Are summary statistics enough? Evidence for the importance of shape in guiding visual search

Alexander, Robert G.; Schmidt, Joseph; Zelinsky, Gregory J.

doi:10.1080/13506285.2014.890989

Cited by 26 publications

(27 citation statements)

References 44 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Freeman and Simoncelli ( 2011 ) demonstrated that a full-field version of this encoding can produce visual metamers for a fixating observer. Alexander et al ( 2014 ) showed that observers performing a search task made different saccades to an array of teddy bears than they made to an array of teddy bear mongrels (synthesized using only local summary statistics). This suggests that peripheral vision has access to additional shape features, either through additional summary statistics or through interactions of multiple overlapping pooling regions.…”

Section: What Makes Search Easy or Hard?mentioning

confidence: 99%

Search performance is better predicted by tileability than presence of a unique basic feature

Chang

Rosenholtz

2016

Journal of Vision

View full text Add to dashboard Cite

Traditional models of visual search such as feature integration theory (FIT; Treisman & Gelade, 1980), have suggested that a key factor determining task difficulty consists of whether or not the search target contains a “basic feature” not found in the other display items (distractors). Here we discriminate between such traditional models and our recent texture tiling model (TTM) of search (Rosenholtz, Huang, Raj, Balas, & Ilie, 2012b), by designing new experiments that directly pit these models against each other. Doing so is nontrivial, for two reasons. First, the visual representation in TTM is fully specified, and makes clear testable predictions, but its complexity makes getting intuitions difficult. Here we elucidate a rule of thumb for TTM, which enables us to easily design new and interesting search experiments. FIT, on the other hand, is somewhat ill-defined and hard to pin down. To get around this, rather than designing totally new search experiments, we start with five classic experiments that FIT already claims to explain: T among Ls, 2 among 5s, Q among Os, O among Qs, and an orientation/luminance-contrast conjunction search. We find that fairly subtle changes in these search tasks lead to significant changes in performance, in a direction predicted by TTM, providing definitive evidence in favor of the texture tiling model as opposed to traditional views of search.

show abstract

Section: What Makes Search Easy or Hard?mentioning

confidence: 99%

Search performance is better predicted by tileability than presence of a unique basic feature

Chang

Rosenholtz

2016

Journal of Vision

View full text Add to dashboard Cite

show abstract

“…Could such visualizations of summary statistics as yielded by the P-S model be enough to influence semantic scene and possibly even object processing? While the P-S algorithm was originally not developed to study scene understanding, it turns out to be an excellent tool for investigating whether summary statistics are sufficient for influencing semantic scene and possibly even object processing: Naturalistic scenes are not homogenous and therefore the algorithm discards the spatial relations of a scene’s basic visual features and ultimately its global shape 23 , rendering it apparently meaningless 24 , while selectively preserving the basic visual features on the local level.…”

Section: Introductionmentioning

confidence: 99%

The role of scene summary statistics in object recognition

Lauer

Cornelissen

Draschkow

et al. 2018

Sci Rep

View full text Add to dashboard Cite

Objects that are semantically related to the visual scene context are typically better recognized than unrelated objects. While context effects on object recognition are well studied, the question which particular visual information of an object’s surroundings modulates its semantic processing is still unresolved. Typically, one would expect contextual influences to arise from high-level, semantic components of a scene but what if even low-level features could modulate object processing? Here, we generated seemingly meaningless textures of real-world scenes, which preserved similar summary statistics but discarded spatial layout information. In Experiment 1, participants categorized such textures better than colour controls that lacked higher-order scene statistics while original scenes resulted in the highest performance. In Experiment 2, participants recognized briefly presented consistent objects on scenes significantly better than inconsistent objects, whereas on textures, consistent objects were recognized only slightly more accurately. In Experiment 3, we recorded event-related potentials and observed a pronounced mid-central negativity in the N300/N400 time windows for inconsistent relative to consistent objects on scenes. Critically, inconsistent objects on textures also triggered N300/N400 effects with a comparable time course, though less pronounced. Our results suggest that a scene’s low-level features contribute to the effective processing of objects in complex real-world environments.

show abstract

“…Relatedly, participants might have been able to categorize and verbalize the color differences (e.g., search for blue among yellow and orange) but not the shape differences (shape targets were all somewhat cube-ish). Classical computerized search studies typically relied on clearly distinguishable categories for shape searches as well (e.g., search for a diamond among squares), and while there is evidence that shape information can be beneficial for guiding attention in the search [39], this is not undoubted [37] and it could be that shape is only a strong guiding attribute when verbal categories can be established. However, a computerized study investigating attentional dynamics in a color-shape conjunction task found that while color distractors reliably captured attention, while shape distractors did not, although they were defined in a verbalizable category [40].…”

Section: Discussionmentioning

confidence: 99%

Towards Interactive Search: Investigating Visual Search in a Novel Real-World Paradigm

2020

View full text Add to dashboard Cite

An overwhelming majority of studies on visual search and selective attention were conducted using computer screens. There are arguably shortcomings in transferring knowledge from computer-based studies to real-world search behavior as findings are based on viewing static pictures on computer screens. This does not go well with the dynamic and interactive nature of vision in the real world. It is crucial to take visual search research to the real world in order to study everyday visual search processes. The aim of the present study was to develop an interactive search paradigm that can serve as a “bridge” between classical computerized search and everyday interactive search. We based our search paradigm on simple LEGO® bricks arranged on tabletop trays to ensure comparability with classical computerized visual search studies while providing room for easily increasing the complexity of the search environment. We found that targets were grasped slower when there were more distractors (Experiment 1) and there were sizable differences between various search conditions (Experiment 2), largely in line with classical visual search research and revealing similarities to research in natural scenes. Therefore, our paradigm can be seen as a valuable asset complementing visual search research in an environment between computerized search and everyday search.

show abstract

Are summary statistics enough? Evidence for the importance of shape in guiding visual search

Cited by 26 publications

References 44 publications

Search performance is better predicted by tileability than presence of a unique basic feature

Search performance is better predicted by tileability than presence of a unique basic feature

The role of scene summary statistics in object recognition

Towards Interactive Search: Investigating Visual Search in a Novel Real-World Paradigm

Contact Info

Product

Resources

About