ARGIRA VIII: Invariant Survival Analysis of Visual–Acoustic Associations Across Competing Sonification Operators

ARGIRA VIII introduces the Invariant Survival Matrix (ISM), a framework for evaluating whether visual–acoustic associations persist across competing sonification operators. Using a corpus of 117 canonical paintings, the study compares a historical operator (M1), a semantic perturbation operator (M2), and an architectural reorganisation operator (M3/M3b). Rather than evaluating correlations alone, associations are assessed according to both their associative strength and their survival under systematic operator perturbation.

The study establishes a methodological shift from correlation analysis to invariance testing. Visual–acoustic relationships are evaluated under changes in mathematical form and architectural assignment, allowing the distinction between associations that are robust and those that depend on specific operator design choices.

The results show that most strong associations do not survive architectural reorganisation and are therefore dependent on operator structure rather than intrinsic properties of the visual corpus. A complete architectural coverage control (M3b) produces a direct falsation of the strongest association reported in ARGIRA VII (hue_std → odd_bias), demonstrating that the relationship is an operator artefact rather than an operator-invariant coupling.

A central finding is that architectural organisation has a substantially greater impact than mathematical formulation. Across the complete matrix, no associations are classified as dependent on mathematical form alone (dep_forma = 0), whereas twenty-one associations are classified as architecture-dependent (dep_arquitectura = 21).

Only a small subset of associations remains stable across all tested perturbations. These surviving relationships are dominated by chromatic–tempo couplings, which emerge as the principal candidates for operator-invariant structure within the ARGIRA framework.

ARGIRA VIII establishes that:

• An invariance test can be applied to visual–acoustic mappings through systematic operator perturbation.

• Correlations alone are insufficient for validation; survival under perturbation must also be evaluated.

• Most strong associations fail under architectural reorganisation and are therefore operator-dependent.

• The strongest association reported in ARGIRA VII (hue_std → odd_bias) is falsified under complete architectural reassignment and reclassified as an operator artefact.

• Architectural organisation exerts a substantially greater influence on outcomes than mathematical formulation.

• A reduced subset of associations survives all perturbations and remains compatible with operator-invariant structure, particularly within chromatic–tempo relationships.

By introducing falsation-oriented survival analysis, ARGIRA VIII reframes sonification validation from the discovery of correlations to the evaluation of their persistence across competing operator architectures. The framework provides a reproducible methodology for distinguishing robust visual–acoustic associations from effects generated by operator design.