Generalization of extinction learning: basic mechanisms, individual differences and clinical implications for the etiology and treatment of anxiety disorders

根据抑制性学习模型，消退学习被认为是焦虑症暴露行为疗法改变的关键机制。在学习灭绝的过程中，根据纠正的信息，学习目前为止具有威胁性的刺激是安全的。为此，受影响的人说服自己，预期的不良后果不会发生。消亡学习作为一种活跃的新学习过程，涉及到记忆形成的一般机制。除记忆习得外，(再)巩固和保留也提出了消退记忆泛化的观点。在记忆泛化过程中，学习经验被转化为类似于原始学习刺激的刺激，但在主动学习过程中并不存在。然而，尽管有了明确的理论概念，但到目前为止，基础科学只是开始研究灭绝泛化的具体过程，并没有测试焦虑症患者可能存在的缺陷。一个详细的知识将有可能得出如何优化仍然有限的暴露疗法的重要建议。本研究计划旨在将一个既定的研究范式从研究恐惧学习的泛化转化为消退泛化的过程。在这篇文章中，我们研究了在3天灭绝范式中，灭绝记忆从先前熄灭的恐惧线索到未熄灭的但视觉相似的恐惧线索的泛化性能，该范式是为研究灭绝泛化而优化的。考虑到概括梯度，还考察了对以前未呈现的概括线索(在已熄灭和未熄灭线索之间的相似性维度上的相似程度的变化)的概括。在第二种范式中，灭绝学习的泛化在与灭绝背景相似的不同背景之间进行测试。概括消亡学习的能力是多模式评估的，包括主观-言语(价态、唤醒和对厌恶结果的预期的评级)和生理(恐惧增强的惊吓、自主神经唤醒、心率)指标。这项研究项目测试了关于概括表现的差异取决于急性焦虑症的倾向风险或存在的假说。因此，我们比较了25例急性焦虑症患者和25例急性焦虑症患者在特质焦虑方面的概括能力。

According to the inhibitory learning model, extinction learning is considered to be a key mechanism of change during the exposure-based behavioral therapy of anxiety disorders. During extinction learning a so far threatening stimulus is learned to be safe based on corrective information. For that, the affected persons convince themselves that expected aversive consequences do not occur. As an active process of new learning, extinction learning involves general mechanisms of memory formation. In addition to memory acquisition, (re-)consolidation and retention also extinction memory generalization is suggested. During memory generalization a learning experience is translated to a stimulus that is similar to the original learning stimulus but was not present during active learning. However, despite the clear theoretical conception, basic science only started to investigate the specific process of extinction generalization and did not test for possible deficits in patients with anxiety disorders so far. A detailed knowledge would make it possible to derive important recommendations how to optimize still limited exposure therapy.This research program here aims to translate an established research paradigm from studying the generalization of fear learning to the process of extinction generalization. Here, the performance of generalization of the extinction memory from an previously extinguished fear cue to a non-extinguished but visual similar fear cue is studied during a 3-day-extinction paradigm, that was optimized for the investigation of extinction generalization. Taking account a generalization gradient also the generalization to previously not presented generalization cues (varying intensity of similarity on the dimension of similarity between the extinguished and non-extinguished cue) is additionally examined. In a second paradigm, the generalization of extinction learning tested between contexts that vary in similarity to the extinction context. The capacity to generalize extinction learning is assessed multimodally including subjective-verbal (ratings of valence, arousal and expectancy of aversive outcomes) and physiological (fear-potentiated startle, autonomic arousal, heart rate) indicators. The research project tests the hypothesis of differences in the generalization performance depending on the dispositional risk for or the presence of acute anxiety disorders. Therefore, the generalization competence are compared between healthy subjects low, moderate and high (with 25 subjects each) in trait anxiety as a pre-clinical model of anxiety and 25 patients suffering form acute anxiety disorders.