TY - CONF
T1 - On the visual span during object search in real-world scenes
AU - Nuthmann, Antje
PY - 2011/8
Y1 - 2011/8
N2 - How much information do viewers obtain from fixations when searching for an object in a real-world scene? This question was addressed in two studies, using the moving window paradigm. Scenes were presented at a viewing distance of 90 cm, each scene occupying a 18.6 x 24.8 deg field of view. Exp. 1 featured window radii measuring 1, 3, 4, 4.7, 5.4, and 6.1 deg. Exp. 2 featured six window radii measuring between 5 and 10 deg. Outside the window, the image was low-pass filtered to impede the parsing of the scene into constituent objects. Visual span was defined as the window size at which object search times became indistinguishable from search times in the no-window control condition; this occurred with windows measuring 7 deg and larger. For windows smaller than 6 deg, search initiation times and fixation durations increased as the window became smaller. Scanning times roughly mirrored the pattern of reaction times. Verification time was inflated for the 1 deg window (only). In all window conditions, viewers adapted smaller average saccade amplitudes than in the control condition. Finally, there were noteworthy individual differences in span size and eye-movement behaviors.
AB - How much information do viewers obtain from fixations when searching for an object in a real-world scene? This question was addressed in two studies, using the moving window paradigm. Scenes were presented at a viewing distance of 90 cm, each scene occupying a 18.6 x 24.8 deg field of view. Exp. 1 featured window radii measuring 1, 3, 4, 4.7, 5.4, and 6.1 deg. Exp. 2 featured six window radii measuring between 5 and 10 deg. Outside the window, the image was low-pass filtered to impede the parsing of the scene into constituent objects. Visual span was defined as the window size at which object search times became indistinguishable from search times in the no-window control condition; this occurred with windows measuring 7 deg and larger. For windows smaller than 6 deg, search initiation times and fixation durations increased as the window became smaller. Scanning times roughly mirrored the pattern of reaction times. Verification time was inflated for the 1 deg window (only). In all window conditions, viewers adapted smaller average saccade amplitudes than in the control condition. Finally, there were noteworthy individual differences in span size and eye-movement behaviors.
M3 - Poster
SP - 148
T2 - 16th European Conference on Eye Movements
Y2 - 21 August 2011 through 25 August 2011
ER -