Faculté de Psychologie et des Sciences de l'Éducation, University of Geneva Geneva, Switzerland.
Front Neurosci. 2014 Dec 4;8:390. doi: 10.3389/fnins.2014.00390. eCollection 2014.
A major effort in cognitive neuroscience of language is to define the temporal and spatial characteristics of the core cognitive processes involved in word production. One approach consists in studying the effects of linguistic and pre-linguistic variables in picture naming tasks. So far, studies have analyzed event-related potentials (ERPs) during word production by examining one or two variables with factorial designs. Here we extended this approach by investigating simultaneously the effects of multiple theoretical relevant predictors in a picture naming task. High density EEG was recorded on 31 participants during overt naming of 100 pictures. ERPs were extracted on a trial by trial basis from picture onset to 100 ms before the onset of articulation. Mixed-effects regression models were conducted to examine which variables affected production latencies and the duration of periods of stable electrophysiological patterns (topographic maps). Results revealed an effect of a pre-linguistic variable, visual complexity, on an early period of stable electric field at scalp, from 140 to 180 ms after picture presentation, a result consistent with the proposal that this time period is associated with visual object recognition processes. Three other variables, word Age of Acquisition, Name Agreement, and Image Agreement influenced response latencies and modulated ERPs from ~380 ms to the end of the analyzed period. These results demonstrate that a topographic analysis fitted into the single trial ERPs and covering the entire processing period allows one to associate the cost generated by psycholinguistic variables to the duration of specific stable electrophysiological processes and to pinpoint the precise time-course of multiple word production predictors at once.
语言认知神经科学的一个主要目标是确定与单词生成相关的核心认知过程的时间和空间特征。一种方法是通过在图片命名任务中研究语言和语言前变量的影响来实现。到目前为止，研究通过使用析因设计分析了一个或两个变量来研究单词生成过程中的事件相关电位 (ERP)。在这里，我们通过在图片命名任务中同时研究多个理论相关预测因子的影响来扩展了该方法。在 31 名参与者进行的显性命名 100 张图片的过程中，记录了高密度脑电图。从图片呈现到发音开始前 100 毫秒，逐试提取 ERP。进行混合效应回归模型以检查哪些变量会影响产生潜伏期和稳定的电生理模式（地形图）的持续时间。结果表明，一个语言前变量，视觉复杂性，对头皮上稳定电场的早期时期有影响，从图片呈现后 140 到 180 毫秒，这一结果与该时间段与视觉对象识别过程相关的假设一致。其他三个变量，词习得年龄、名称一致性和图像一致性，影响反应潜伏期并调制从 380 毫秒到分析期结束的 ERP。这些结果表明，将适合于单个试次 ERP 的地形分析并覆盖整个处理周期，可以将心理语言变量产生的成本与特定稳定电生理过程的持续时间相关联，并一次确定多个单词生成预测因子的精确时间过程。
