Fast and involuntary reactions to the social message signaled by emotional faces. The role of moderating variables

这个项目旨在测试是哪个过程驱动了对面部表情的快速和无意识的反应。一个普遍的假设是，积极的表达总是自动引起积极的评价反应（例如，接近），而消极的表达总是自动引起消极的评价反应（例如，回避）。相反，我们假设反应也可以由情绪表达所发出的社会信息触发。例如，内部群体成员表现出的快乐表情可能会立即被视为表达友好的愿望，并引发积极的反应，而同样的表情如果出现在受到负面评价的外部群体成员身上，可能会立即被视为傲慢或幸灾乐祸的迹象，并引发消极的反应。在第一个资助期，我们通过不同的因变量发现了这一社会信息假说的证据：在接近/回避范式中，群体内成员表达的恐惧和快乐引发了和谐行为（即快乐接近和恐惧回避比反向分配产生更快的反应），而群体外成员表达的情绪引发了相反的模式。我们在惊吓反射范式中发现了相同的模式，我们表明，对愤怒表达的反应受到对手体力的影响。然而，我们的研究结果也显示了社会信息假设的边界条件：我们发现，在接近/回避任务中使用的对比情绪会影响驱动反应的过程，例如，相同的表达（即愤怒）如果与快乐配对，则产生回避反应，而如果与恐惧配对，则产生接近反应。我们将这种模式解释为基于价值的反应与基于社会信息的反应。我们还发现了意想不到的局限性：在评价性启动任务中没有发现情绪与群体的相互作用。因此，在第一个资助阶段的基础上，我们可以得出结论，对情绪表达的快速和无意反应只有在特定条件下才会受到各自社会信息的影响。因此，重要的是要检查哪些方面触发了哪些反应。这是第二个筹资期的目标。因此，我们假设对情绪表达的（名义）效价作出反应是默认的，因为这是最显著的特征。然而，在社会信息与感知者特别相关的情况下，对表达的社会信息作出反应更具适应性。我们认为，如果满足两个先决条件，情感表达所传达的社会信息的相关性就会建立起来：情感表达对感知者来说是任务相关的（“任务相关性”），并且存在指向表达所传达的社会信息的信息价值的语境线索（“语境线索”）。

This project aims to test which process drives fast and involuntary reactions to emotional facial expressions. A prevailing hypothesis is that positive expressions automatically always elicit a positive evaluative response (e.g., approach) whereas negative expressions automatically always elicit a negative evaluative response (e.g., avoidance). In contrast, we hypothesized that responses can also be triggered by the social message signaled by an emotional expression. For example, the happy expression shown by a member of an in-group might immediately be seen as expressing a friendly wish for affiliation and elicit a positive reaction, whereas the same expression might be immediately taken as a sign of arrogance or Schadenfreude if shown by a member of a negatively evaluated out-group and elicit a negative reaction. We found evidence for this social message hypothesis with different dependent variables in the first funding period: In the approach/avoidance paradigm, fear and happiness expressed by in-group members elicited concordant behavior (i.e, happy-approach and fear-avoidance produced faster responses than the reversed assignment), whereas emotions expressed by out-group members elicited a reversed pattern. We found the same pattern with the startle reflex paradigm and we showed that reactions to expressions of anger are influenced by the physical strength of the opponent. However, our findings also showed boundary conditions of the social message hypothesis: We found that the contrast emotion employed in the approach/avoidance task influences which process drives the reactions such that the same expression (i.e., anger) produced an avoidance reaction if paired with happiness but an approach reaction if paired with fear. We interpreted this pattern as valence-based responses versus social message-based responses. We also found unexpected limitations: The emotion x group interaction was not found in the evaluative priming task. Thus, on the basis of the first funding phase, we can conclude that fast and unintentional reactions to emotional expressions are influenced by their respective social message only under specific conditions. Thus, it is important to examine the circumstances under which reactions are triggered by which aspect. This is the goal of the second funding period. We thereby hypothesize that it is the default to react to the (nominal) valence of an emotional expression since this is the most salient feature. However, in situations in which the social message is especially relevant for the perceiver, it is more adaptive to react to the social message of the expression. We propose that relevance of the social message signaled by the emotional expression is established if two prerequisites are fulfilled: the emotion expression is task-relevant for the perceiver (“task-relevance”) and context cues which point to the informative value of the social message signaled by the expression for the perceiver are present (“context cues”).