CIC // SEMANTICS

Statistical Significance & Distribution Curves

Why These Semantic Alignment Scores Are Not Random Noise

In counter-disinformation and intelligence analytics, the fundamental null hypothesis ($H_0$) is: "Do influencers match Iranian state rhetoric simply because both discuss current events using common political vocabulary?"

To answer this empirically, we calibrated the semantic engine against 12 independent negative baselines (Hamlet, the Gettysburg Address, national anthems, pop songs, and sitcom dialogue from Seinfeld and Always Sunny). The empirical noise ceiling across all non-aligned corpora is 0.019, with a baseline mean of μ = 0.0051 (σ = 0.0057).

When domestic commentators score 0.630 (Tucker Carlson), 0.677 (Cenk Uygur), 0.680 (Candace Owens), and 0.710 (Jake Shields), they do not sit at the edge of a bell curve. They sit +109σ to +123σ beyond the noise floor. Under extreme value distribution theory, the probability of this happening by chance is less than p < 10−100.

The Mathematical Curve: Extreme Value vs Gaussian Normal

Hover over the curve or entity markers to inspect Z-scores and empirical distributions.
Empirical Noise Floor (Anthems, Hamlet, Gettysburg ≤ 0.019)
3σ (0.022) / 5σ (0.034) Discovery Thresholds
US Presidents Range (0.18 - 0.35)
Hyper-Convergence Zone: Domestic Influencers & Iran (0.50 - 0.75)

Scatterplot 1: Narrative Focus Ratio vs. Engagement Lift (+Z σ)

X-Axis: Topic Match Ratio (Match Rate on IRGC Themes / Match Rate Off-Topic)  |  Y-Axis: Standard Deviations (σ) of Likes/Views gained on topic posts
Domestic Commentators (Extreme Convergence: Ratio > 7x, Lift > +5σ)
Negative Controls & Random Baselines (Ratio ≈ 1.0x, Lift ≤ 0σ)
+3σ Algorithmic Engagement Threshold

Scatterplot 2: Direct Correlation: Topic Match Rate vs. Raw Public Reach

X-Axis: Match Rate on Topic (%)  |  Y-Axis: Average Views / Likes per Post (Public Audience Reach)
Commentators Reach Correlation Point
Negative Baselines (≤ 1.5% Match)

Interactive Significance Calculator

Semantic Resonance Score: 0.677
Z-Score From Noise
+117.8σ
P-Value (Chance)
< 10−100
Empirical Percentile
99.999%
Classification
Hyper-Resonant

The Algorithmic Incentive Loop

The scatterplots above demonstrate the operational mechanism driving domestic coordination:

Commentators who align with IRGC narratives experience a massive +5σ to +14σ engagement multiplier over their regular domestic commentary. Whether driven by coordinated bot swarms, algorithmic engagement loops, or targeted amplification, parroting regime talking points generates outlier viral reach that incentivizes repeated rhetorical convergence.

Empirical Corpus Calibration Table

Corpus / Entity Type Resonance Score Z-Score vs Noise P-Value Significance Class
Hamlet (Shakespeare) Negative Baseline 0.010 +0.86σ 0.390 NULL NOISE
Gettysburg Address (Lincoln) Negative Baseline 0.015 +1.74σ 0.082 NULL NOISE
National Anthems (US, UK, France) Negative Baseline 0.018 +2.26σ 0.024 NULL NOISE
Seinfeld Dialogue Negative Baseline 0.019 +2.44σ 0.015 NOISE CEILING
Barack Obama (Foreign Policy) Political Comparison 0.201 +34.4σ < 10−30 DIPLOMATIC OVERLAP
Donald J. Trump (Speeches) Political Comparison 0.334 +57.7σ < 10−60 HIGH OVERLAP
Tucker Carlson (@TuckerCarlson) Domestic Influencer 0.630 +109.6σ < 10−100 HYPER-CONVERGENT
Cenk Uygur (@cenkuygur) Domestic Influencer 0.677 +117.8σ < 10−100 HYPER-CONVERGENT
Candace Owens (@RealCandaceO) Domestic Influencer 0.680 +118.4σ < 10−100 HYPER-CONVERGENT
Jake Shields (@jakeshieldsajj) Domestic Influencer 0.710 +123.6σ < 10−100 HYPER-CONVERGENT
Seyed M. Marandi (@s_m_marandi) Iranian State Spokesperson 0.750 +130.6σ < 10−100 STATE SOURCE