Domain Authority Gates Citation Eligibility: Entity Density as a Within-Domain Signal in AI Search

Domain authority gates citation eligibility; entity density operates within that gate. No published study has directly measured the relationship between named entity density and AI citation rate using a within-domain competition design. We ran a two-phase citation protocol — 30 general queries and 22 competition queries with pre-registered plausible-answer sets — on 10 pages from a single domain (thegeolab.net) spanning 12.9–29.5 unique named entities per 1,000 words, using Perplexity sonar-pro with web search enabled. The domain achieved 20% citation rate on proprietary concept queries and 0% on all 22 competition queries (zero citations across all 10 pages regardless of entity density). We interpret this as a domain authority threshold effect: below the competitive citation gate, page-level content signals including entity density are not operative. The refined hypothesis — that entity density differentiates citation probability within an authority tier, among citation-eligible pages on the same domain — generates a falsifiable prediction testable by a researcher with access to a citation-eligible domain, and we provide a complete replication protocol.