Centre de Recherche en Automatique de Nancy (CRAN), Université de Lorraine, France.
Department of Psychology, University of Victoria, Canada.
Neuropsychologia. 2019 Feb 18;124:44-54. doi: 10.1016/j.neuropsychologia.2019.01.005. Epub 2019 Jan 17.
While expert face discrimination develops naturally in humans, expert discrimination in non-face object categories, such as birds, cars and dogs, is acquired through years of experience and explicit practice. The current study used an implicit visual discrimination paradigm and electroencephalography (EEG) - Fast Periodic Visual Stimulation - to examine whether within-category discrimination of faces and non-face objects of expertise rely on shared mechanisms despite their distinct learning histories. Electroencephalogram was recorded while bird experts and bird novices viewed 60 s sequences of bird images or face images presented at a periodic rate of six images per second (i.e., 6.0 Hz). In the sequence, an adapting base image of a family-level bird (e.g., robin), a species-level bird (e.g., purple finch) or a face (e.g., Face A) was presented repeatedly for four consecutive cycles, followed by a different within-category "oddball" image at every fifth cycle (e.g., warbler, house finch, Face B). A differential response between the adapting base and the oddball images (6.0 Hz/fifth cycle = 1.20 Hz) provided an index of within-category discriminability. The results showed that both experts and novices demonstrated a robust EEG signal of equal magnitudes to the 6.00 Hz base face and bird images at medial-occipital channels and to the oddball 1.20 Hz face and bird images at the more anterior occipito-temporal channels. To examine whether the responses to faces and birds were generated by shared neural mechanisms, we correlated the responses to birds and faces at the participant-level. For the base signal at medial-occipital channels, all object categories positively correlated in both the experts and the novices, as expected given that the base signal indexes visual responses that are shared by all object categories (e.g., low-level). In contrast, for the discrimination signal at the more anterior occipito-temporal channels, the response to family- and species-level birds positively correlated with faces for the experts, but no face-bird association was found for the novices. These findings indicate the existence of partially shared neural mechanisms for within-category discrimination of faces and birds in the experts, but not in the novices.
虽然人类天生就能熟练地识别面部，但对鸟类、汽车和狗等非面部物体类别的熟练辨别则需要多年的经验和明确的实践。本研究采用内隐视觉辨别范式和脑电图（EEG）-快速周期性视觉刺激，来检验尽管面部和非面部物体的专长具有不同的学习历史，但它们的类别内辨别是否依赖于共同的机制。在实验中，鸟类专家和鸟类新手观看以每秒 6 个图像的周期性速度（即 6.0 Hz）呈现的 60 秒鸟类或面部图像序列，同时记录脑电图。在序列中，一个家族水平的鸟类（如知更鸟）、一个物种水平的鸟类（如紫翅雀）或一个面部（如 A 面）的适应基图像会在四个连续周期中重复出现，然后在每第五个周期会出现不同的类别内“异常”图像（如莺、黑头金丝雀、B 面）。适应基图像和异常图像之间的差异反应（6.0 Hz/第五周期=1.20 Hz）提供了类别内可辨别性的指标。结果表明，专家和新手在内侧枕叶通道对 6.00 Hz 的基底面部和鸟类图像以及在前侧枕颞通道对 1.20 Hz 的异常面部和鸟类图像都表现出同等幅度的强脑电图信号。为了检验对面孔和鸟类的反应是否是由共同的神经机制产生的，我们在参与者层面上对鸟类和面孔的反应进行了相关性分析。对于内侧枕叶通道的基信号，所有物体类别在专家和新手群体中都呈正相关，这是意料之中的，因为基信号索引了所有物体类别都共有的视觉反应（例如，低水平的）。相比之下，在前侧枕颞通道的辨别信号中，专家对家族和物种水平的鸟类的反应与面部呈正相关，但新手则没有发现这种关联。这些发现表明，在专家中存在部分共享的神经机制来进行面部和鸟类的类别内辨别，但在新手群体中则不存在。
