The primary objectives of this study were to evaluate the structure and age-related stability of social attention in English and Arabic-speaking youth and to compare social attention between children with autism spectrum disorder (ASD), other developmental disabilities (DD), and typically-developing controls. Eye-tracking data were collected from US (N = 270) and Qatari (N = 242) youth ages 1–17, including children evaluated for possible ASD. Participants viewed 44 stimuli from seven social paradigms. Fixation was computed for areas of interest within each stimulus. Latent variable models examined the structure of social attention. Generalized estimating equation models examined the effect of age, sex, culture, and diagnostic group on social attention. The best-fitting model included a general social attention factor and six specific factors. Cultural differences in social attention were minimal and social attention was stable across age (r = 0.03), but females showed significantly greater social attention than males (d = 0.28). Social attention was weaker in DD (d = −0.17) and lowest in ASD (d = −0.38) relative to controls. Differences were of sufficient magnitude across areas-of-interest to reliably differentiate DD from controls (AUC = 0.80) and ASD-only from all other cases (AUC = 0.76). A social attention dimension that represents an early-life preference for socially salient information was identified. This preference was cross-culturally consistent and stable across development but stronger in females and weaker in DD, especially ASD. Given rapid and easy-to-collect remote eye tracking administration, social attention measurement may be useful for developmental monitoring. Acquisition of population norms, analogous to height/weight/head circumference, might enhance early screening and tracking of neurodevelopment.

中文翻译：

---

社会关注作为跨文化跨诊断神经发育风险标志物

本研究的主要目的是评估英语和阿拉伯语青年社会注意力的结构和年龄相关稳定性，并比较自闭症谱系障碍 (ASD)、其他发育障碍 (DD) 和典型儿童之间的社会注意力。 - 开发控制。眼动追踪数据收集自美国 ( N  = 270) 和卡塔尔 ( N = 242) 1-17 岁的青少年，包括评估可能存在 ASD 的儿童。参与者观察了来自七种社会范式的 44 种刺激。针对每个刺激内的感兴趣区域计算注视。潜变量模型检验了社会注意力的结构。广义估计方程模型检验了年龄、性别、文化和诊断群体对社会关注的影响。最佳拟合模型包括一般社会关注因素和六个特定因素。社会关注的文化差异很小，社会关注跨年龄稳定（r = 0.03），但女性表现出明显高于男性的社会关注（d = 0.28）。与对照组相比，DD (d = -0.17) 的社会注意力较弱，而 ASD (d = -0.38) 的社会注意力最低。感兴趣区域之间的差异足够大，可以可靠地区分 DD 与对照 (AUC = 0.80) 和仅 ASD 与所有其他情况 (AUC = 0.76)。确定了代表早年对社会显着信息的偏好的社会注意力维度。这种偏好在整个发展过程中具有跨文化一致性和稳定性，但在女性中更强，而在 DD 中更弱，尤其是 ASD。鉴于快速且易于收集的远程眼动追踪管理，社会注意力测量可能对发育监测有用。获取人口规范，类似于身高/体重/头围，