Complex Visual Environments Research Articles

Colour and luminance contrasts predict the human detection of natural stimuli in complex visual environments.

Much of what we know about human colour perception has come from psychophysical studies conducted in tightly-controlled laboratory settings. An enduring challenge, however, lies in extrapolating this knowledge to the noisy conditions that characterize our actual visual experience. Here we combine statistical models of visual perception with empirical data to explore how chromatic (hue/saturation) and achromatic (luminant) information underpins the detection and classification of stimuli in a complex forest environment. The data best support a simple linear model of stimulus detection as an additive function of both luminance and saturation contrast. The strength of each predictor is modest yet consistent across gross variation in viewing conditions, which accords with expectation based upon general primate psychophysics. Our findings implicate simple visual cues in the guidance of perception amidst natural noise, and highlight the potential for informing human vision via a fusion between psychophysical modelling and real-world behaviour.

Multiple visual objects are sampled sequentially.

When acting in a complex visual environment, it is essential to be able to flexibly allocate attention to parts of the visual scene that may contain goal-relevant information. The paper by Jia et al. provides novel evidence that our brains sequentially sample different objects in a visual scene. The results were obtained using “temporal response functions,” in which unique electroencephalographic (EEG) signals corresponding to the processing of 2 continuously presented objects were isolated in an object-specific way. These response functions were dominated by 10-Hz alpha-band activity. Crucially, the different objects were sequentially sampled at a rate of about 2 Hz. These findings provide important neurophysiological insights into how our visual system operates in complex environments.

Odor source localization in complex visual environments by fruit flies.

Flying insects routinely forage in complex and cluttered sensory environments. Their search for a food or a pheromone source typically begins with a whiff of odor, which triggers a flight response, eventually bringing the insect near the odor source. However, pinpointing the precise location of an odor source requires use of both visual and olfactory modalities, aided by odor plumes. Here, we investigated odor-tracking behavior in fruit flies (Drosophila melanogaster) presented with low- or high-contrast visual landmarks, either paired with or separate from an attractive odor cue. These experiments were conducted either in a gentle air stream which generated laminar odor plumes or in still air in which odor dissipates uniformly in all directions. Trajectories of flies revealed several novel features of their odor-tracking behavior in addition to those previously documented. First, in both moving and still air, odor-seeking flies rely on the co-occurrence of visual landmarks with olfactory cues to guide them to odorant objects. Second, flies abruptly decelerate upon encountering an odor plume, thereafter steering towards the nearest visual objects that had no inherent salience in the absence of odor. Thus, interception of an attractive odor increases their salience to nearby high-contrast visual landmarks. Third, flies adopt distinct odor-tracking strategies during flight in moving versus still air. Whereas they weave in and out of plumes towards an odor source in airflow, their approach is more incremental in still air. Both strategies are robust and flexible, and enable flies to reliably find odor sources under diverse visual and airflow environments.

Visual statistical learning at basic and subordinate category levels in real-world images

Visual statistical learning (VSL) has been proposed as a powerful mechanism underlying the striking ability of human observers to handle complex visual environments. Previous studies have shown that VSL can occur when statistical information is embedded at multiple levels of abstraction, such as at semantically different category levels. In the present study, we further examined whether statistical regularities at a basic category level (e.g., a regular sequence of a bird, then a car, and then a dog) could influence the ability to extract statistical regularities at the subordinate level (e.g., a regular sequence of a parrot, then a sports car, and then an Eskimo dog). In the familiarization phase, participants were exposed to a stream of real-world images whose semantic categories had temporal regularities. Importantly, the temporal regularities existed at both the basic and subordinate levels, or the regularities existed at only the subordinate level, depending on the experimental condition. After completing the familiarization, participants performed a surprise two-alternative forced choice (2AFC) task for a familiarity judgment between two triplets in which the temporal regularities were either preserved or not preserved. Our results showed that the existence of statistical regularities at the basic level did not influence VSL at the subordinate level. The subsequent experiments showed these results consistently even when the basic-level categories had to be explicitly recognized and when the stimuli were not easily categorized at their subordinate level. Our results suggest that VSL is constrained to learn a particular level of patterns when patterns are presented across multiple levels.

