Jane Jacob

Spatial attention effects during conscious and nonconscious processing of visual features and objects

Flanker congruency effects were measured in a masked flanker task to assess the properties of spatial attention during conscious and nonconscious processing of form, color, and conjunctions of these features. We found that (1) consciously and nonconsciously processed colored shape distractors (i.e., flankers) produce flanker congruency effects; (2) these congruency effects decrease with increasing separation between the probe and the flanking stimuli; (3) this decrease occurs even when flankers are suppressed from awareness; and (4) regardless of whether the form, color, or conjunction of these features have to be attended. Importantly, (5) flanker congruency effects decrease at nearly comparable rates when flankers are visible or masked, which suggests that for separate and conjoint features (i.e., objects), the gradient of spatial attention at the conscious and nonconscious levels of processing is almost identical, with the possibility of a slightly broader gradient for nonconsciously processed visual information.

Metacontrast masking with texture-defined second-order stimuli.

We examine metacontrast masking with texture-defined second-order stimuli. Our results reveal that (1) the monotonic type A as well as the nonmonotonic (U-shaped) type B metacontrast effect, which has been extensively examined with first-order luminance-defined stimuli, can be obtained with texture-defined second-order stimuli; and (2) while variations of luminance contrast are known to affect the magnitude of metacontrast with first-order stimuli, neither the size nor orientation contrast between texture elements defining the second-order stimuli have a significant impact on the magnitude or shape of metacontrast. These findings bear on theories of metacontrast masking by showing that the mechanism giving rise to nonmonotonic masking effects can operate beyond the level of first-order stimulus processing.

Microgenesis of surface completion in visual objects: Evidence for filling-out

Using metacontrast masking we examined the temporal dynamics of surface completion in object vision. By varying the stimulus onset asynchrony between the target object and the flanking mask(s), we obtained estimates of the time required for the entire surface contrast to fill out within the area delimited by the contours/edges of the target. The estimated speed of the filling-out process was 36.0 deg/s. Using existing estimates of cortical magnification, the computed filling-out speed in terms of cortical distance is .385 m/s, a value that approximates the estimated cortical filling-in speed and the speed of horizontal propagation in monkey V1. We discuss our results in relation to (1) prior findings of filling-in and filling-out phenomena, using surface completion in cortical space as the unifying principle, and (2) extant computational models of object vision.

Temporal dynamics of contour and surface processing of texture-defined second-order stimuli

Interaction between SOA and feature (p=0.063) Figure 2. Examples of target (left) and mask (right) stimuli used in Experiments 1 (top panel: texture contrast based on size differences) and 2 (bottom panel: texture contrast based on orientation differences). The observers judged whether the left or right corner of the target stimulus was missing (Expt 1 & Expt 2 contour task) and whether the texture elements defining the target were tilted to the left or right (Expt 2 surface task). Figure 5. Target accuracy in contour and surface feature discrimination tasks as a function of SOA for two observers (Expt 2). The baseline accuracy for both observers in each task was 1. Notice that optimal contour suppression (green lines) occurred at an SOA of 27 for both observers whereas optimal surface suppression (orange lines) occurred at an SOA of 53 ms (left panel) for one observer and at an SOA of 80 ms (right panel) for another observer.