Abstract:One partner says "Fine" meaning <i>resolution</i>; the other hears <i>surrender</i>. The word is shared; the affective uptake is not. We formalize this as <b>affective meaning divergence (AMD)</b>, the total-variation distance between interlocutors' anchor-conditioned affect distributions. Building on speech-act theory, common-ground accumulation, and entropy-regularized game theory, we derive a logit best-response map whose dynamics undergo a saddle-node bifurcation: when $\beta\alpha > 4$, a monotone increase in AMD-driven load produces an abrupt, hysteretic collapse of repair coordination. On Conversations Gone Awry (CGA-Wiki; $N=652$), derailing conversations exhibit critical-slowing-down (CSD) signatures across multiple levels: lexical divergence variance ($p<0.001$, $d=0.36$), AMD variance ($p=0.001$, $d=0.26$), and dialog-act repair variance ($p=0.016$, $d=0.20$), all significant after correction and stronger than toxicity and sentiment baselines. AMD provides a distinct temporal signature, with retrospectively measured variance peaking at the bifurcation point while toxicity variance peaks earlier, and is the only indicator grounded in the theoretical framework. Boundary-condition analysis on CGA-CMV ($N=1{,}169$) yields mixed but directionally consistent evidence.
| Comments: | Accepted to the ACL 2026 Student Research Workshop |
| Subjects: | Computation and Language (cs.CL); Artificial Intelligence (cs.AI) |
| Cite as: | arXiv:2605.09043 [cs.CL] |
| (or arXiv:2605.09043v1 [cs.CL] for this version) | |
| https://doi.org/10.48550/arXiv.2605.09043 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Napassorn Litchiowong [view email]
[v1]
Sat, 9 May 2026 16:30:47 UTC (151 KB)