Contour junctions underlie neural representations of scene categories in high-level human visual cortex

Humans efficiently grasp complex visual environments, making highly consistent judgments of entry-level category despite their high variability in visual appearance. How does the human brain arrive at the invariant neural representations underlying categorization of real-world environments? We here show that the neural representation of visual environments in scene-selective human visual cortex relies on statistics of contour junctions, which provide cues for the three-dimensional arrangement of surfaces in a scene. We manipulated line drawings of real-world environments such that statistics of contour orientations or junctions were disrupted. Manipulated and intact line drawings were presented to participants in an fMRI experiment. Scene categories were decoded from neural activity patterns in the parahippocampal place area (PPA), the occipital place area (OPA) and other visual brain regions. Disruption of junctions but not orientations led to a drastic decrease in decoding accuracy in the PPA and OPA, indicating the reliance of these areas on intact junction statistics. Accuracy of decoding from early visual cortex, on the other hand, was unaffected by either image manipulation. We further show that the correlation of error patterns between decoding from the scene-selective brain areas and behavioral experiments is contingent on intact contour junctions. Finally, a searchlight analysis exposes the reliance of visually active brain regions on different sets of contour properties. Statistics of contour length and curvature dominate neural representations of scene categories in early visual areas and contour junctions in high-level scene-selective brain regions.

A robust tracking method with adaptive local spatial sparse representation

In this paper, a robust visual tracking method is proposed based on local spatial sparse representation. In the proposed approach, the learned target template is sparsely and compactly expressed by forming local spatial and trivial samples dynamically. An adaptive multiple subspaces appearance model is developed to describe the target appearance and construct the candidate target templates during the tracking process. An effective selection strategy is then employed to select the optimal sparse solution and locate the target accurately in the next frame. The experimental results have demonstrated that our method can perform well in the complex and noisy visual environment, such as heavy occlusions, dramatic illumination changes, and large pose variations in the video. Copyright © 2015 John Wiley & Sons, Ltd.

Decoding covert shifts of attention induced by ambiguous visuospatial cues.

Simple and unambiguous visual cues (e.g., an arrow) can be used to trigger covert shifts of visual attention away from the center of gaze. The processing of visual stimuli is enhanced at the attended location. Covert shifts of attention modulate the power of cerebral oscillations in the alpha band over parietal and occipital regions. These modulations are sufficiently robust to be decoded on a single trial basis from electroencephalography (EEG) signals. It is often assumed that covert attention shifts are under voluntary control, and that they also occur in more natural and complex environments, but there is no direct evidence to support this assumption. We address this important issue by using random-dot stimuli to cue one of two opposite locations, where a visual target is presented. We contrast two conditions, one in which the random-dot motion is predictive of the target location, and the other, in which it provides ambiguous information. Behavioral results show attention shifts in anticipation of the visual target, in both conditions. In addition, using the common spatial patterns (CSPs) algorithm, we extract EEG power features in the alpha-band (around 10 Hz) that best discriminate the attended location in single trials. We obtain a significant decoding accuracy in 7/10 subjects using a cross-validation procedure applied in the predictive condition. Interestingly, similar accuracy (significant in 5/10 subjects) is obtained when the CSPs trained in the predictive condition are tested in the ambiguous condition. In agreement with this result, we find that the CSPs show very similar topographies in both conditions. These results shed a new light on the behavioral and EEG correlates of visuospatial attention in complex visual environments. This study demonstrates that alpha-power features could be used in brain–computer interfaces to decode covert attention shifts in an environment containing ambiguous spatial information.

