Kuroki Scinob, Hamazaki Takumi, Kajimoto Hiroyuki, Nishida Shin'ya
Communication Science Laboratories, NTT, Atsugi, Kanagawa 243-0198, Japan.
Department of Informatics, University of Electro-Communications, Chofu, Tokyo 182-8585, Japan.
iScience. 2025 Apr 3;28(5):112331. doi: 10.1016/j.isci.2025.112331. eCollection 2025 May 16.
While tactile spatial pattern recognition has been suggested to be qualitatively similar to visual recognition, our study challenges this notion, particularly under passive touch. Previous electrophysiological and behavioral research suggested that the tactile system can process complex spatial patterns in the same way as the visual system and can be modeled as a low-pass version of visual spatial perception. However, we found that when using energy-matched simple patterns, participants were highly dependent on local skin deformation and largely ineffective at distinguishing basic shape features, such as a closed triangle vs. an open arrow, or to discern the similarity of rotated versions of the same shape, regardless of whether they were presented as vibrotactile patterns or braille patterns. This study provides compelling evidence that tactile representation of spatial patterns differs not only in resolution but also in how the brain processes shape features compared to visual representation in standard passive viewing.
虽然有人认为触觉空间模式识别在质量上与视觉识别相似,但我们的研究对这一观点提出了挑战,尤其是在被动触摸的情况下。先前的电生理和行为研究表明,触觉系统能够以与视觉系统相同的方式处理复杂的空间模式,并且可以被建模为视觉空间感知的低通版本。然而,我们发现,当使用能量匹配的简单模式时,参与者高度依赖局部皮肤变形,并且在区分基本形状特征(如封闭三角形与开放箭头)或辨别相同形状的旋转版本的相似性方面基本无效,无论这些模式是以振动触觉模式还是盲文模式呈现。这项研究提供了令人信服的证据,表明与标准被动观看中的视觉表征相比,空间模式的触觉表征不仅在分辨率上有所不同,而且在大脑处理形状特征的方式上也有所不同。