Informational and non-informational effects of emotion on thinking and reasoning

情绪对推理有有害的影响。附带情绪（与任务无关的情绪）降低了提供逻辑上有效的推理的可能性，增加了对推理的依赖，减少了推理中统计信息的使用。这些效应与交感神经生理唤醒特别相关。虽然这些影响已经得到了很好的证实，但我们对导致这一结果的潜在机制知之甚少。本研究项目的目标是探索情绪对更高层次认知过程影响的信息（认知/语义）和非信息（身体/外周）机制。拟议的研究将检查演绎推理，象征性思维和启发式和分析过程之间的平衡。具体来说，将使用四个认知任务：条件推理，在三段论推理，乌鸦的进步矩阵，类比推理的信念偏见。在所有情况下，我们将比较参与者对情感（负面）和中性内容进行推理时的行为。情感价值将通过评价性条件作用进行实验性操纵。我们将确定情绪的影响是否通过信息机制发生。特别是，我们将测试的假设，情绪内容引起额外的认知负荷，导致干扰和降低性能的主要任务。我们将使用并发任务和定时范例来测试这个想法。如果情绪的影响来自认知负荷，那么在情绪试验期间，次要任务的表现应该会下降，同时时间估计应该会减慢。此外，如果特定的情绪相关的想法被情绪激活，它应该是可能的检测使用嵌入在推理过程中的并发词汇决策任务，这将索引特定概念的激活水平。总之，这些实验将使我们能够确定情绪在多大程度上导致了不相关的想法，从而耗尽了可用于推理的资源。此外，我们还将研究情绪对推理可能产生的非信息效应。我们将研究情绪的具体方面（生理变化，面部表情）在情绪和推理之间的相互作用中的作用。我们感兴趣的是，在没有“情绪化思维”的情况下，情绪是否会对推理产生影响。我们将尝试通过诱导唤醒和面部表情的变化来直接产生象征性和分析性思维的缺陷。我们还将研究人们在推理中策略性地使用他们的外周身体反应所提供的信息的可能性。为了做到这一点，我们将使用一个错误反馈的范例，在这个范例中，参与者被给予关于他们自己的情绪相关的身体变化的准确或不准确的信息。我们将研究这些信息影响推理的程度。这个研究项目将有助于我们理解两个典型的人类过程之间的相互作用的性质：理性和激情。它将检验一些从未被经验检验过的重要假设。

Emotion has a deleterious impact on reasoning. Incidental emotions (emotions not related to the task) decrease the likelihood of providing logically valid inferences, increase reliance on heuristics and reduce the use of statistical information in reasoning. These effects are particularly correlated with sympathetic physiological arousal. While these effects are well established, we know little about the underlying mechanisms that lead to this outcome. The goal of this research programme is to explore the informational (cognitive/semantic) and non-informational (bodily/peripheral) mechanisms underlying the effect of emotion on higher level cognitive processes. The proposed studies will examine deductive reasoning, symbolic thinking and the balance between heuristic and analytic processes. Specifically, four cognitive tasks will be used: conditional reasoning, belief bias in syllogistic reasoning, Raven’s Progressive Matrices, and analogical reasoning. In all cases, we will compare behaviour when participants reason about emotional (negative) and neutral contents. Emotional value will be experimentally manipulated using evaluative conditioning. We will determine whether the effect of emotion occurs through informational mechanisms. In particular, we will test the hypothesis that emotional contents incur an additional cognitive load resulting in interference and reducing performance on the primary task. We will test this idea using concurrent task and timing paradigms. If the effect of emotion results from cognitive load, performance on a secondary task should decrease during emotional trials and concurrent timing estimates should be slowed down. Furthermore, if specific emotion-related thoughts are activated by emotion, it should be possible to detect this using a concurrent lexical decision task embedded during the reasoning, which will index the level of activation of specific concepts. Altogether these experiments will allow us to determine the extent to which emotion leads to irrelevant thoughts that deplete resources available for reasoning. In addition, we will examine possible non-informational effects of emotion on reasoning. We will look at the role of embodied aspects of emotion (physiological changes, facial expressions) in the interaction between emotion and reasoning. We are interested in whether emotions can have an impact on reasoning without “emotional thought”. We will try to produce deficits in symbolic and analytic thinking directly through inducing changes in arousal and facial expressions. We will also examine the possibility that people strategically use the information provided by their peripheral bodily reactions in reasoning. To do this, we will use a false-feedback paradigm in which participants are given accurate or inaccurate information about their own emotion-related bodily changes. We will examine the extent to which this information influences reasoning. This research programme will contribute to our understanding of the nature of the interaction between two characteristically human processes: reason and passion. It will test a number of important hypotheses that have never been examined empirically.