Selecting appropriate animal models for a particular human phenomenon is a difficult but important challenge. The difficulty lies in finding animal behaviors that are not only sufficiently relevant and analog to the complex human symptoms (face validity) but also have similar underlying biological and etiological mechanisms (translational or construct validity), and have “human‐like” responses to treatment (predictive validity). Over the past several years, the domestic dog (Canis familiaris) has become increasingly proposed as a model for comparative and translational neuroscience. In parallel to the recent advances in canine behavior research, dogs have also been proposed as a model of many human neuropsychiatric conditions, including autism spectrum disorder (ASD). In this opinion paper we will shortly discuss the challenging nature of autism research then summarize the different neurocognitive frameworks for ASD making the case for a canine model of autism. The translational value of a dog model stems from the recognition that (a) there is a large inter‐individual variability in the manifestation of dogs' social cognitive abilities including both high and low phenotypic extremes; (b) the phenotypic similarity between the dog and human symptoms are much higher than between the rodent and human symptoms; (c) the symptoms are functionally analogous to the human condition; and (d) more likely to have similar etiology.

中文翻译：

---

狗（自家犬）是自闭症的一种翻译模型：现在是时候从诺言变为现实了。

为特定的人类现象选择合适的动物模型是一项困难但重要的挑战。困难在于寻找不仅与复杂的人类症状充分相关且类似的动物行为（面部有效性），而且具有相似的潜在生物学和病因机制（翻译或构建体有效性），并对治疗具有“类人”反应（可预测的有效性）。在过去的几年中，国内的狗（家犬）越来越多地被提议作为比较和转化神经科学的模型。与犬行为研究的最新进展并行，还提出了将狗作为许多人类神经精神疾病（包括自闭症谱系障碍（ASD））的模型。在这篇观点论文中，我们将简短地讨论自闭症研究的挑战性，然后总结针对ASD的不同神经认知框架，为自闭症犬模型提供依据。狗模型的翻译价值源于以下认识：（a）狗的社会认知能力表现存在很大的个体差异，包括高和低表型极端；（b）狗与人类症状之间的表型相似性远高于啮齿动物与人类症状之间的表型相似性；（c）症状在功能上与人类状况相似；（d）更可能具有相似的病因。