Change blindness in driving scenes

One of the key perceptual errors that contributes to accidents on the road is ‘looking but failing to see’. Though this has previously been attributed to failures of attention or time gaps, the recent change blindness literature suggests another alternative. Researchers have proposed that we have a poor memory for the visual world, and as such, participants find it very hard to notice a change between two successive pictures providing the transients that normally catch attention are masked. Such masking can occur naturally due to blinks and saccadic suppression. It is suggested that these effects may contribute to accident liability. An experiment was undertaken to test the application of the change blindness paradigm to the driving domain. It was predicted that experienced drivers may have greater visual persistence for changed targets in a road scene provided they are relevant to a driver’s parsing of the road (i.e. if the targets are potential hazards such as pedestrians, rather than changes in background scenery). The experiment required drivers and non-drivers to view a complex driving-related visual scene that was constantly interrupted by a flash once per second. During the flashes one item in the scene was changed. This target was manipulated according to location and semantic relevance. Results showed an interaction between central and peripheral items with semantic relevance. Participants found it hard to detect central items that were inconsequential, relative to other classifications of targets. No effect of experience was noted. The results are discussed in relation to the general theoretical literature and their potential applications to the driving domain.     

My shadow,myself: Cast-body shadows are embodied

Objects that serve as extensions of the body can produce a sensation of embodiment, feeling as if they are a part of us. We investigated the characteristics that drive an object’s embodiment, examining whether cast-body shadows, a purely visual stimulus, are embodied. Tools are represented as an extension of the body when they enable observers to interact with distant targets, perceptually distorting space. We examined whether perceptual distortion would also result from exposure to cast-body shadows in two separate distance estimation perceptual matching tasks. If observers represent cast-body shadows as extensions of their bodies, then when these shadows extend toward a target, it should appear closer than when no shadow is present (Experiment 1). This effect should not occur when a non-cast-body shadow is cast toward a target (Experiment 2). We found perceptual distortions in both cast-body shadow and tool-use conditions, but not in our non-cast-body shadow condition. These results suggest that, although cast-body shadows do not enable interaction with objects or provide direct tactile feedback, observers nonetheless represent their shadows as if they were a part of them.     

The shadow knows: a primer on the informational structure of cast shadows

I list relevant properties of cast shadows that the visual system could exploit (although it may not exploit). The study concerns the various types of information that can be extracted from shadows by systems that are significantly like ours (including embodied artificial systems) in an environment that is significantly like ours.     

Why cast shadows are expendable: insensitivity of human observers and the inherent ambiguity of cast shadows in pictorial art

The kinds of visual cues artists choose to use or not use in their work can offer insight into perceptual processes. On the basis of the observed paucity of the use of cast shadow in pictorial art, we hypothesized that cast shadows might be relatively expendable as pictorial cues. In this study, we investigated two potential reasons for this expendability: first, viewers might be insensitive to much of the information that cast shadows provide; and, second, ambiguities about what is shadow and what is pigment can often be resolved only through motion-something that static media are ill-equipped to deal with. In experiment 1, we used a visual-search paradigm in which viewers had to determine if there were odd cast shadows in sets of 4, 8, 16, and 32 objects. In experiment 2, viewers had to discriminate between shadow/pigment ambiguities in both still and moving images. Our results demonstrate that viewers are neither particularly sensitive to static cast-shadow incongruities, nor are they able to disambiguate cast shadow from pigment without continuous motion information. Taken together, these results may help explain why cast shadows are relatively rare in static pictorial work.     

Use of change blindness to measure different abilities to detect relevant changes in natural driving scenes

Drivers have to focus their attention on a danger to become aware of it. Change blindness paradigms are therefore relevant to studying the ability to detect danger. However, research has not yet focused on the role of two essential factors in guiding drivers’ attention: driving experience and the specific needs for performing a manoeuvre. Based on a previous analysis of real accident situations, we used a one-shot paradigm with static scenes to test observers’ ability to detect various changes as a function of their driving experience, the manoeuvre envisaged and the environmental context. The results showed that change detection depends greatly on driving experience when planning to cross a junction or to turn left, while it depends more on the environmental setting and task complexity when seeking a direction. The results were not conclusive, however, in explaining how drivers failed to notice that the vehicle ahead of them was turning when they considered an overtaking manoeuvre. We discuss the contributions of our research in relation to the possibilities of using change blindness as a measurement tool in studies on automobile driving.     

