| C01 | Persistent conduct problems are physiologically heterogeneous; high-CU/low-anxiety and high-anxiety pathways can show different or opposite fear profiles. | Supported | Association / subgroup differentiation | Subtype definitions vary by study. |
| C02 | CU-related affective atypicality is detectable across multiple response systems. | Supported with constraints | Association | No single physiological marker uniquely defines CU traits. |
| C03 | Primary/low-anxiety and secondary/high-anxiety variants can differ in defensive/threat physiology. | Supported | Association | Developmental and task dependence remain. |
| C04 | Psychopathy dimensions have partially distinct physiological correlates. | Supported | Association | Dimensional measures and samples vary. |
| C05 | A universal low-arousal biomarker model of conduct problems is inadequate. | Strongly supported constraint | Meta-analytic synthesis | Some measures/tasks still show reliable group effects. |
| C06 | Emotion-processing differences can involve attention, recognition, expression, physiology, and neural processing, and these components need not move together. | Supported synthesis | Association → selected manipulation | Cross-study measures differ. |
| C07 | CU-related emotion-recognition difficulties cannot be reduced to a simple 'does not look at the eyes' explanation. | Supported constraint | Association | Eye allocation remains relevant in some child samples and operationalizations. |
| C08 | Neural findings do not support a single generalized hypoactivation account. | Supported constraint | Association | Region/task/subtype effects require finer mapping. |
| C09 | DLPFC activity contributes causally to selected emotion-recognition performance. | Supported in experimental sample | Experimental causal manipulation | Proximal outcome; nonclinical/symptom-stratified young adults. |
| C10 | A proximal emotional mechanism can be changed without demonstrating durable antisocial-behavior change. | Strongly supported inference boundary | Experimental manipulation / intervention | Requires distal follow-up trials. |
| C11 | Physiological reactivity can moderate person-by-context effects involving parenting and later behavior. | Supported longitudinally | Temporal prediction / moderation | Does not alone prove mediation or mechanism. |
| C12 | CU/LPE measurement choice can materially alter detected attentional differences. | Supported | Association / construct-validity constraint | Requires replication across samples. |
| C13 | Preschool conduct-problem groups can show atypical emotional reactivity without clear CU-specific differentiation. | Supported constraint | Association | Sample and task specificity. |
| C14 | VR enters the bridge corpus as a controlled emotion-elicitation/measurement environment before it is evidenced as treatment. | Supported historical classification | Methodological transition | Later treatment work must be mapped separately. |
| C15 | By 2024, mechanism-targeted intervention is present, but proximal efficacy and durable clinical effectiveness remain separate gates. | Supported | Intervention / proximal outcome | Distal and real-world outcomes remain uneven. |
