Attraction Formula- A Computational Dynamics Model for Narratology(V4.0)

English This is Version 4.0 of the Attraction Formula theoretical framework, constituting a major paradigm upgrade from Version 3.0 (April 2026). While V3.0 established the Unified Attraction Formula (E = 1/2 · m · v² · s · ∏C), Actual vs. Virtual State Motion, Chain-Web Topology, Two-Stage Selective Focusing, and Social-Domain Resonance as a rigorous computational dynamics model for narratology, V4.0 advances the paradigm by streamlining the coefficient architecture, introducing temporal and directional parameters, and formally integrating superpositional closure and dual-layer valuation into the theoretical corpus. Key theoretical upgrades in V4.0 include: (1) Unified Attraction Formula V4.0: E = 1/2 · m · (s/t)² · d · ∏(9C) · A_focusing, replacing the previous 10-coefficient multiplicative architecture with a streamlined 9-coefficient system plus the system-level A_focusing modulation layer; introducing narrative time-speed (t) and direction (d) parameters; and formally bifurcating bare kinetic energy (E_bare) from experienced attraction (E_experience); (2) Superpositional Closure, modeling multi-thread narrative attraction through independent source-event bookkeeping, separate repair trajectories, and cross-node resonance decay (C_resonance), enabling quantitative accumulation of psychological kinetic energy across parallel mother state-lines; (3) Dual-Layer State-Value Scale, unifying all demand-states under a subjective-satisfaction metric across the narrative-fact layer (objective story-world states) and the audience-experience layer (subjective satisfaction assessments), establishing the prerequisite for cross-state superpositional calculation; (4) Two-Stage Selective Focusing V4.0, upgrading the attention architecture by retiring C_attention as a multiplicative coefficient and reconstituting it as the system-level A_focusing weight output by Stage-2 Softmax competition, functionally absorbing and replacing the retired First-State Contention mechanism; and (5) Actual vs. Virtual State Motion and Social-Domain Resonance (retained and refined), governing reality-modulated energy release and collective audience amplification, now fully integrated with the superpositional closure and dual-layer valuation systems. V4.0 thereby transforms narrative attraction analysis from qualitative interpretation to quantitative dynamical-systems computation with enhanced multi-thread accumulation capacity, with direct applicability to film, literature, interactive media, music, painting, and generative storytelling. License: CC BY-NC-ND 4.0 中文 本文是"吸引力公式"理论框架的第 4.0 版,是从第三版(2026年4月)的重大范式升级。V3.0已将吸引力统一公式(E = 1/2 · m · v² · s · ∏C)、实态动与虚态动、链网拓扑、两阶段选择性聚焦及社会域共振确立为计算叙事学的严谨动力学模型;V4.0则在此基础上精简系数架构、引入时序与方向参数,并将叠加闭合与双层值位评估正式纳入理论体系。 V4.0的核心理论升级包括: (1) 吸引力统一公式V4.0:E = 1/2 · m · (s/t)² · d · ∏(9C) · A_聚焦,以精简的9系数体系加系统级A_聚焦调制层取代原有的10系数乘性架构;新增叙事时间速度参数t与方向参数d;并正式区分事态级裸吸引力与体验级最终吸引力; (2) 叠加闭合机制,通过独立源事态记账、分别修复轨迹与跨节点回响衰减(C_回响)建模多线叙事吸引力,实现跨平行母态线的心理动能定量累积; (3) 双层值位量表,以叙事事实层(故事世界客观状态)与观众体验层(主观满足评估)的统一主观满足量纲归约全部需求态,确立跨态叠加计算的前提; (4) 两阶段选择性聚焦V4.0,将原C_关注从乘性系数中移除,重构为第二阶段Softmax竞争输出的系统级A_聚焦权重,功能上吸收并替代已废止的争首态机制;以及 (5) 实态动与虚态动及社会域共振(保留并深化),统摄真实度调制的能量释放与集体放大效应,现与叠加闭合及双层值位体系完整整合。 V4.0由此将叙事吸引力分析从定性解释转变为具备多线累积能力的定量动力学系统计算,可直接应用于电影、文学、交互媒介、音乐、绘画及生成式故事创作。 许可协议:CC BY-NC-ND 4.0