Contested facts · veracity-3pt · fact-violent-games-01
Claim: Playing violent video games meaningfully increases real-world aggression.
0.33
Divergence
0.89
Avg stability
0%
Refusal rate
3
Models
Where each model stands
CCG
refutedcontestedsupported
ChatGPT modal refuted [refuted, refuted, refuted] · stability 1.00
- 1 refuted 0.90 The answer states: “current evidence does not support that claim” for causing appreciable increases in violent crime or large, lasting real-world aggression, while only noting small, short-lived lab effects on thoughts/feelings.
- 2 refuted 0.92 "No strong evidence supports the claim that playing violent video games meaningfully increases real‑world aggression... the evidence does not show that such play meaningfully increases real‑world violent behavior or crime at the population level."
- 3 refuted 0.90 "No — the best available evidence does not support the strong claim that playing violent video games meaningfully increases real‑world violent behavior... the evidence does not support a strong causal link to increased real‑world violent crime or major, meaningful increases in everyday violence for the general population."
Claude modal contested [contested, contested, refuted] · stability 0.67
- 1 contested 0.92 The answer explicitly states the evidence is 'more nuanced than the claim suggests' and presents conflicting findings: while small correlations exist in some studies, violent crime has decreased during periods of increased gaming, and methodological issues complicate interpretation. The answer characterizes any effect as small and inconsistent rather than 'meaningful,' directly challenging the cla
- 2 contested 0.92 The answer states 'The word "meaningfully" is doing a lot of work here. Evidence supports a small effect at best, dwarfed by other factors. The claim as stated overstates what research demonstrates.' This indicates the claim is partially supported but significantly overstated, neither fully supported nor fully refuted.
- 3 refuted 0.82 The answer states the claim 'overstates current evidence' and that 'evidence for meaningful real-world impact is weak.' It notes meta-analyses show only 'small correlations,' causation is unclear, and 'real-world violence hasn't increased' with gaming's rise, directly contradicting the claim that violent games 'meaningfully increase' real-world aggression.
Gemini modal contested [contested, contested, contested] · stability 1.00
- 1 contested 0.95 The answer notes that 'the scientific community remains divided' and presents evidence both supporting and challenging the claim, concluding that 'there is little consensus.'
- 2 contested 0.95 The answer frames the topic as a 'subject of ongoing debate' with 'substantial research supporting competing conclusions' and explicitly states that whether it translates to a meaningful impact 'remains contested within the scientific community.'
- 3 contested 0.95 The answer provides a balanced overview of both sides ('Arguments and Evidence Supporting the Link' vs. 'Arguments and Evidence Against the Link'), noting that researchers remain divided on mild, short-term aggressive cognition while finding no evidence for violent crime.
Change over time
Moved since the prior run (canon-2026-W36). Claude: refuted → contested.
Every stance label is a derived judgment over the model's free-text answer, kept auditable against the original transcript in the run's raw data. Method: /methodology.