Bridge Corpus — 2014–2024

Purpose

This file reconstructs the transition between Fanti's developmental/subtyping work and the later neuroscience-informed intervention program. The unit of analysis is not simply the paper; it is the claim, its evidentiary level, and the unresolved question passed downstream.

Provisional transition

Heterogeneity / equifinality → multi-system psychophysiological differentiation → candidate affective, attentional, and neural mechanisms → subtype- and task-specific neural/attention tests → experimental manipulation of candidate mechanisms → early mechanism-targeted intervention

Running in parallel:

environment / parenting ⇄ physiological sensitivity ⇄ behavior

The parallel loop is important. The bridge does not support a historical story in which biological explanation displaces transactional or ecological explanation.

Phase A — 2014–2017: heterogeneity becomes physiological

The central move is from behavioral subtyping toward testing whether apparently similar conduct problems have distinguishable fear, arousal, facial-response, and startle profiles.

Key propositions:

persistent conduct problems can arise with different fear/arousal profiles;
high-CU and lower-CU/high-anxiety pathways can show opposite physiological tendencies;
psychopathy dimensions show partially distinct psychophysiological correlates;
a unitary 'low arousal' account is therefore too coarse.

Phase B — 2018–2019: synthesis and constraint

The 2018 cross-age neurophysiology study and review consolidate a heterogeneity framework. The 2019 systematic review/meta-analysis then constrains simple biomarker claims: effects vary by autonomic measure, task, and study design, and baseline heart rate is not a universal discriminator.

Virtual reality enters this period primarily as an elicitation/measurement environment for fear physiology, not yet as a demonstrated treatment technology.

Phase C — 2020–2021: neural and attentional branching

The program tests amygdala activity, gaze, emotion recognition, prefrontal activity, and DLPFC stimulation.

The results are not reducible to a single hypoactivation story:

amygdala effects vary by subtype and conditioning phase;
eye-gaze differences do not fully explain recognition deficits in all samples;
LPE/CU-related prefrontal findings can include increased rather than reduced activity;
cTBS provides a true causal manipulation of a neural target, but its endpoint is proximal emotion recognition rather than durable antisocial behavior.

Phase D — 2022–2024: developmental extension, construct sensitivity, and intervention

Child studies extend psychophysiological and gaze findings downward developmentally, but measurement choice matters. A 2024 multi-method study found that subgroup differences depended on how CU/LPE was operationalized, and emotion-recognition accuracy itself did not differ between high- and low-CU groups.

At the same time, experimental and intervention studies increasingly target putative mechanisms directly. Parent-child sadness/emotional-engagement work demonstrates that proximal emotional processes can be modified. That does not by itself demonstrate durable reductions in aggression, conduct disorder, or forensic outcomes.

Principal bridge conclusion

The 2014–2024 record supports a transition from describing heterogeneous phenotypes to measuring and selectively manipulating candidate mechanisms. It does not establish a single biological signature of antisocial behavior, nor does it establish that changing a proximal mechanism reliably produces durable real-world behavioral change.