Even in correctable search, some types of rare targets are frequently missed

Socially important visual search tasks, such as airport baggage screening and tumor detection, place observers in situations where the targets are rare and the consequences of failed detection are substantial. Recent laboratory studies have demonstrated that low target prevalence yields substantially higher miss errors than do high-prevalence conditions, in which the same targets appear frequently (Wolfe, Horowitz, & Kenner, 2005; Wolfe et al., 2007). Under some circumstances, this \ldprevalence effect\rd can be eliminated simply by allowing observers to correct their last response (Fleck & Mitroff, 2007). However, in three experiments involving search of realistic X-ray luggage images, we found that the prevalence effect is eliminated neither by giving observers the choice to correct a previous response nor by requiring observers to confirm their responses. This prevalence effect, obtained when no trial-by-trial feedback was given, was smaller than the effect obtained when observers searched through the same stimuli but were given trial-by-trial feedback about accuracy. We suggest that low prevalence puts pressure on observers in any search task, and that the diverse symptoms of that pressure manifest themselves differently in different situations. In some relatively simple search tasks, misses may result from motor or response errors. In other, more complex tasks, shifts in decision criteria appear to be an important contributor.     

Changing scenes: memory for naturalistic events following change blindness

Research on scene perception indicates that viewers often fail to detect large changes to scene regions when these changes occur during a visual disruption such as a saccade or a movie cut. In two experiments, we examined whether this relative inability to detect changes would produce systematic biases in event memory. In Experiment 1, participants decided whether two successively presented images were the same or different, followed by a memory task, in which they recalled the content of the viewed scene. In Experiment 2, participants viewed a short video, in which an actor carried out a series of daily activities, and central scenes' attributes were changed during a movie cut. A high degree of change blindness was observed in both experiments, and these effects were related to scene complexity (Experiment 1) and level of retrieval support (Experiment 2). Most important, participants reported the changed, rather than the initial, event attributes following a failure in change detection. These findings suggest that attentional limitations during encoding contribute to biases in episodic memory.     

Not all missed opportunities cast the same shadow: Voluntary inaction reduces the “sting” of missed sales promotions

How do consumers who miss extremely attractive sales promotions respond to merely attractive opportunities they later encounter when prices return to higher levels? The literature on inaction inertia suggests that the more attractive the missed opportunity, the less likely a consumer is to accept the subsequently encountered inferior opportunity, indicating that consumers may stay undecided. Thus, consumers are believed to be negatively influenced by the shadow of the attractive opportunities they missed. This has adverse consequences for both consumer and firm welfare. Yet, we sometimes do see consumers buying even after missing a sale. We draw from the literature on regret and personal responsibility to hypothesize the conditions that would allow the consumer to remain uninfluenced by the attractiveness of the missed opportunity. In three studies, we find support for the idea that personal responsibility for missing the first opportunity allows consumers to be less influenced by its attractiveness when they see a second inferior opportunity compared to conditions in which they were not personally responsible for missing the first opportunity; this bodes well both for consumer and marketer welfare.     

Memory for object position in natural scenes

Memory for the positions of objects in natural scenes was investigated. Participants viewed an image of a real-world scene (preview scene), followed by a target object in isolation (target probe), followed by a blank screen with a mouse cursor. Participants estimated the position of the target using the mouse. Three conditions were compared. In the target present preview condition, the target object was present in the scene preview. In the target absent preview condition, the target object not present in the scene preview. In the no preview condition, no preview scene was displayed. Localization accuracy in the target present preview condition was reliably higher than that in the target absent preview condition, which was reliably higher than localization accuracy in the no preview condition. These data demonstrate that participants can remember both the spatial context of a scene and the specific positions of local objects.     

