The effect of landmarks on visual stability in naturalistic scenes.

Undergraduate Just-In-Time Abstract

Poster Presentation 43.363: Monday, May 20, 2024, 8:30 am – 12:30 pm, Banyan Breezeway
Session: Undergraduate Just-In-Time 2

Garrett Hensley1, Jessica L. Parker1, A. Caglar Tas1; 1University of Tennessee, Knoxville

The present study investigated the effect of landmarks on visual stability in naturalistic scenes. We have previously that spatial position shifts in saccade target (ST) and saccade source (the fixated object prior to saccade) are better detected during the saccade compared to shifts in either background or whole scene, suggesting a special role of ST and source objects in establishing visual stability (Parker & Tas, VSS22). The present study expanded these findings by including a close (2dva) landmark object either above or below the ST. If presaccadic shift of attention spreads to the landmarks, then we expect to find similarly accurate displacement detection for ST and landmarks. If, however, STs are preferentially processed then we expect better displacement detection for STs than landmarks. On each trial, ST was cued with a red highlight, and participants were instructed to execute a saccade to it. During the saccade, one of the five possible horizontal shifts could occur: ST, landmark, whole image, background, no-shift (control). The shifts could be in the same (forward) or opposite direction (backward) as the saccade. Further, we manipulated visual stability with target blanking paradigm. Blanking significantly improved displacement detection, replicating previous work. We also found a significant effect of condition where shifts in ST (.91) were better detected than landmark (.58) and background shifts (.47). Importantly, shifts in landmark objects were better detected than background shifts, but not as good as whole image shifts (.88). We also found a significant interaction: Blanking significantly improved displacement detection for landmarks and whole image shifts, but not background or target shifts, possibly due to floor and ceiling effects respectively. These results show that the visual system uses various sources of information when establishing visual stability in naturalistic scenes, including the landmark objects with the priority given to the ST.