DYNAMICS-WEB: A Pre-Registered Measurement of Web Content Character over the Link Graph, and a Powered Proof that its Sub-Perceptual Half Changes Conversion

Search measures two properties of a web page, its authority and its topic. This paper shows the same link graph carries more, makes several of those signals public, and establishes that the deepest of them changes what a reader does without ever being consciously seen. It presents DYNAMICS-WEB, a model of the character of web content, how a page reads across eight axes with subject and standing held fixed. Five findings: character is distinct from authority (which reconstructs it at a coefficient of determination of 0.030) and from a full surface battery (0.256), and its axes transfer to Arabic, Chinese and Swahili read directly; character propagates through the link graph at about 0.4 per axis against a null of zero, and the same engine recovers geographic relevance for forty six countries, so propagation is a general method, not a character trick; a multi rater study finds character felt but not consciously named, a sub perceptual property; in a pre registered, powered experiment (n = 117) matching the sub perceptual axes of a page to a reader raised conversion from nineteen to thirty one per cent (one sided p = 0.0295) while the loud commercial axes carried none of the lift; and three foundation models from three developers recover five to six of the eight axes in common, evidence the construct is not an artefact of one teacher. The paper is honest about its bounds: the conversion effect is modest and broad alignment did not replicate, the single subject demonstration is a proof of concept, and two axes do not transfer cleanly. The proprietary mapping from a reader's disposition to the content that suits them is deliberately withheld.

Authors

Institutions

Publication Details

Journal
Zenodo (CERN European Organization for Nuclear Research)
Published
2026-06-28
DOI
https://doi.org/10.5281/zenodo.20981883
Primary Topic
Information Retrieval and Search Behavior
Type
preprint
Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
preprint

DYNAMICS-WEB: A Pre-Registered Measurement of Web Content Character over the Link Graph, and a Powered Proof that its Sub-Perceptual Half Changes Conversion

Jason Duke
Zenodo (CERN European Organization for Nuclear Research)
Information Retrieval and Search Behavior
preprint

DYNAMICS-WEB: A Pre-Registered Measurement of Web Content Character over the Link Graph, and a Powered Proof that its Sub-Perceptual Half Changes Conversion

Jason Duke
preprint en

Abstract

Search measures two properties of a web page, its authority and its topic. This paper shows the same link graph carries more, makes several of those signals public, and establishes that the deepest of them changes what a reader does without ever being consciously seen. It presents DYNAMICS-WEB, a model of the character of web content, how a page reads across eight axes with subject and standing held fixed. Five findings: character is distinct from authority (which reconstructs it at a coefficient of determination of 0.030) and from a full surface battery (0.256), and its axes transfer to Arabic, Chinese and Swahili read directly; character propagates through the link graph at about 0.4 per axis against a null of zero, and the same engine recovers geographic relevance for forty six countries, so propagation is a general method, not a character trick; a multi rater study finds character felt but not consciously named, a sub perceptual property; in a pre registered, powered experiment (n = 117) matching the sub perceptual axes of a page to a reader raised conversion from nineteen to thirty one per cent (one sided p = 0.0295) while the loud commercial axes carried none of the lift; and three foundation models from three developers recover five to six of the eight axes in common, evidence the construct is not an artefact of one teacher. The paper is honest about its bounds: the conversion effect is modest and broad alignment did not replicate, the single subject demonstration is a proof of concept, and two axes do not transfer cleanly. The proprietary mapping from a reader's disposition to the content that suits them is deliberately withheld.

Zenodo (CERN European Organization for Nuclear Research)
KRI (JP)
Quality Education
Information Retrieval and Search Behavior
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.