverges from the performer's actual wishes, because excessive identification transforms empathy into projection. This version introduces two internal subtypes, formally demarcates PPPM from adjacent phenomena, and identifies the paradoxical consequence in which protective behavior harms the object of protection. AI-assisted writing (Claude, ChatGPT) was used for drafting; observation, conceptualization, and theoretical judgment are the author's own.
TekitouQ (2026) studied this question.