Neurobehavioral signatures of social avoidance and reinforcement learning - a transdiagnostic approach in anorexia nervosa and social anxiety

越来越多地，精神病学疾病是使用跨诊断和维度方法(例如RDoC)来概念化的。找出精神症状和诊断实体的神经行为机制的重叠、区别和维度差异对于最终推进临床翻译是至关重要的。支持接近和回避学习的神经行为系统在许多精神疾病中起着特别重要的作用，并在动物和人类研究中得到了广泛的研究。它们可以在计算模型方法中使用反馈学习范例进行优雅的评估，方法是将预测错误反应与强化和回避环境中的大脑激活联系起来。这种方法有巨大的潜力为精神症状的特征提供简明的神经行为描述，特别是对于调查不同诊断实体之间的重叠。在这里，我们使用这种跨诊断的方法来识别神经性厌食症(AN)和社交焦虑障碍(SAD)之间的异同，以及社交焦虑症状的特定维度影响。AN与高度功能失调的方法和回避行为有关，例如，对食物的无情关注和对苗条的追求，但对体重增加和避免进食的强烈恐惧。神经成像研究表明，在大脑皮质(如纹状体、脑岛和杏仁核)中，过度活跃的大脑系统用于反馈学习。最近的证据和长期存在的临床直觉表明，社交焦虑作为AN的前驱和维持因素具有重要作用，然而，目前还没有关于AN中反馈处理的社会方面的神经成像研究。另一方面，大量证据表明，社交焦虑症(SAD)有一种特定的神经加工模式，包括对积极社会反馈预期的激活减少，但对负面社会结果的激活增加。一些证据表明，社交焦虑和过度活跃的奖励系统的特征可以被认为是特质效应，这可能是疾病的前兆，并在康复后持续存在。我们将首次在跨精神病学诊断实体(AN和SAD)的计算建模的统一框架内，以及与康复AN和典型人群中的社交焦虑症状相关的统一框架内，对社会和非社会环境中的接近和回避系统及其神经实现进行系统的调查。这种方法与“RDoC”框架相兼容，可能使我们更接近对AN和SAD症状学的机械性理解，并可能最终指导开发更个性化的治疗方案，并根据精神症状群和疾病的神经行为表型确定风险概况。

Increasingly, psychiatric disease is conceptualized using transdiagnostic and dimensional approaches (e.g.RDoC). Pinpointing overlap, distinction and dimensional variance in neurobehavioral mechanisms of psychiatric symptoms and diagnostic entities is of utmost importance to ultimately advance clinical translation. Neurobehavioral systems supporting approach and avoidance learning play a particular important role in many psychiatric diseases and have been studied extensively in animal and human studies. They can be assessed elegantly using feedback learning paradigms in a computational modelling approach by relating prediction error responses to brain activation during reinforcement and avoidance contexts. This approach has an enormous potential to offer a parsimonious neurobehavioral description of the characteristics of psychiatric symptoms, in particular for the investigation of overlap across diagnostic entities. Here, we employ such a transdiagnostic approach to identify similarities and differences between Anorexia nervosa (AN) and social anxiety disorder (SAD), as well as the specific dimensional influence of social anxiety symptoms. AN is associated with highly dysfunctional approach and avoidance behaviors, e.g. relentless preoccupation with food and pursuit of thinness, but intense fear of gaining weight and avoidance of eating. Neuroimaging studies suggest hyperactive brains systems for feedback learning in AN (e.g. striatum, insula, and amygdala). Recent evidence and longstanding clinical intuition point towards an important role for social anxiety as a precursor and maintaining factor in AN, however, no respective neuroimaging studies on social aspects of feedback processing in AN exist. On the other hand, ample evidence suggests a specific pattern of disturbed neural processing in social anxiety disorder (SAD), including diminished activation for the anticipation of positive social feedback, but increased activation for negative social outcomes. Several lines of evidence suggest, that the characteristics of social anxiety and hyperactive reward system in AN can be considered trait effects, which may be a precursor to the disease and persist after recovery. We will, for the first time, provide a systematic investigation of approach and avoidance systems in social and non-social contexts and their neural implementation within a unifying framework of computational modelling across psychiatric diagnostic entities (AN and SAD), and in relation to social anxiety symptoms within recovered AN and the typical population. Compatible to the “RDoC” framework, this type of approach may take us closer towards a mechanistic understanding of the symptomatology of AN and SAD and may ultimately guide the development of more individualized treatment options and identification of risk-profiles based on neurobehavioral phenotypes of psychiatric symptom clusters and disease.