Abstract
Central vision loss disrupts patterns of eye movements during visual search due to the strong habit to fixate using the fovea. We investigated the effect of central vision loss on spatial attentional learning by simulating a scotoma in participants searching for a target that frequently appeared in one visual quadrant. Attentional biases may arise from an oculomotor habit to fixate that quadrant, in which case central vision loss may disrupt learning. Alternatively, if attentional learning relies on spatial representations not tied to oculomotor control, it may be unaffected by central vision loss. To dissociate these possibilities, we used a gaze-contingent paradigm to simulate a central scotoma 7º in radius during a location probability cueing task. Participants in Experiment 1 completed a two-phase T-among-L search task designed to induce attentional learning. In a training phase with simulated central vision loss, targets occurred in one screen quadrant with 50% probability. Despite the central scotoma, participants acquired location probability learning, responding faster and more accurately when targets appeared in the high probability quadrant compared with other quadrants. In a testing phase, targets appeared equally often in each location and participants searched with and without the scotoma in alternating blocks. Despite different eye movement patterns with and without the scotoma, the training effect persisted to the testing phase for both viewing conditions. Experiment 2 investigated whether learning with healthy vision transfers to a testing phase with central vision loss. This experiment was identical to Experiment 1, except the training phase did not include a scotoma. Learning acquired with healthy vision persisted to the testing phase whether participants searched with or without the scotoma. Overall, location probability learning is present during simulated central vision loss, suggesting that the acquisition of spatial attentional learning may not require healthy eye movement patterns.