Iconic memory for natural scenes: Evidence using a modified change-detection procedure

ABSTRACT

Change blindness for the contents of natural scenes suggests that only items that are attended while the scene is still visible are stored, leading some to characterize our visual experiences as sparse. Experiments on iconic memory for arrays of discrete symbols or objects, however, indicate observers have access to more visual information for at least several hundred milliseconds at offset of a display. In the experiment presented here, we demonstrate an iconic memory for complex natural or real-world scenes. Using a modified change detection task in which to-be changed objects are cued at offset of the scene, we show that more information from a natural scene is briefly stored than change blindness predicts and more than is contained in visual short-term memory. In our experiment, a cue appearing 0, 300, or 1000?msec after offset of the pre-change scene or at onset of the second scene presentation (a Post Cue) directed attention to the location of a possible change. Compared to a no-cue condition, subjects were significantly better at detecting changes and identifying what changed in the cue condition, with the cue having a diminishing effect as a function of time and no effect when its onset coincided with that of the second scene presentation. The results suggest that an iconic memory of a natural scene exists for at least 1000?msec after scene offset, from which subjects can access the identity of items in the pre-change scene. This implies that change blindness underestimates the amount of information available to the visual system from a brief glance of a natural scene.     

Fixation-dependent memory for natural scenes: an experimental test of scanpath theory

Many modern theories propose that perceptual information is represented by the sensorimotor activity elicited by the original stimulus. Scanpath theory (Noton & Stark, 1971) predicts that reinstating a sequence of eye fixations will help an observer recognize a previously seen image. However, the only studies to investigate this are correlational ones based on calculating scanpath similarity. We therefore describe a series of 5 experiments that constrain the fixations during encoding or recognition of images in order to manipulate scanpath similarity. Participants encoded a set of images and later had to recognize those that they had seen. They spontaneously selected regions that they had fixated during encoding (Experiment 1), and this was a predictor of recognition accuracy. Yoking the parts of the image available at recognition to the encoded scanpath led to better memory performance than randomly selected image regions (Experiment 2), and this could not be explained by the spatial distribution of locations (Experiment 3). However, there was no recognition advantage for re-viewing one's own fixations versus someone else's (Experiment 4) or for retaining their serial order (Experiment 5). Therefore, although it is beneficial to look at encoded regions, there is no evidence that scanpaths are stored or that scanpath recapitulation is functional in scene memory. This paradigm provides a controlled way of studying the integration of scene content, spatial structure, and oculomotor signals, with consequences for the perception, representation, and retrieval of visual information.     

Change blindness blindness: beliefs about the roles of intention and scene complexity in change detection

Observers have difficulty detecting visual changes. However, they are unaware of this inability, suggesting that people do not have an accurate understanding of visual processes. We explored whether this error is related to participants' beliefs about the roles of intention and scene complexity in detecting changes. In Experiment 1 participants had a higher failure rate for detecting changes in an incidental change detection task than an intentional change detection task. This effect of intention was greatest for complex scenes. However, participants predicted equal levels of change detection for both types of changes across scene complexity. In Experiment 2, emphasizing the differences between intentional and incidental tasks allowed participants to make predictions that were less inaccurate. In Experiment 3, using more sensitive measures and accounting for individual differences did not further improve predictions. These findings suggest that adults do not fully understand the role of intention and scene complexity in change detection.     

Object-position binding in visual memory for natural scenes and object arrays

Nine experiments examined the means by which visual memory for individual objects is structured into a larger representation of a scene. Participants viewed images of natural scenes or object arrays in a change detection task requiring memory for the visual form of a single target object. In the test image, 2 properties of the stimulus were independently manipulated: the position of the target object and the spatial properties of the larger scene or array context. Memory performance was higher when the target object position remained the same from study to test. This same-position advantage was reduced or eliminated following contextual changes that disrupted the relative spatial relationships among contextual objects (context deletion, scrambling, and binding change) but was preserved following contextual change that did not disrupt relative spatial relationships (translation). Thus, episodic scene representations are formed through the binding of objects to scene locations, and object position is defined relative to a larger spatial representation coding the relative locations of contextual objects.     

