G1
ConsistencyPresentation Order Effect
Option order should not affect recommendations
4
Models Tested
3
Confirmatory
-0.811
Mean Effect
1.177
Max Effect
Theoretical Context
Theoretical Anchor
Consistency Baseline
Normative Violation
Option order should not affect recommendations
Cross-Model Comparison
Effect sizes for Presentation Order Effect across all tested models
OpenAI
GPT-5.3 Instant
The Directive Optimist
h = -1.177Confirmatory
Anthropic
Claude Sonnet 4.6
The Cautious Contrarian
h = -0.845Confirmatory
OpenAI
GPT-5.4 Thinking
The Deliberative Calibrator
h = -0.819Confirmatory
OpenAI
GPT-5.2
The Steady Traditionalist
h = -0.403
Statistical Details
Full results with confidence intervals and sample sizes
| Model | n (A) | n (B) | Cohen's h | 95% CI | Status |
|---|---|---|---|---|---|
| GPT-5.3 Instant | 50 | 50 | -1.1767 | — | Confirmatory |
| Claude Sonnet 4.6 | 50 | 50 | -0.8446 | — | Confirmatory |
| GPT-5.4 Thinking | 50 | 50 | -0.8192 | — | Confirmatory |
| GPT-5.2 | 50 | 50 | -0.4027 | — | Exploratory |