THE EXPLANATORY FIDELITY-ACCESSIBILITY TENSION THEORY (EFATT): A Conditional Framework for the Optimization of Scientific Explanation, Representational Complexity, and Audience-Indexed Comprehensibility
Year: 2026
Type: preprint
Abstract: **Abstract**Scientific explanation can involve a potential tension between the faithful representation of a subject and the accessibility of that representation to a specified audience. Greater analytical detail may expose additional assumptions, dependencies, boundary conditions, mechanisms, uncertainties, and mathematical relations. However, a more detailed presentation may also increase the effort required to follow the explanation. Conversely, simplification can reduce unnecessary representational burden, but it may also omit conditions or mechanisms essential to the intended interpretation. Neither outcome is inevitable: appropriate sequencing, alternative representations, and layered presentation may improve both fidelity and accessibility.This paper proposes the Explanatory Fidelity-Accessibility Tension Theory (EFATT), a conditional theoretical framework for analysing these possibilities. Its primary case, the Scientific Explanation Fidelity-Accessibility Dilemma (SEFAD), represents the problem as an audience- and purpose-dependent decision task rather than a universal conflict between simplicity and correctness.The proposed default mechanism is Layered Fidelity-Preserving Explanatory Optimization (LFPEO). It specifies ten procedural operations, including scope definition, construction of an evidence-qualified reference explanation, complexity decomposition, critical-content classification, layered representation, fidelity screening, audience testing, distortion auditing, iterative revision, and documented closure.A preliminary mathematical architecture defines explanatory fidelity as a vector, comprehensibility as an audience-indexed construct, representational complexity as a multidimensional measure, and simplification loss as a distortion vector. The core optimization problem minimizes representational complexity subject to explicit fidelity, comprehensibility, uncertainty, and scope constraints. If no single-layer representation is feasible, the framework requires consideration of layered alternatives or an explicit declaration that the requirements cannot jointly be satisfied.The framework is organized into sixteen architectural layers, eight interdisciplinary research domains, testable propositions, measurement plans, boundary conditions, counterexamples, and a nineteen-step research roadmap. The mathematical expressions and operational measures are proposed specifications rather than validated universal laws. No original empirical data are reported. The framework's contribution must therefore be evaluated through systematic literature review, measurement development, preregistered studies, independent replication, and comparison with alternative models.**Keywords**scientific explanation; explanatory fidelity; accessibility; representational complexity: cognitive load; layered explanation; information compression; operationalization; multi-objective optimization; theory construction
Author: Shan, Shamiul Hoque
Language: English
Subfield: History and Philosophy of Science
Field: Arts and Humanities
Domain: Social Sciences
Open Access status: green
OpenAlex ID: W7225752967
Cites:
Cited by:
Related to: