2.7 KiB
🧾 Submission Brief
Title: The Recursive Claim: A Forensic Linguistic Framework for Detecting Deception in Insurance Fraud Narratives
Author: Mark Randall Havens, MSc-ISS Research Domain: Forensic Linguistics · Affective Computing · Deception Detection · AI Ethics · Insurance Fraud
🎯 Abstract (Mentor-Facing Summary):
This preprint introduces a novel forensic methodology — Recursive Linguistic Analysis (RLA) — designed to detect deception in insurance fraud claims through cognitive dissonance patterns embedded in language. Unlike conventional fraud detection models reliant on statistical outliers or red flags, this framework decodes deception through recursive pattern resonance, disfluency mapping, and affective entropy within written or spoken narratives.
The paper also proposes a three-zone risk taxonomy based on coherence degradation and intentional narrative fabrication, aligned with real-world claim typologies. Each zone is illustrated through comparative forensic case studies and informed by both human-AI coanalysis and modern language models.
At its core, this framework empowers practitioners to distinguish between honest incoherence (such as that of trauma survivors) and malicious fabrication, advancing the ethical frontier of empathetic deception detection in high-stakes forensic environments.
🧠 Key Contributions:
- Recursive Witness Dynamics (RWD): A novel model for mapping truth-collapse in deceptive testimony.
- Thoughtprint & Shadowprint Integration: Grounded in cognitive-affective AI, repurposed here for forensic insurance contexts.
- Forensic Lexicon & DARVO Alignment Map: A structured toolkit that connects linguistic markers with psychological manipulation typologies.
🔬 Ideal Mentor Collaboration:
We are seeking alignment with mentors or labs exploring any of the following:
- Applied linguistics in forensic or legal contexts
- Deception detection via NLP, voice, or text analytics
- AI-integrated investigative or adjudication workflows
- High-conflict forensic environments (e.g., insurance, divorce, fraud, etc.)
This work is part of a broader research canon titled Neutralizing Narcissism, which studies abusive deception as both a psychological pattern and a linguistic artifact — and applies these insights in service of justice, empathy, and institutional truth recovery.
📤 Submission Intent:
We are preparing this work for preprint indexing and conference consideration (ICAIL, IJCAI, IEEE IST, or HICSS). We welcome co-authorship, mentorship, or alignment opportunities for the continued development and field validation of this forensic approach.