信任神经生物学中的任务结构与构念效度：一项系统结构综合

The neuroscience of interpersonal trust has expanded across functional magnetic resonance imaging, electroencephalography, functional near-infrared spectroscopy, pharmacology, genetics, physiology, and brain stimulation. Although this literature implicates distributed reward-, salience-, mentalizing-, and executive-control systems, findings are often difficult to compare because nominally similar Trust Game paradigms differ in their task architecture and analytic definition of trust. Here, we provided a PRISMA-guided systematic structural synthesis of 198 empirical studies comprising 200 experiments. Rather than conducting an effect-size or coordinate-based meta-analysis, we coded paradigm architecture, economic structure, temporal dynamics, trustee information, control conditions, participant populations, measurement modalities, and analytic strategies to characterize conditions relevant to neurobiological interpretation. Our synthesis showed that trust neuroscience is dominated by standard and binary Trust Game variants, simulated partners, limited explicit controls, and decision- and outcome-phase analyses. Exploratory contingency analyses identified associations among task architecture, measurement approach, and analytic operationalization without testing neurobiological effects. These features may emphasize different trust-related processes, including valuation of expected reciprocity, betrayal aversion under vulnerability, social inference about partner intentions, executive regulation, and feedback-based updating. We therefore propose a structural-computational alignment framework in which neurobiological findings can be interpreted relative to the task features and analytic contrasts under which they are observed. This framework reframes heterogeneity in trust neuroscience as a construct-validity problem rather than simple theoretical fragmentation. Explicit alignment among task design, inferred computation, analytic choices, and neurobiological measures may support a coherent mechanistic account of interpersonal trust.

中文翻译还没有生成，先显示原文。