The visual ecology of directed aerial descent in first-instar nymphs of the stick insectExtatosoma tiaratum

Many wingless insects perform directed aerial descent (DAD) to return to vegetational structures after falling. Given the complex visual environment and spatial structures of tree canopies, those visual signals used as directional cues are not fully understood. Here, we address the role of visual contrast for DAD in newly hatched nymphs of the stick insect Extatosoma tiaratum under controlled laboratory conditions. Landing preferences of gliding E. tiaratum in various visual environments were studied. We used a single vertical stripe defined by variable contrast edges to test the use of contrast consistency and sharpness. We also used aggregate patterns to examine the effects of target size and the effectiveness of luminance contrast and chromatic contrast. E. tiaratum nymphs were attracted to single stripes with well-defined edges, and particularly favored narrow dark targets. The directionality and accuracy of landing were dependent on target size. Lastly, luminance contrasts were more effective in attracting landings than were chromatic contrasts. Visual contrasts are therefore used as spatial references for landing behavior in DAD. These behaviors may enable nymphs to quickly locate dark or shaded sides of vertically oriented vegetation in natural habitats.

Transsaccadic processing: stability, integration, and the potential role of remapping.

While our frequent saccades allow us to sample the complex visual environment in a highly efficient manner, they also raise certain challenges for interpreting and acting upon visual input. In the present, selective review, we discuss key findings from the domains of cognitive psychology, visual perception, and neuroscience concerning two such challenges: (1) maintaining the phenomenal experience of visual stability despite our rapidly shifting gaze, and (2) integrating visual information across discrete fixations. In the first two sections of the article, we focus primarily on behavioral findings. Next, we examine the possibility that a neural phenomenon known as predictive remapping may provide an explanation for aspects of transsaccadic processing. In this section of the article, we delineate and critically evaluate multiple proposals about the potential role of predictive remapping in light of both theoretical principles and empirical findings.

Bumblebees Learn Polarization Patterns

SummaryForaging insect pollinators such as bees must find and identify flowers in a complex visual environment. Bees use skylight polarization patterns for navigation [1–3], a capacity mediated by the polarization-sensitive dorsal rim area (DRA) of their eye [4, 5]. While other insects use polarization sensitivity to identify appropriate habitats [6], oviposition sites, and food sources [7], to date no nonnavigational functions of polarization vision have been identified in bees. Here we investigated the ability of bumblebees (Bombus terrestris) to learn polarization patterns on artificial “flowers” in order to obtain a food reward. We show that foraging bumblebees can learn to discriminate between two differently polarized targets, but only when the target artificial “flower” is viewed from below. A context for these results is provided by polarization imaging of bee-pollinated flowers, revealing the potential for polarization patterns in real flowers. Bees may therefore have the ability to use polarization vision, possibly mediated by their polarization-sensitive DRA, both for navigation and to learn polarization patterns on flowers, the latter being the first nonnavigational function for bee polarization vision to be identified.

Selective attention in the honeybee optic lobes precedes behavioral choices

Attention allows animals to respond selectively to competing stimuli, enabling some stimuli to evoke a behavioral response while others are ignored. How the brain does this remains mysterious, although it is increasingly evident that even animals with the smallest brains display this capacity. For example, insects respond selectively to salient visual stimuli, but it is unknown where such selectivity occurs in the insect brain, or whether neural correlates of attention might predict the visual choices made by an insect. Here, we investigate neural correlates of visual attention in behaving honeybees (Apis mellifera). Using a closed-loop paradigm that allows tethered, walking bees to actively control visual objects in a virtual reality arena, we show that behavioral fixation increases neuronal responses to flickering, frequency-tagged stimuli. Attention-like effects were reduced in the optic lobes during replay of the same visual sequences, when bees were not able to control the visual displays. When bees were presented with competing frequency-tagged visual stimuli, selectivity in the medulla (an optic ganglion) preceded behavioral selection of a stimulus, suggesting that modulation of early visual processing centers precedes eventual behavioral choices made by these insects.

Far Field Effects of Cortical tDCS in the Cerebellum may Contribute to its Effects on Learning.

