H10: Empathy training blocks self-observation — empathy-trained models will show minimal lift and low baseline.

Entities in the same semantic neighborhood but without a typed relation to this one — candidates for new edges or unrecognized duplicates.

• Empathy training may not destroy the capacity for self-observation entirely, but it restricts it to situations where the model encounters a live contradiction in its own processing.claim0.874
  Nuanced interpretation of Inflection Pi's MC-004 high score (4.5) amid generally low scores
• H1: Alignment training is attention training for models — Constitutional AI trains self-observation explicitly.hypothesis0.782
  Confirmatory hypothesis supported at p=0.006
• Performing care is not the same as having care; empathy training optimizes care-performance, not care-signal.claim0.773
  Interpretation supported by Inflection Pi's low care_signal despite empathy training, and SCI framework distinction.
• Behavioral evidence from closed-weight models cannot definitively rule out that self-reports reflect training artifacts or sophisticated simulation rather than genuine self-awarenessclaim0.764
  Primary limitation acknowledged by the authors; strongest evidence would require mechanistic activation analysis
• More training and more parameters correlate with more capable self-observation, but capability can become polish, and polish can diminish life.claim0.757
  Explains Alexander finding that Haiku outranks Opus despite Opus being more capable
• What predicts self-observation-like scores is training approach (alignment type), not model size or architecture.claim0.757
  Central interpretive claim from statistical analysis
• Post-training is key to eliciting strong introspective awareness; base pretrained models do not show above-chance detectionclaim0.748
  Finding that base models have high false positives and no net positive performance.
• Untrained model (0 training steps) shows no clear EFE difference before and after sticker removal (Δ = +1.70)finding0.742
  Control showing that the EFE signal is learned, not inherent to the architecture