Abstract
Due to qualitative differences of visual information obtained from the periphery and the fovea, perception of objects can differ depending on their location in the visual field (Rosenholz, 2016). Together with an inherent tendency to perceive the environment as stable, this might result in biases in how we perceive changes in the environment. To investigate this, we examined change detection performance across saccades within one feature in respect to peripheral (presaccadic) and foveal (postsaccadic) perception.
We changed an object’s shape during a saccade from triangular to circular (TC), or vice versa (CT), and measured the point of subjective stability. Additionally, we tested the differences between peripheral and foveal shape perception, by measuring the point of subjective equality when the object appeared in the periphery or in the fovea only. The change discrimination task revealed that small CT changes were perceived as stable, whereas no change was more likely perceived as a TC change. In the shape perception task, we found peripheral perception being biased towards rounder percepts compared to foveal perception. Therefore, the bias for TC reports cannot be explained by relative biases in peripheral and foveal perception. Moreover, the bias does not represent a simple response bias, because it was also present in a criterion-free paradigm.
This asymmetry in intrasaccadic change perception might be explained by a bias in transsaccadic prediction. Since angular shapes provide finer spatial detail than circular shapes, one might assume that the TC direction leads to a reduction in detail across a saccade. However, the usual perceptual experience might be that objects become richer in detail in the fovea after a saccade. Hence, there might be a weaker violation of transsaccadic prediction for a CT-direction change. This might explain why the environment appears stable, although the fovea reveals new details with every saccade.