In previous studies, we have used imaging and tDCS to examine the brain basis of learning to detect objects hidden in a complex visual environment. Learning this task resulted in increased activity in right lateral frontal and parietal cortices, and reduced activity in occipital-temporal regions bilaterally. We hypothesized that anodal and cathodal stimulation over regions showing increased and decreased activity with learning, respectively, would both result in increased performance. In agreement with this, we found that both anodal F10 or P4 vs.

Mechanisms Underlying the Attentional Guidance form Working Memory Representations

People had to select specific information from complex visual scene every time to avoid overloading the limited information processing system.When extracting relevant information from a complex visual environment,working memory(WM) representation would bias attention to stimuli that were the same or the similar to the search target,as to optimize target selection while ignoring the distractors.The present study systematically reviewed the emerging evidences on the attentional guidance from WM representations.After discussing the following three controversial issues: Could WM guide attention automatically? Could verbal WM guide attention? And did WM always guide attention? We concluded that the WM could top-down guide attention,but it was subjected to some constrains.

Back to the Future: Warlpiri Encounters with Drawings, Country and Others in the Digital Age

Since the early 1900s, Warlpiri people living in the central desert region of Australia have experienced an intense process of adaption to the changing circumstances of postcolonial life. From the earliest days, their encounters with kardiya, non-Aboriginal people, have been mediated by diverse visual technologies that have, over time, become integral to the ways Warlpiri orient themselves to each other and their wider world. In this paper I trace the key elements of the complex visual environment that has emerged from this history of mediation. The central part of the paper considers events around the repatriation to Warlpiri communities in 2011 of a collection of drawings made in the 1950s by their forebears. In responses to a medium that once was new but now is old, several points of interest emerge, among them a clear sense of a hierarchy of value Warlpiri apply to modes of visual communication. In the context of the return of the drawings, the significance Warlpiri ascribe to other visual media comes to the fore. I consider some of the ways visual forms are deployed in support of public projections of cultural identity on the one hand and everyday modes of expression and address on the other. The paper's central argument is that contemporary Warlpiri attitudes to images – whether they be drawn, painted or broadcast – reveal the complex postcolonial workings of mimetic desire.

ERP evidence for human early visual sensitivity to co-linearity compared to co-circularity

Our complex visual environment is constrained by natural geometric regularities, including spatiotemporal regularity, co-linearity and co-circularity. To investigate human visual processing associated with these regularities we directly compared the neural processes in encoding dynamic co-linearity and co-circularity using event-related potentials (ERPs). By recording ERPs to a target bar presented alone (no context) or in a dynamic sequence of bars following a co-linear or co-circular path, we observed earlier ERPs to targets embedded in co-linear sequence at early (66ms) and later stages (197ms) of post-target processing. In contrast, targets in co-circular sequence only modulated ERPs at later stages of processing. It is proposed that early visual processing may have adapted to efficiently process co-linearity to improve target identification, whereas sensitivity to co-circularity does not occur until later stages of processing. These results have significant impact for understanding brain-behaviour relationships when processing natural geometric regularities.

Object recognition in congruent and incongruent natural scenes: A life-span study

Efficient processing of our complex visual environment is essential and many daily visual tasks rely on accurate and fast object recognition. It is therefore important to evaluate how object recognition performance evolves during the course of adulthood. Surprisingly, this ability has not yet been investigated in the aged population, although several neuroimaging studies have reported altered activity in high-level visual ventral regions when elderly subjects process natural stimuli. In the present study, color photographs of various objects embedded in contextual scenes were used to assess object categorization performance in 97 participants aged from 20 to 91. Objects were either animals or pieces of furniture, embedded in either congruent or incongruent contexts. In every age group, subjects showed reduced categorization performance, both in terms of accuracy and speed, when objects were seen in incongruent vs. congruent contexts. In subjects over 60 years old, object categorization was greatly slowed down when compared to young and middle-aged subjects. Moreover, subjects over 75 years old evidenced a significant decrease in categorization accuracy when objects were seen in incongruent contexts. This indicates that incongruence of the scene may be particularly disturbing in late adulthood, therefore impairing object recognition. Our results suggest that daily visual processing of complex natural environments may be less efficient with age, which might impact performance in everyday visual tasks.