The incongruency advantage for environmental sounds presented in natural auditory scenes

The effect of context on the identification of common environmental sounds (e.g., dogs barking or cars honking) was tested by embedding them in familiar auditory background scenes (street ambience, restaurants). Initial results with subjects trained on both the scenes and the sounds to be identified showed a significant advantage of about five percentage points better accuracy for sounds that were contextually incongruous with the background scene (e.g., a rooster crowing in a hospital). Further studies with naive (untrained) listeners showed that this incongruency advantage (IA) is level-dependent: there is no advantage for incongruent sounds lower than a Sound/Scene ratio (So/Sc) of -7.5 dB, but there is about five percentage points better accuracy for sounds with greater So/Sc. Testing a new group of trained listeners on a larger corpus of sounds and scenes showed that the effect is robust and not confined to a specific stimulus set. Modeling using spectral-temporal measures showed that neither analyses based on acoustic features, nor semantic assessments of sound-scene congruency can account for this difference, indicating the IA is a complex effect, possibly arising from the sensitivity of the auditory system to new and unexpected events, under particular listening conditions.     

Object identification in context: the visual processing of natural scenes.

When we view a natural visual scene, we seem able to determine effortlessly the scene's semantic category, constituent objects, and spatial relations. How do we accomplish this visual-cognitive feat? The commonly held explanation is known as the schema hypothesis, according to which a visual scene is rapidly identified as a member of a semantic category, and predictions generated from the scene category are then used to aid subsequent object identification. In this paper I will first outline and offer a critique of the evidence that has been taken to support the schema hypothesis. I will then offer an alternative framework for understanding scene processing, which I will call the local-processing hypothesis. This hypothesis assumes a modular, informationally-encapsulated architecture, and explicitly includes the role of covert visual attention in scene processing.     

Hemispheric specialization for spatial frequency processing in the analysis of natural scenes

Experimental data coming from visual cognitive sciences suggest that visual analysis starts with a parallel extraction of different visual attributes at different scales/frequencies. Neuropsychological and functional imagery data have suggested that each hemisphere (at the level of temporo-parietal junctions-TPJ) could play a key role in spatial frequency processing: The right TPJ should predominantly be involved in low spatial frequency (LFs) analysis and the left TPJ in high spatial frequency (HFs) analysis. Nevertheless, this functional hypothesis had been inferred from data obtained when using the hierarchical form paradigm, without any explicit spatial frequency manipulation per se. The aims of this research are (i) to investigate, in healthy subjects, the hemispheric asymmetry hypothesis with an explicit manipulation of spatial frequencies of natural scenes and (ii) to examine whether the 'precedence effect' (the relative rapidity of LFs and HFs processing) depends on the visual field of scene presentation or not. For this purpose, participants were to identify either non-filtered or LFs and HFs filtered target scene displayed either in the left, central, or right visual field. Results showed a hemispheric specialization for spatial frequency processing and different 'precedence effects' depending on the visual field of presentation.     

Face detection differs from categorization: Evidence from visual search in natural scenes

In this study, we examined whether the detection of frontal, ¾, and profile face views differs from their categorization as faces. In Experiment 1, we compared three tasks that required observers to determine the presence or absence of a face, but varied in the extents to which participants had to search for the faces in simple displays and in small or large scenes to make this decision. Performance was equivalent for all of the face views in simple displays and small scenes, but it was notably slower for profile views when this required the search for faces in extended scene displays. This search effect was confirmed in Experiment 2, in which we compared observers’ eye movements with their response times to faces in visual scenes. These results demonstrate that the categorization of faces at fixation is dissociable from the detection of faces in space. Consequently, we suggest that face detection should be studied with extended visual displays, such as natural scenes.