Hübner Carolin, Schütz Alexander C
AG Allgemeine und Biologische Psychologie, Philipps-Universität Marburg, Marburg, Germany.
J Vis. 2017 Nov 1;17(13):12. doi: 10.1167/17.13.12.
Humans achieve a stable and homogeneous representation of their visual environment, although visual processing varies across the visual field. Here we investigated the circumstances under which peripheral and foveal information is integrated for numerosity estimation across saccades. We asked our participants to judge the number of black and white dots on a screen. Information was presented either in the periphery before a saccade, in the fovea after a saccade, or in both areas consecutively to measure transsaccadic integration. In contrast to previous findings, we found an underestimation of numerosity for foveal presentation and an overestimation for peripheral presentation. We used a maximum-likelihood model to predict accuracy and reliability in the transsaccadic condition based on peripheral and foveal values. We found near-optimal integration of peripheral and foveal information, consistently with previous findings about orientation integration. In three consecutive experiments, we disrupted object continuity between the peripheral and foveal presentations to probe the limits of transsaccadic integration. Even for global changes on our numerosity stimuli, no influence of object discontinuity was observed. Overall, our results suggest that transsaccadic integration is a robust mechanism that also works for complex visual features such as numerosity and is operative despite internal or external mismatches between foveal and peripheral information. Transsaccadic integration facilitates an accurate and reliable perception of our environment.
尽管视觉处理在整个视野中存在差异,但人类能够实现对视觉环境的稳定且均匀的表征。在此,我们研究了在扫视过程中,周边信息和中央凹信息在数量估计时进行整合的情况。我们要求参与者判断屏幕上黑白点的数量。信息呈现方式有三种:在扫视前呈现于周边区域,在扫视后呈现于中央凹区域,或者在两个区域连续呈现以测量跨扫视整合。与之前的研究结果不同,我们发现中央凹呈现时数量估计被低估,周边呈现时数量估计被高估。我们使用最大似然模型,根据周边和中央凹的值来预测跨扫视条件下的准确性和可靠性。我们发现周边和中央凹信息实现了近乎最优的整合,这与之前关于方向整合的研究结果一致。在连续三个实验中,我们破坏了周边和中央凹呈现之间的物体连续性,以探究跨扫视整合的限度。即便对于数量刺激的全局变化,也未观察到物体不连续性的影响。总体而言,我们的结果表明,跨扫视整合是一种强大的机制,它同样适用于诸如数量等复杂视觉特征,并且在中央凹和周边信息存在内部或外部不匹配的情况下仍能起作用。跨扫视整合有助于对我们的环境进行准确可靠的感知。