The extraction of natural scene gist in visual crowding

Abstract

The gist of natural scenes can be extracted very rapidly and even without focal attention. However, it is unclear whether and to what extent the gist of natural scenes can break through the bottleneck of crowding, a phenomenon in which object recognition will be immensely impaired. In the first two experiments, a target scene, either presented alone or surrounded by four flankers, was categorized at basic (Experiment 1) or global levels (Experiment 2). It was showed that the elimination of high-level semantic information of flankers greatly alleviated the crowding effect, demonstrating that high-level information played an important role in crowding of scene gist. More importantly, participants were able to categorize the scenes in crowding at rather high accuracies, suggesting that the extraction of scene gist might be a prioritized process. To test this hypothesis, in Experiment 3 we compared the crowding effect of three types of stimuli, namely, scenes, facial expressions and letter “E”s. The results showed that scenes could be better categorized than the other two types of stimuli in the crowding condition. This scene gist advantage thus supported our hypothesis. Together, the present studies suggest that scene gist is highly recognizable in crowding, probably due to its prioritization in visual processing.

Flexible contextual modulation of naturalistic texture perception in peripheral vision

“…Interestingly, other studies have previously shown that contextual modulation can be reduced by removing the natural HOS from the surround. Perceptually, this has been observed for tasks involving recognition and discrimination of natural scenes in peripheral vision (Wallis, Bethge, & Wichmann, 2016; Gong et al, 2018), and for local orientation processing during scene perception (Neri, 2017). Neurally, it has been shown that phase-scrambling the surround (i.e.…”

“…Understanding the effects of surround structure on contextual modulation of texture perception is important because during natural scene perception there is abundant variability in texture properties and arrangement. Furthermore, different levels of surround structure are often used as proxies for different stages of neural processing (Manassi & Whitney, 2018; Farzin et al, 2009; Louie et al, 2007; Gong, Xuan, Smart, & Olzak, 2018), which could provide insights on the mechanisms behind our observations. For these reasons, we next asked how target-surround dissimilarity affects peripheral texture perception, and how it interacts with segmentation.…”

“…This interpretation seems also in line with previous work with artificial textures, proposing that the asymmetries between textures with uniform and random orientation in texture filling-in could be related to a process of perceptual inference (Hindi Attar, Hamburger, Rosenholtz, Götzl, & Spillmann, 2007). In other work using natural and phase-scrambled scenes, the effect of phase-scrambling has been explained (Gong et al, 2018) as resulting from a weaker engagement of higher areas in the visual hierarchy, leading to reduced contextual modulation in these higher areas. In the context of this literature, our finding of a relatively weak effect of HOS dissimilarity in the previous experiment raises the question of whether the presence of natural HOS is necessary for recruiting contextual modulation for textures.…”

Visual crowding in driving

“…However, the “dead leave” patches were added artificially and could be inconsistent with the perspective and depth of the natural scene. Gong et al (2018) studied crowding in the recognition of the gist of natural scenes, but they did not study crowding in object recognition in natural scenes. Previous studies also showed crowding of moving targets, but limited to oriented gratings and simple shapes ( Bex & Dakin, 2005 ; Bex et al, 2003 ; Dakin et al, 2011 ; Maus et al, 2011 ).…”

“…However, the vast majority of studies in the field have been mainly restricted to experiments in psychophysical laboratories. Stimuli typically used are usually static, artificial, and relatively simple, such as oriented gratings, shapes, letters, symbols, and even faces; only a few experiments have studied the impact of crowding in static natural scenes or of natural textures ( Gong, Xuan, Smart, & Olzak, 2018 ; Wallis & Bex, 2012 ). Furthermore, most crowding studies rely on participants’ explicit responses made during psychophysical tasks (e.g.…”