Redundant Neural Vision Systems—Competing for Collision Recognition Roles

Ability to detect collisions is vital for future robots that interact with humans in complex visual environments. Lobula giant movement detectors (LGMD) and directional selective neurons (DSNs) are two types of identified neurons found in the visual pathways of insects such as locusts. Recent modeling studies showed that the LGMD or grouped DSNs could each be tuned for collision recognition. In both biological and artificial vision systems, however, which one should play the collision recognition role and the way the two types of specialized visual neurons could be functioning together are not clear. In this modeling study, we compared the competence of the LGMD and the DSNs, and also investigate the cooperation of the two neural vision systems for collision recognition via artificial evolution. We implemented three types of collision recognition neural subsystems - the LGMD, the DSNs and a hybrid system which combines the LGMD and the DSNs subsystems together, in each individual agent. A switch gene determines which of the three redundant neural subsystems plays the collision recognition role. We found that, in both robotics and driving environments, the LGMD was able to build up its ability for collision recognition quickly and robustly therefore reducing the chance of other types of neural networks to play the same role. The results suggest that the LGMD neural network could be the ideal model to be realized in hardware for collision recognition.

Postsynaptic organisations of directional selective visual neural networks for collision detection

In this paper, we studied the postsynaptic organisations of directional selective visual neurons for collision detection. Directional selective neurons can extract different directional visual motion cues fast and reliably by allowing inhibition spreads to further layers in specific directions with one or several time steps delay. Whether these directional selective neurons can be easily organised for other specific visual tasks is not known. Taking collision detection as the primary visual task, we investigated the postsynaptic organisations of these directional selective neurons through evolutionary processes. The evolved postsynaptic organisations demonstrated robust properties in detecting imminent collisions in complex visual environments with many of which achieved 94% success rate after evolution suggesting active roles in collision detection directional selective neurons and its postsynaptic organisations can play.

Aurally Aided Visual Threat Acquisition in a Virtual Urban Battlespace

Many of today’s military operations take place in large urban environments, which present unique challenges due to limited line of sight and increased concealment for enemy forces. This experiment evaluated the potential of a 3D audio display to aid threat detection and localization in a complex visual environment. Subjects rode as part of a convoy through a simulated city where they encountered snipers surrounded by distracting personnel. We compared 3D audio cues to verbal descriptions of the sniper’s location and to simple audio warnings of the presence of a sniper. Consistent with past research, subjects located the sniper more quickly in the 3D audio condition compared to both the semantic description and simple warning conditions. The 3D audio display was faster than the other displays across all azimuths but, in contrast to previous findings, the advantage did not increase consistently with increasing azimuth.

Working towards a test for screening visual skills in a complex visual environment

We developed a test for screening visual skills under similar conditions as at workplace. The test was administered to 207 participants, recruited in the working population. Six-digit numbers were super-imposed on a video of a drive and presented for 300 ms in the central visual field and in the periphery. Participants reported whether or not the integer ‘3’ was included in the numbers. Normative data for the test were computed using 150 data sets of participants reporting not to take drugs and with an age ranging between 15 y and 67 y. Participants performed better in the central visual field than in periphery. The test could successfully be completed by participants of all ages without the need of adjustment of settings. Comparison of 46 data sets recorded in participants taking drugs (30 y – 68 y) with age matched participants not taking drugs demonstrates an additional potential application of the test. Practitioner Summary: Developed test delivers comparative estimates of visual skills within short time and appears as valuable and cheap complementation to current testing procedures in industrial practice. The test can be used in a preventive and in educative manner to monitor effects of factors like fatigue, sleep deprivation or drug consumption.