Scene context influences without scene gist: Eye movements guided by spatial associations in visual search

Abstract

Although the use of semantic information about the world seems ubiquitous in every task we perform, it is not clear whether we rely on a scene's semantic information to guide attention when searching for something in a specific scene context (e.g., keys in one's living room). To address this question, we compared contribution of a scene's semantic information (i.e., scene gist) versus learned spatial associations between objects and context. Using the flash-previewmoving-window paradigm Castelhano and Henderson (Journal of Experimental Psychology: Human Perception and Performance 33:753-763, 2007), participants searched for target objects that were placed in either consistent or inconsistent locations and were semantically consistent or inconsistent with the scene gist. The results showed that learned spatial associations were used to guide search even in inconsistent contexts, providing evidence that scene context can affect search performance without consistent scene gist information. We discuss the results in terms of hierarchical organization of top-down influences of scene context.

Visual search in divided areas: Dividers initially interfere with and later facilitate visual search

“…Recent studies have reported that the global information of a scene (e.g., gist or layout) can influence visual search (e.g., Castelhano & Heaven, 2011;Castelhano & Henderson, 2007;Torralba, Oliva, Castelhano, & Henderson, 2006;Võ & Henderson, 2010). For example, Võ and Henderson suggested that scene representations generated from only a brief glimpse of a whole scene may provide sufficient information to guide gaze during an object search.…”

Cat and mouse search: the influence of scene and object analysis on eye movements when targets change locations during search

“…Their findings showed that an inconsistent object was fixated on earlier and for a longer duration than a consistent object, though attempts to replicate this finding have had mixed results [12 -15]. When the task is to search for objects that might be semantically consistent or inconsistent with scene context, attention is not initially drawn to inconsistent targets [6,16] but response times are longer. Both of the studies showing this result limited visibility of the scene to a small window during search; we could find no studies that investigated the speed of searching for semantically inconsistent targets in fully visible scenes, and so the influence of semantic consistency on guidance of later stages of search is not yet fully understood.…”

“…In two studies that manipulated the plausibility of target locations in scenes, objects that appeared in likely locations were found more quickly than targets that appeared in unlikely locations, even though the use of windowed viewing during search meant that detailed scene syntax could only be derived during the preview of the scene [6,16]. Another study used fully visible scenes, and showed that the first saccade during search landed closer to the likely target location than to the unlikely target location, when the target was both present and absent [9].…”

“…The role of scene knowledge has been demonstrated by giving foreknowledge about a scene by showing a brief preview of it prior to defining the search target. Doing so reduces the time to find objects [5][6][7][8]. Influential factors include the pre-existing representations of scene structure and the 'fit' of the object to semantics of the scene and to locations in the scene [6][7][8][9].…”

“…Doing so reduces the time to find objects [5][6][7][8]. Influential factors include the pre-existing representations of scene structure and the 'fit' of the object to semantics of the scene and to locations in the scene [6][7][8][9]. Importantly, the scene-gist information provided by a preview can guide initial eye movements but has little impact on targeting of later eye fixations [10].…”