MIX-45: Ultra‑High‑Rigor Protocol for Electronic Voice Phenomenon (EVP)

Protocol MIX-45 implements the unified metrological framework of Protocol 9.0 to test the claimed phenomenon of Electronic Voice Phenomenon (EVP) – the alleged appearance of intelligible voices or messages in audio recordings that were not audible at the time of recording, often interpreted as communication from discarnate entities. The primary observable is the agreement between independent blind judges on the presence and content of anomalous voice‑like signals in recordings made under controlled conditions, compared to control recordings (silence, pure noise, and sham sessions). The experimental design includes high‑fidelity, low‑noise audio recording in a soundproofed environment, standardised question protocols, multiple control conditions (no‑question silence, pure noise, sham inquirer absent), triple blinding (participants, judges, and analysts are blind to condition), objective acoustic analysis (spectral analysis, phonetic feature extraction), comprehensive environmental monitoring, and positive controls (known acoustic targets to validate the detection pipeline). The protocol mandates a pilot phase to calibrate recording equipment, validate the judge agreement metric, and estimate baseline pareidolia rates. The primary analysis uses a Linear Mixed Model (LMM) with fixed effects of Condition (active EVP recording vs. controls), with α = 0.025 (one‑tailed) and power ≥ 0.90 for a small to moderate effect (Cohen's d ≥ 0.10). The sample size is N = 200 independent recording sessions (baseline, with adaptive increase to N = 300 if pilot shows high variability). Statistical analyses are preregistered, include Bayesian inference (BF₁₀), permutation tests, and multiverse robustness checks (≥70% of 12 forks). The aim is to detect a reproducible EVP signal or to establish an upper bound on the effect size with unprecedented rigor.