Cross-Disciplinary Taxonomy and Modeling of Misunderstanding Generation, Amplification, and Detection, from Pragmatics to AI Agents
06:00 · August 17, 2026 · arXiv cs.AI RSS
Detection of misunderstanding is an urgent problem to solve because communication has moved away from real-time, in-person interaction and is increasingly handled by AI-mediated channels. This shift cuts communicators off from the resources repair depends on faster than new means of detection are being built. In this paper we analyse misunderstanding as a layered process in which a divergence is generated, may then be amplified, and is either detected and repaired or left to persist unnoticed. Consolidating accounts from nine fields of research that do not ordinarily cite one another, we identify eleven exact failure modes and show that each operates at a specific point in a communicative process rather than anywhere within it. Those points give eight analytical layers, derived from the literature rather than adopted from an existing model. Eight of the mechanisms primarily generate a divergence, two primarily amplify one already present, and one governs whether a divergence is detected and repaired. We model the eight layers formally, extending information and communication theory from the transmission of signals to the reconstruction of meaning, and we supply a source-by-source evidence matrix that makes every rating auditable, a coding manual, and nine analysed dialogue cases. No prior classification of misunderstanding both locates mechanisms at points in the process and types them by function.
Summary
Misunderstanding in communication has become harder to detect and repair as interactions shift from face-to-face exchanges to AI-mediated channels that remove immediate feedback cues. The paper frames misunderstanding as a sequential process in which a divergence between intended and interpreted meaning is first generated, may then be amplified, and is either noticed and corrected or allowed to persist. Drawing on research from nine fields that rarely intersect, the authors isolate eleven specific failure modes and locate each at a distinct stage of the communicative sequence rather than treating them as interchangeable.
These stages coalesce into eight analytical layers derived directly from the source literature. Eight of the identified mechanisms primarily create an initial divergence, two act to magnify an existing one, and one controls whether the divergence is detected and repaired. The authors formalize the eight layers by extending classical information and communication theory beyond signal transmission to encompass the reconstruction of meaning across successive interpretive steps.
To support application and verification, the work includes an auditable source-by-source evidence matrix, a coding manual for applying the taxonomy, and nine fully analyzed dialogue cases. The resulting classification is the first to combine precise functional typing of each mechanism with its location in the overall process, offering a structured basis for both detection systems and further modeling in AI-mediated settings.
Why it matters
This research is highly relevant for Dutch AI researchers and developers working on conversational AI and NLP. Its formal modeling of misunderstandings provides actionable frameworks to improve the reliability and transparency of AI agents, aligning with the EU's strong emphasis on trustworthy AI.
