Answers are grounded in this paper and cite their sources.
This document preregisters six predictions of steretic logic for the growth of the concept network of the biomedical literature (MEDLINE, MeSH major-topic co-occurrence), before any MEDLINE data were downloaded or inspected. A missing link between two concepts is a void when the concepts share at least two common neighbours. Beyond the known fact that common neighbours predict new links (P0, a sanity check), the predictions concern the role of activity and dormancy in whether a void is filled (P1), the inability of a single mediator to join two parts of the network (P2), a distance threshold for absorption into dense cores (P3), protection against homogenisation by dormant concepts (P4), whether dormant mediators still carry a void between active concepts (P5), and the life cycle of the space of voids of a research direction (P6). Data construction, thresholds, models and decision criteria (confirmed, refuted, inconclusive) are fixed in advance; a held-out replication year is analysed only after the main results are written up. All verdicts will be published whatever they turn out to be.
No takes yet. Share an insight, caveat, or question.
Oleksandr Krol (2026) studied this question.
Synapse has enriched 8 closely related papers on similar clinical questions. Consider them for comparative context: