Liu-Shuang Joan, Torfs Katrien, Rossion Bruno
Psychological Science Research Institute & Institute of Neuroscience, University of Louvain, Belgium.
Psychological Science Research Institute & Institute of Neuroscience, University of Louvain, Belgium.
Neuropsychologia. 2016 Mar;83:100-113. doi: 10.1016/j.neuropsychologia.2015.08.023. Epub 2015 Aug 28.
One of the most striking pieces of evidence for a specialised face processing system in humans is acquired prosopagnosia, i.e. the inability to individualise faces following brain damage. However, a sensitive and objective non-behavioural marker for this deficit is difficult to provide with standard event-related potentials (ERPs), such as the well-known face-related N170 component reported and investigated in-depth by our late distinguished colleague Shlomo Bentin. Here we demonstrate that fast periodic visual stimulation (FPVS) in electrophysiology can quantify face individualisation impairment in acquired prosopagnosia. In Experiment 1 (Liu-Shuang et al., 2014), identical faces were presented at a rate of 5.88 Hz (i.e., ≈ 6 images/s, SOA=170 ms, 1 fixation per image), with different faces appearing every 5th face (5.88 Hz/5=1.18 Hz). Responses of interest were identified at these predetermined frequencies (i.e., objectively) in the EEG frequency-domain data. A well-studied case of acquired prosopagnosia (PS) and a group of age- and gender-matched controls completed only 4 × 1-min stimulation sequences, with an orthogonal fixation cross task. Contrarily to controls, PS did not show face individualisation responses at 1.18 Hz, in line with her prosopagnosia. However, her response at 5.88 Hz, reflecting general visual processing, was within the normal range. In Experiment 2 (Rossion et al., 2015), we presented natural (i.e., unsegmented) images of objects at 5.88 Hz, with face images shown every 5th image (1.18 Hz). In accordance with her preserved ability to categorise a face as a face, and despite extensive brain lesions potentially affecting the overall EEG signal-to-noise ratio, PS showed 1.18 Hz face-selective responses within the normal range. Collectively, these findings show that fast periodic visual stimulation provides objective and sensitive electrophysiological markers of preserved and impaired face processing abilities in the neuropsychological population.
人类存在专门面部处理系统的最显著证据之一是后天性面孔失认症，即脑损伤后无法识别个体面孔。然而，使用标准的事件相关电位（ERP），如我们已故的杰出同事什洛莫·本廷深入报道和研究的著名的与面孔相关的N170成分，很难为这种缺陷提供一个敏感且客观的非行为学标记。在这里，我们证明电生理学中的快速周期性视觉刺激（FPVS）可以量化后天性面孔失认症中的面孔个体化损伤。在实验1（刘双等人，2014年）中，相同面孔以5.88赫兹的频率呈现（即约每秒6张图像，刺激间隔=170毫秒，每张图像1次注视），每5张面孔会出现不同的面孔（5.88赫兹/5 = 1.18赫兹）。在脑电图频域数据中，在这些预定频率（即客观地）识别出感兴趣的反应。一个经过充分研究的后天性面孔失认症（PS）病例和一组年龄及性别匹配的对照组仅完成了4个×1分钟的刺激序列，并伴有正交注视十字任务。与对照组不同，PS在1.18赫兹时未表现出面孔个体化反应，这与她的面孔失认症相符。然而，她在5.88赫兹时的反应反映了一般视觉处理，处于正常范围内。在实验2（罗西昂等人，2015年）中，我们以5.88赫兹的频率呈现物体的自然（即未分割）图像，每5张图像会出现一张面孔图像（1.18赫兹）。鉴于她保留了将面孔归类为面孔的能力，并且尽管广泛的脑损伤可能影响整体脑电图的信噪比，但PS在正常范围内表现出1.18赫兹的面孔选择性反应。总体而言，这些发现表明，快速周期性视觉刺激为神经心理学人群中保留和受损的面孔处理能力提供了客观且敏感的电生理标记。
