Recursive Engagement After Disclosure A Longitudinal Human Control Case Under Semantic Austerity Conditions (Single Image Artifact)

This dataset consists of a single image documenting a post-publication continuation of engagement behavior by the same human participant analyzed in The Engagement Trap: How Humans and AI React to Online Boredom. The image captures a YouTube comment made after the participant’s behavior had already been explicitly framed, archived, and DOI-registered as part of a digital ethnographic study. Despite this disclosure, the participant continues engagement through ironic dismissal and explicit metric enumeration, reinforcing the original findings. This artifact functions as a within-subject longitudinal human control case under Semantic Austerity Testing (SAT) conditions. It demonstrates that awareness of algorithmic capture and research framing does not reliably restore disengagement or silence. Instead, engagement persists recursively, even when its futility and instrumental absorption are acknowledged. By isolating this behavior into a single, timestamped visual artifact, the dataset preserves temporal clarity and evidentiary restraint while strengthening the comparative parallel between human participants and large language models subjected to the AI Boredom Test.