Canário Nádia, Jorge Lília, Loureiro Silva M F, Alberto Soares Mário, Castelo-Branco Miguel
IBILI - Institute for Biomedical Imaging in Life Sciences, Faculty of Medicine, University of Coimbra, Azinhaga Santa Comba - Celas, 3000-548 Coimbra, Portugal; ICNAS - Institute for Nuclear Sciences Applied to Health, Brain Imaging Network of Portugal, Portugal.
IBILI - Institute for Biomedical Imaging in Life Sciences, Faculty of Medicine, University of Coimbra, Azinhaga Santa Comba - Celas, 3000-548 Coimbra, Portugal.
Neuropsychologia. 2016 Jul 1;87:110-119. doi: 10.1016/j.neuropsychologia.2016.05.010. Epub 2016 May 11.
The ventral visual pathway receives both inputs from parvocellular and magnocellular pathways, and combines information from distinct high and low spatial frequency channels (HSF and LSF). Using a random effects region of interest general linear model approach (n=21), we aimed to compare the selectivity to different spatial frequency channels in eight key areas involved in visual object recognition: FFA, OFA, and STS, for face processing; FBA, and EBA as body selective regions; (dorsal and ventral) LOC for object perception; PPA for processing information of places and VWFA as a region which responds to written verbal material. We found that face and body selective regions had significantly higher response to LSF, suggesting an important contribution of holistic processing favoring LSF channels, while other object responsive regions had a higher response to HSF, suggesting a more important role for detailed component processing. Both FBA and VWFA failed to reveal a preference to SF content. These findings apply in general to the preferred category, with the notable exception of PPA, which revealed a higher response to HSF for all categories of stimuli. Our results suggest that areas along the ventral stream have distinct spatial frequency preferences that seem to reflect both the nature of visual objects being processed, their position in the visual hierarchy, task demands and the relevance of holistic versus detailed processing.
腹侧视觉通路接收来自小细胞通路和大细胞通路的输入,并整合来自不同高空间频率通道和低空间频率通道(HSF和LSF)的信息。我们采用随机效应感兴趣区域一般线性模型方法(n = 21),旨在比较参与视觉物体识别的八个关键区域对不同空间频率通道的选择性:用于面部处理的梭状回面孔区(FFA)、枕颞叶面孔区(OFA)和颞上沟(STS);作为身体选择性区域的梭状回身体区(FBA)和枕叶身体区(EBA);用于物体感知的(背侧和腹侧)外侧枕叶复合体(LOC);用于处理地点信息的海马旁回地点区(PPA)以及作为对书面语言材料作出反应区域的视觉词形区(VWFA)。我们发现,面部和身体选择性区域对低空间频率的反应显著更高,这表明有利于低空间频率通道的整体处理具有重要作用,而其他物体反应区域对高空间频率的反应更高,这表明详细成分处理发挥着更重要的作用。FBA和VWFA均未表现出对空间频率内容的偏好。这些发现总体上适用于偏好的类别,但海马旁回地点区是个显著例外,它对所有类别的刺激对高空间频率的反应都更高。我们的结果表明,腹侧流沿线的区域具有不同的空间频率偏好,这似乎既反映了所处理视觉物体的性质、它们在视觉层次结构中的位置、任务需求,也反映了整体处理与详细处理的相关性。