Structural and functional connectivity markers of developmental speech and language disorders

摘要 发育性言语和语言障碍影响了大约15%的儿童，并具有终身影响。 在社会和情感发展和就业方面。两种常见的神经发育障碍是 发育性语言障碍（DLD;也称为特定语言障碍）和发育性口吃， 分别影响7%和5%的儿童。尽管其流行和巨大的影响， 这些常见神经发育障碍的神经原因、相关因素和后果;因此， 有效的治疗仍然是难以捉摸的。在拟议的项目中，我们将研究这些神经基础， 使用磁共振成像（MRI）研究结构和功能神经连接的疾病。 以前对这些人群的连接性的研究是有限的，并且几乎没有达成共识，部分原因可能是 小样本量。对这两种疾病的理论解释都涉及到神经回路的功能障碍， 基底神经节在目前的提案中，我们将测试和比较神经系统的结构和功能完整性， 在DLD患者（N=80）和口吃患者（PWS; N=80）的大队列中， 在年龄和性别匹配的具有典型发育的对照组（N=160）中获得了类似的数据。 首先，我们将评估语音/语言特定网络中的连接性，使用扩散数据来评估结构 连接性和静息状态数据来评估功能连接性。结果将表明 在PWS和DLD患者的大队列中的连接。在每一种紊乱中，我们也将确定连接性 对个体行为差异的贡献。这将揭示不同的连接模式是如何 与严重性沿着相关维度的差异相关（例如，流利度，语言测量），理想情况下 导致神经相关的疾病。最后，我们将评估全脑功能连接 使用数据驱动的机器学习，每个障碍组与其匹配的对照组之间的差异 接近。结果将表明神经活动的模式，区分这些疾病从控制。的 这一提议的结果将是这些群体中潜在网络差异的表征， 这将理想地导致这些目标行为和神经调节治疗的发展， 多方面和广泛的疾病。研究和培训将在牛津大学进行，这是一个理想的地方。 在这样的环境中，我们可以进行这方面的研究。申请人将由世界领先的研究人员指导 与知识需要指导他在这项工作中，包括在发展的神经基础的专业知识， 言语和语言障碍，神经成像的尖端方法，以及机器学习。实现 这些目标将阐明这些言语和语言障碍的神经相关性， 申请人在该领域的独立研究生涯。

ABSTRACT Developmental speech and language disorders affect an estimated 15% of children and have lifelong impacts on social and emotional development and employment. Two common neurodevelopmental disorders are developmental language disorder (DLD; also called specific language impairment) and developmental stuttering, affecting 7% and 5% of children respectively. Despite their prevalence and immense impact, little is known of the neural causes, correlates, and consequences of these common neurodevelopmental disorders; thus, effective treatment remains elusive. In the proposed project, we will study the neural underpinnings of these disorders using magnetic resonance imaging (MRI) to study structural and functional neural connectivity. Previous studies of connectivity in these populations are limited and show little consensus, likely due in part to small sample sizes. Theoretical accounts of both disorders implicate dysfunctional neural circuits through the basal ganglia. In the current proposal, we will test and compare the structural and functional integrity of neural pathways in large cohorts of people with DLD (N=80) and people who stutter (PWS; N=80) and compare them with similar data obtained in age- and sex-matched control groups of people with typical development (N=160). First, we will assess connectivity in speech/language-specific networks, using diffusion data to assess structural connectivity and resting-state data to assess functional connectivity. Results will indicate abnormalities in connectivity in large cohorts of PWS and people with DLD. In each disorder, we will also determine connectivity contributions to individual differences in behavior. This will reveal how different connectivity patterns are correlated to differences in severity along relevant dimensions (e.g., fluency, language measures), ideally resulting in neural correlates of the disorders. Finally, we will evaluate whole-brain functional connectivity differences between each disorder group and its matched control group using data-driven machine learning approaches. Results will indicate patterns of neural activity that differentiate these disorders from controls. The outcome of this proposal will be the characterization of underlying network differences in these populations, which will ideally lead to the development of targeted behavioral and neuro-modulatory treatments of these multifaceted and pervasive disorders. Research and training will take place at the University of Oxford, an ideal environment in which to pursue this line of research. The applicant will be mentored by world-leading researchers with the knowledge needed to guide him in this work, including expertise in the neural bases of developmental speech and language disorders, cutting-edge methodology in neuroimaging, and machine learning. Achieving these aims will illuminate the neural correlates of these speech and language disorders as well as prepare the applicant for an independent research career in this area.