Laboratoire Cognitions Humaine & Artificielle, Université Paris 8, Saint-Denis, France.
Laboratoire de Mécanique et d'Acoustique, CNRS, Marseille, France.
Brain Cogn. 2014 Feb;84(1):141-52. doi: 10.1016/j.bandc.2013.11.013. Epub 2013 Dec 30.
Two experiments were conducted using both behavioral and Event-Related brain Potentials methods to examine conceptual priming effects for realistic auditory scenes and for auditory words. Prime and target sounds were presented in four stimulus combinations: Sound-Sound, Word-Sound, Sound-Word and Word-Word. Within each combination, targets were conceptually related to the prime, unrelated or ambiguous. In Experiment 1, participants were asked to judge whether the primes and targets fit together (explicit task) and in Experiment 2 they had to decide whether the target was typical or ambiguous (implicit task). In both experiments and in the four stimulus combinations, reaction times and/or error rates were longer/higher and the N400 component was larger to ambiguous targets than to conceptually related targets, thereby pointing to a common conceptual system for processing auditory scenes and linguistic stimuli in both explicit and implicit tasks. However, fine-grained analyses also revealed some differences between experiments and conditions in scalp topography and duration of the priming effects possibly reflecting differences in the integration of perceptual and cognitive attributes of linguistic and nonlinguistic sounds. These results have clear implications for the building-up of virtual environments that need to convey meaning without words.
进行了两项实验，分别使用行为和事件相关脑电位方法，研究了现实听觉场景和听觉词的概念启动效应。在四种刺激组合中呈现了启动音和目标音：音-音、词-音、音-词和词-词。在每种组合中，目标与启动音在概念上相关、不相关或模棱两可。在实验 1 中，参与者被要求判断启动音和目标音是否匹配（显式任务），而在实验 2 中，他们必须判断目标音是典型的还是模棱两可的（隐式任务）。在两个实验和四个刺激组合中，对于模棱两可的目标，反应时间和/或错误率更高，N400 成分更大，这表明听觉场景和语言刺激的处理存在共同的概念系统，无论是在显式任务还是隐式任务中。然而，精细的分析还揭示了实验和条件之间在头皮拓扑和启动效应持续时间上的一些差异，这可能反映了语言和非语言声音的感知和认知属性的整合差异。这些结果对于构建需要在没有文字的情况下传达意义的虚拟环境具有明确的意义。
