持续性姿势-知觉性头晕中的注意功能障碍：机制证据的叙述性综述

文献信息

中文摘要

背景： 持续性姿势-知觉性头晕（PPPD）是一种常见的慢性功能性神经耳科疾病，其特征为持续的非旋转性头晕、不稳感，以及对直立姿势、运动和视觉复杂环境的高度敏感。目前已提出若干概念模型来解释其持续性，包括在失稳事件后重新适应失败、适应不良的多感觉整合伴视觉或躯体感觉权重过度，以及预测-实际感觉不匹配伴预测误差放大。本叙述性综述综合神经影像学、行为学、心理物理学和临床证据，以考察注意功能障碍作为一种候选的组织性机制，其可能将这些部分重叠的通路联系起来。
方法： 结果：涉及背外侧前额叶皮层（DLPFC）、前扣带皮层（ACC）、后岛叶、楔前叶、视觉皮层、前庭皮层区域，以及大规模注意、突显、感觉运动、视觉和默认模式网络的结构与功能改变，提示为分布式的功能重组，而非单一病灶性基础。行为学研究提示，在部分患者中，姿势控制和自我运动知觉可能变得过度依赖注意，表现为双任务效应、视觉探索改变、姿势策略僵化以及感觉-知觉标度异常。临床研究进一步显示平衡警觉、注意偏向和异质性症状表型，包括视觉不耐受、安静站立或坐位不耐受、被动运动不耐受和主动运动不耐受。我们提出一个修订的主动推断模型，其中注意参与调节视觉、前庭、躯体感觉、内感受和姿势预测误差信号之间失调的精度加权。该模型具有产生假设的性质，而非定论性。它可容纳有或无明确先前前庭疾病的患者，以及症状在无明显身体运动情况下发生的患者。该框架突出了未来研究的可检验靶点，包括事件相关电位、计算建模、双任务范式和表型分层治疗试验。
结论：

英文摘要

Persistent postural-perceptual dizziness (PPPD) is a common chronic functional neuro-otological disorder characterized by persistent non-spinning dizziness, unsteadiness, and heightened sensitivity to upright posture, motion, and visually complex environments. Several conceptual models have been proposed to explain its persistence, including failure of readaptation after a destabilizing event, maladaptive multisensory integration with excessive visual or somatosensory weighting, and prediction-actual sensory mismatch with amplified prediction errors. This narrative review synthesizes neuroimaging, behavioral, psychophysical, and clinical evidence to examine attentional dysfunction as a candidate organizing mechanism that may link these partly overlapping pathways. Structural and functional alterations involving the dorsolateral prefrontal cortex (DLPFC), anterior cingulate cortex (ACC), posterior insula, precuneus, visual cortex, vestibular cortical regions, and large-scale attention, salience, sensorimotor, visual, and default-mode networks suggest distributed functional reorganization rather than a single lesional substrate. Behavioral studies indicate that postural control and self-motion perception may become excessively attention-dependent in subsets of patients, with dual-task effects, altered visual exploration, stiffened postural strategies, and abnormal sensory-perceptual scaling. Clinical studies further show balance vigilance, attentional bias, and heterogeneous symptom phenotypes, including visual intolerance, quiet standing or sitting intolerance, passive motion intolerance, and active motion intolerance. We propose a revised active-inference model in which attention contributes to dysregulated precision weighting across visual, vestibular, somatosensory, interoceptive, and postural prediction-error signals. This model is hypothesis-generating rather than definitive. It accommodates patients with or without a clear preceding vestibular disorder and patients whose symptoms occur without overt body motion. The framework highlights testable targets for future studies using event-related potentials, computational modeling, dual-task paradigms, and phenotype-stratified treatment trials.