What question did this study set out to answer?

This research aims to explore a framework for evaluating cognitive manipulation risk in textual content.

May 30, 2026Open Access

R8: An Exploratory Lexical Framework for Approximating Cognitive Manipulation Risk in Text

Key Points

This research aims to explore a framework for evaluating cognitive manipulation risk in textual content.
Utilized System 1 cognitive processing framing for analysis.
Revised performance metrics and inter-rater reliability values.
Included markdown files for reproducibility.
Recall achieved was 35.2%, indicating moderate retrieval effectiveness.
F1 score was 51.8, showing a balance between precision and recall.
Inter-rater reliability κ values were 0.231 for Claude Sonnet and 0.103 for Gemini 2.5 Flash, suggesting low agreement.

Abstract

Version 1.8 of the R8 preprint. Changes from v1.7: (1) title updated to reflect exploratory scope; (2) Abstract shortened with System 1 cognitive processing framing added; (3) APA 7th edition reference formatting applied throughout; (4) display math for CMI formula and sigmoid equation; (5) corrected performance metrics (Recall=35.2%, F1=51.8) and inter-rater reliability values (Claude Sonnet 4.6: κ=0.231; Gemini 2.5 Flash: κ=0.103); (6) Japanese translation added. Source markdown files included for reproducibility.

Read Full Paperexternally

Mark Helpful

Bookmark

Relay

View Full Paper