Repetition probability effects for inverted faces
Repetition probability effects for inverted faces
复制标题
DOI：
10.1016/j.neuroimage.2014.08.006
复制
发表时间：
2014-11-15
期刊：
影响因子：
5.7
通讯作者：
Kovacs, Gyula
中科院分区：
文献类型：
--
作者：
Grotheer, Mareike;Hermann, Petra;Kovacs, Gyula
已经表明，血氧水平依赖（BOLD）信号的重复相关降低由面部的重复概率（P（rep））调制（萨默菲尔德et al.，2008），为视觉感知的预测编码（PC）模型提供支持（Rao和Ballard，1999）。然而，重复抑制（RS）调制的P（rep）的人脸处理的阶段仍然不清楚。已知面部倒置会中断更高级别的面部/整体面部处理步骤，并且如果在面部处理的这些阶段发生P（rep）对RS的调制，则与直立面部相比，预计倒置的P（rep）效应会降低。因此，在这里，我们的目的是调查是否P（REP）的影响RS观察到的面部刺激起源于更高级别的面部处理的神经/整体阶段，通过比较这些影响直立和倒置的脸。类似于以前的研究，我们操纵P（rep）的刺激对在个别区块的功能磁共振成像记录。这种操作显着影响重复抑制后FFA，OFA和LO，独立的刺激方向。因此，我们的研究结果表明，RS在腹侧视觉流调制P（REP），即使在脸反转的情况下，因此强烈妥协的神经/整体的脸处理。一个额外的全脑分析也无法确定概率的调制效应是定向特异性的任何区域。这些结果表明，P（rep）效应可能起源于面孔加工的早期阶段。(c)2014爱思唯尔公司All rights reserved.
It has been shown, that the repetition related reduction of the blood-oxygen level dependent (BOLD) signal is modulated by the probability of repetitions (P(rep)) for faces (Summerfield et al., 2008), providing support for the predictive coding (PC) model of visual perception (Rao and Ballard, 1999). However, the stage of face processing where repetition suppression (RS) is modulated by P(rep) is still unclear. Face inversion is known to interrupt higher level configural/holistic face processing steps and if modulation of RS by P(rep) takes place at these stages of face processing, P(rep) effects are expected to be reduced for inverted when compared to upright faces. Therefore, here we aimed at investigating whether P(rep) effects on RS observed for face stimuli originate at the higher-level configural/holistic stages of face processing by comparing these effects for upright and inverted faces. Similarly to previous studies, we manipulated P(rep) for pairs of stimuli in individual blocks of fMRI recordings. This manipulation significantly influenced repetition suppression in the posterior FFA, the OFA and the LO, independently of stimulus orientation. Our results thus reveal that RS in the ventral visual stream is modulated by P(rep) even in the case of face inversion and hence strongly compromised configural/holistic face processing. An additional whole-brain analysis could not identify any areas where the modulatory effect of probability was orientation specific either. These findings imply that P(rep) effects on RS might originate from the earlier stages of face processing. (c) 2014 Elsevier Inc. All rights reserved.