Regional effects of clutter on human target detection performance

Abstract

Clutter is something that is encountered in everyday life, from a messy desk to a crowded street. Such clutter may interfere with our ability to search for objects in such environments, like our car keys or the person we are trying to meet. A number of computational models of clutter have been proposed and shown to work well for artificial and other simplified scene search tasks. In this paper, we correlate the performance of different models of visual clutter to human performance in a visual search task using natural scenes. The models we evaluate are Feature Congestion (Rosenholtz, Li, & Nakano, 2007), Sub-band Entropy (Rosenholtz et al., 2007), Segmentation (Bravo & Farid, 2008), and Edge Density (Mack & Oliva, 2004) measures. The correlations were performed across a range of target-centered subregions to produce a correlation profile, indicating the scale at which clutter was affecting search performance. Overall clutter was rather weakly correlated with performance (r ≈ 0.2). However, different measures of clutter appear to reflect different aspects of the search task: correlations with Feature Congestion are greatest for the actual target patch, whereas the Sub-band Entropy is most highly correlated in a region 12° × 12° centered on the target.

Targets Need Their Own Personal Space: Effects of Clutter on Multiple-Target Search Accuracy

“…Previous research has demonstrated an impact of visual clutter on search performance in single-target searches (e.g., Rosenholtz et al, 2007) and, more recently, within a 6 to 7 radius around a target (Asher et al, 2013). However, the current results did not demonstrate a significant effect of clutter on the low-or high-salience, single-target hit rates or response time measures.…”

“…In the present study, we adopted the radius-specific clutter approach from Asher et al (2013) and used a simplified search array. This allowed us to define clutter purely as a function of the number of items near a target.…”

Display Clutter

“…This finding was attributed to the smaller difference in clutter between the Neider and Zelinsky images, suggesting that Rosenholtz et al's algorithms are sensitive only to large changes in clutter. In addition, Asher et al (2013) found no significant correlation between search time and any of the metrics of Rosenholtz et al, although there were some significant correlations with true and false positive rates for subband entropy and edge density (both r = -.21, p = .02; measures were averaged per image). In this case, the authors explained the discrepancy with the fact that they used real-life images of natural scenes as opposed to maps.…”

Improving Route Directions: The Role of Intersection Type and Visual Clutter for Spatial Reference

“…In complicated intersections, people take longer to navigate and make more mistakes (Montello, 2005), while the level of visual clutter was shown to increase the time needed for finding an object in a visual scene (Asher, Tolhurst, Troscianko, & Gilchrist, 2013). In a complex situation, even if the target street can be successfully referred to a via path reference, mentioning landmarks could be truly helpful for the addressee.…”

“…The excess of items and their disorganized display lead to crowding and occlusion, thus decreasing object recognition performance (Bravo & Farid, 2006) and increasing the difficulty of both segmenting a scene (Bravo & Farid, 2004) and performing visual search (Asher et al, 2013;Henderson, Chanceaux, & Smith, 2009;Neider & Zelinsky, 2011). The number of objects in a scene positively correlates with the reaction times for finding a target (Rosenholtz et al, 2007).…”