Purchase this article with an account.
Corinna Osterbrink, Arvid Herwig; Transsaccadic prediction of real-world objects. Journal of Vision 2019;19(10):237c. doi: https://doi.org/10.1167/19.10.237c.
Download citation file:
© ARVO (1962-2015); The Authors (2016-present)
During a saccade, a target object’s retinal position and spatial resolution change. Nevertheless, we do not perceive these changes, but instead experience a stable percept of our environment. One mechanism of how the brain deals with these saccade induced changes is that it forms associations between presaccadic peripheral and postsaccadic foveal input of saccade-target objects. Based on these transsaccadic associations the visual system can predict visual features across saccades. Up to now, studies investigating how transsaccadic feature prediction affects peripheral perception have focused primarily on simple visual features (e.g., spatial frequency, size and shape). They could show that peripheral perception is biased toward previously associated foveal input. The present study tested whether also complex visual features constituting real-world objects (fruits and balls) are predicted across saccades. In an eyetracking experiment, twenty-four participants first underwent an acquisition phase, in which they learned new object-specific transsaccadic associations. Six out of twelve objects were systematically swapped to an object of the opposite category (from fruit to ball or vice versa) during saccades. In the following test phase, participants were briefly presented peripheral saccade target objects and had to identify which object they saw. Objects which had previously been swapped during the acquisition phase were more often perceived as belonging to a different category compared to objects which stayed the same during acquisition. These category errors occurred mainly because participants confused the peripherally presented object with its transsaccadically associated foveal counterpart. This result indicates that transsaccadic prediction is object-specific and not limited to a small set of simple visual features.
This PDF is available to Subscribers Only