Distilling complexity: stylized facts as explicit, auditable foundation for ecological theory and modeling
Ecological models routinely rely on broad empirical regularities to constrain theory and validate predictions, yet these regularities are typically applied implicitly under alternative labels such as "patterns", "laws", or "benchmarks." Such implicit applications obscure a model's empirical commitments, hide unverified assumptions, mask scale dependencies, and hinder transfer across systems. Explicitly naming these regularities as "stylized facts" — a formalization with roots in economics and a robust application in ecology through Dynamic Energy Budget (DEB) theory — changes their epistemic status: foundational constraints become searchable, comparable across studies, and directly auditable against empirical support. This review evaluates the epistemological role and methodological utility of explicit stylized facts across ecological modeling. We first survey their widespread but implicit use across subfields. We then use DEB as a focal example to examine how explicit stylized facts drive theoretical development through three modes — expansion, adjustment, and structural change — and how contradictory evidence, rather than invalidating a theory, defines its precise domain of applicability and motivates accountable theoretical pluralism. Auditing empirical support at the scale of the published literature is beyond what any author team can do by hand; what we report instead is a structured authors' confidence assessment covering both DEB's core stylized facts and a set of additional statements formulated by the author panel. Recognizing the bias inherent in any small panel and the sheer volume of the modern literature, we call for an AI-assisted, community-curated system for cataloguing and auditing stylized facts. Several long-standing assumptions already command markedly less confidence than their foundational status implies — and making that visible, and open to revision, is what turns a set of modeling conventions into an auditable empirical foundation.
Authors
- Branimir K. Hackenberger (ORCID: https://orcid.org/0000-0003-4317-2067)
- Lav Bavčević
- Domagoj K. Hackenberger
- Tin Klanjšček (ORCID: https://orcid.org/0000-0002-9341-344X)
- Sunčana Geček (ORCID: https://orcid.org/0000-0001-8258-4895)
- Jasminka Klanjšček (ORCID: https://orcid.org/0000-0001-8329-9260)
- Ines Haberle (ORCID: https://orcid.org/0000-0001-8978-4435)
- Tamara Đerđ
- Jadranka Pečar Ilić
- Ivana Terzić (ORCID: https://orcid.org/0009-0007-9367-5814)
Institutions
- Ruđer Bošković Institute (HR)
- Agricultural Institute Osijek (HR)
- University of Osijek (HR)
- University of Zadar (HR)
Publication Details
- Journal
- Ecological Modelling
- Published
- 2026-10-07
- DOI
- https://doi.org/10.1016/j.ecolmodel.2026.111864
- Primary Topic
- Philosophy and History of Science
- Type
- article
- Field-Weighted Citation Impact
- 0.00