Crowding

Chromatic crowding-effect for cone-isolating stimuli

Tatiana Bryakileva and Marina Danilova

A cone-isolating Landolt C (size 2 deg) was presented 5 deg to the left or right from fixation and was surrounded by four flanking bars isolating also one type of cones. The separation between the test and the flanks varied to estimate inhibitory areas. The observers' task was to indicate the orientation of the test. Colour contrast thresholds were measured for each separation between the test and the flanking bars. The inhibitory areas were estimated in two experiments: (a) The test Landolt C was presented to L-cones, the flanks were presented either to L-, or to M-, or to S-cones; (b) The test Landolt C and the flanking bars were presented to the same type of cones. A repeated-measures ANOVA showed that factor 'cone type' was significant in both experiments. The inhibitory areas were defined using post hoc pairwise comparisons with the contrast threshold for the isolated Landolt C. In first experiment the inhibitory areas were significantly smaller when the flanks were addressed to S-cones and increased in the case of the flanks addressed to L- and M- cones. In the second experiment the size of the area is largest in the case of S-cones and the smallest in the case of L-cones. Support: RFBR grant 12-04-01797-a.

Crowding suppresses the N1 wave

Vitaly Chicherov, Gijs Plomp and Michael H. Herzog

Flankers can strongly deteriorate performance on a visual target (crowding). For example, vernier offset discrimination is strongly affected by neighboring flankers. Interestingly, performance for longer and shorter flankers is better than performance for equal length flankers. We previously found that crowding is strongest when the vernier and the flankers group (same length flankers) and weaker when they ungroup (shorter or longer flankers). Here, we used high density EEG to investigate the mechanisms underlying crowding. The P1 wave amplitude correlated with the length of flankers (stimulus energy). The N1 wave amplitude correlated with performance, with the highest amplitudes occurring for long flankers, intermediate ones for short flankers, and low amplitudes for medium flankers. The N1 amplitude did not depend on the task difficulty per se but on target-flanker grouping, because increasing vernier offset size did not change the N1 amplitude. We show that the N1 reduction corresponds to suppression in crowding.

When crowding of crowding leads to uncrowding

Mauro Manassi and Michael H Herzog

In peripheral crowding, target perception can be strongly deteriorated by nearby flankers. What happens if flanker “A” crowds flanker “B” and “B” crowds target “C”? At 9° eccentricity, we determined offset discrimination thresholds for verniers. When the vernier was embedded in a square, performance deteriorated compared to the unflanked threshold (B crowds C). Interestingly, when adding more squares (flanker A), performance did not deteriorate as expected but strongly improved. Hence, when “A” crowds “B” and “B” crowds “C”, “A” un-crowds “C”. We propose that this un-crowding effect can be explained in terms of grouping. Grouping between the vernier and the central square leads to crowding. Grouping the central square and the nearby squares leads to grouping of the squares, ungrouping of the vernier, and un-crowding of the vernier.

Temporal dynamics of the remote facilitation effect in crowding

Bilge Sayim, John A. Greenwood and Patrick Cavanagh

In peripheral vision, a target letter is harder to identify when it is flanked by nearby letters, an effect called crowding. Here we show that items presented in the fovea that matched a crowded target reduced crowding (the remote facilitation effect), but only when the foveal item was presented within 200 ms of the target. An array of three horizontally arranged letters and numbers, with the target at the center, was presented in the periphery. Additionally, a single number or letter was presented at fixation. Observers indicated whether the target was a number or a letter. To ensure the allocation of attention to both positions, observers also indicated whether the foveal item was the same as the crowded target. The foveal item either matched (Matched condition) or did not match the target (Unmatched condition). We varied the stimulus onset asynchrony between the foveal item and the crowded array. Crowding was reduced in the Matched compared to the Unmatched condition when the foveal item appeared within 200 ms of the target. This limited temporal window for facilitation argues against cognitive biases as a factor in the effect. We suggest that the result is due to long-range grouping processes that precede crowding.

Perceptual learning enables the brain to overcome the blurred image received from the aging eye.

Uri Polat, Maria Lev, Oren Yehezkel and Anna Sterkin

In presbyopia (aging eye), the visual input to the brain is limited by the degraded optics of the eye resulting in multiple negative effects on near vision, specifically on reading, near visual acuity (VA), contrast sensitivity (CS), and slower processing speed. We recently showed that perceptual learning (PL) in presbyopia leads to remarkable improvement in VA (≈ 81%), CS and contrast discrimination, without changes in optical functions. We adapted our PL method for presbyopia to mobile devices (iPhone, iPad, iPod). The subjects were trained on contrast detection of Gabor targets under backward masking conditions, posing temporal constraints on the visual processing. The training (15 minutes/session, 3 times/week) covered a range of spatial frequencies. The CS on the mobile devices is better by factor of 2.5 than on PC. The improved visual performance enabled the presbyopes to overcome the disabilities imposed by the aging eye. Since PL improved CS and backward masking by decreasing the image processing latency and shortening the reaction time, the improved processing speed of presbyopes may enable recovering a reliable percept from the blurred image received from the aging eye. The results suggest that the aging brain retains enough plasticity to overcome the natural biological deterioration.

When perceived global shape is different from the sum of its parts

Branka Spehar and Luke Vu

Visual coding of shapes is crucial to object recognition and successful interaction within visual environment. Global shape representation depends on integration of local information about curvature (part shape) and its location in object-centered coordinates (part position). Our previous work has shown an important and differential role of contrast polarity in combining smooth parts of object contours compared to those at points of high curvature (Spehar and Clifford, 2002, Vision Research, 43, 1915-1919; Schira and Spehar, 2011, Frontiers in Psychology: Perception Science, 2: 47). Here we report a distortion in the perceived shape of a circle or a square when they are bisected in two halves that differ in either contrast polarity or brightness. When the two differently colored halves are oriented vertically, the perceived shape is vertically elongated, and when the two differently colored parts are oriented horizontally, the perceived shape appears horizontally elongated too. Previous investigations in areas of shape perception and shape illusions have predominantly focused on the role of geometrical factors. Our results demonstrate strong effects of other types of information, known to be important in perceptual segmentation, organization and grouping.