Testing for trans-saccadic prediction error signaling by foveal SC neurons. Top: Monkeys generated a delayed, visually-guided saccade towards an extrafoveal target. The authors used a delayed paradigm to make sure that there was a stable visual image upon saccade generation. In some trials, the saccade target was unchanged throughout the whole trial (high spatial frequency grating embedded within a circular patch for this shown example). In other trials, they detected saccade onset and immediately flipped the saccade target to another feature (from a low to a high spatial frequency texture in the shown example). Bottom: The authors only selected foveal SC neurons with response fields (RF’s) not extending towards the pre-saccadic extrafoveal stimulus location. In this example, the RF was almost entirely contained within <2 deg eccentricity. Each black dot is a stimulus onset location during RF mapping, and the white circle (3 deg radius) shows the extent of the saccade target if it was perfectly foveated post-saccadically. The target covered the RF post-saccadically but not pre-saccadically. The z-axis indicates the visual response strength of the neuron at each stimulus location.
Why don't #saccades disrupt our continuous #visual experience? This study shows that neurons of the #SuperiorColliculus are sensitive to the pre-movement peripheral appearance of the eye movement targets, potentially explaining the experienced perceptual stability @plosbiology.org 🧪 plos.io/44cxQIe