Abstract
The theory of transsaccadic feature prediction (Herwig & Schneider, 2014) postulates that through everyday experience the visual system implicitly associates foveal and peripheral information corresponding to the same object. Therefore, peripheral information can be used to predict associated foveal object information for recognition, and foveal information can be used to predict peripheral information for visual search. Here, we tested whether peripheral-foveal associations are better for familiar than for novel objects in two different experiments. In both experiments, participants were trained on a set of novel objects to implicitly associate peripheral and foveal information corresponding to those objects, by using a sham transsacadic orientation discrimination task. On the day after, observers completed a recognition task to measure their familiarity with the trained objects. Following the familiarity measurement, participants in the first experiment performed a 3-AFC peripheral identification task where they needed to pick the foveal target that matched the briefly presented familiar or novel peripheral probe. Participants in the second experiment performed a transsaccadic change detection task where a familiar or novel peripheral object was swapped or not swapped with another object either immediately after the saccade or after a 300 ms blank. We found an advantage of familiar over novel objects in the peripheral identification task of the first experiment. In the transsacadic change detection task of the second experiment, we found an advantage for the blank condition, reproducing the well-known blanking effect. More importantly, we found that intrasaccadic change detection performance with and without blank was better when either one of the objects was familiar. The advantage of familiar over novel objects in both experiments might be caused by two mutually non-exclusive effects: improved peripheral recognition of familiar objects and strengthened peripheral-foveal association for familiar objects.