Bayesian multi-level modelling for predicting single and double feature visual search.

Performance in visual search tasks is frequently summarised by "search slopes" - the additional cost in reaction time for each additional distractor. While search tasks with a shallow search slopes are termed efficient (pop-out, parallel, feature), there is no clear dichotomy between efficient and inefficient (serial, conjunction) search. Indeed, a range of search slopes are observed in empirical data. The Target Contrast Signal (TCS) Theory is a rare example of quantitative model that attempts to predict search slopes for efficient visual search. One study using the TCS framework has shown that the search slope in a double-feature search (where the target differs in both colour and shape from the distractors) can be estimated from the slopes of the associated single-feature searches. This estimation is done using a contrast combination model, and a collinear contrast integration model was shown to outperform other options. In our work, we extend TCS to a Bayesian multi-level framework. We investigate modelling using normal and shifted-lognormal distributions, and show that the latter allows for a better fit to previously published data. We run a new fully within-subjects experiment to attempt to replicate the key original findings, and show that overall, TCS does a good job of predicting the data. However, we do not replicate the finding that the collinear combination model outperforms the other contrast combination models, instead finding that it may be difficult to conclusively distinguish between them.

Varying benefits of generalist and specialist camouflage in two versus four background environments.

Background-matching camouflage is a well-established strategy to reduce detection, but implementing this on heterogeneous backgrounds is challenging. For prey with fixed color patterns, solutions include specializing on a particular visual microhabitat, or adopting a compromise or generalist appearance, matching multiple backgrounds less well. Existing studies suggest both approaches can succeed, but most consider relatively simple scenarios, where artificial prey appear against two backgrounds differing in a single visual characteristic. Here, we used computer-based search tasks with human participants to test the relative benefits of specializing and generalizing for complex targets, displayed on either two or four types of naturalistic backgrounds. Across two background types, specialization was beneficial on average. However, the success of this strategy varied with search duration, such that generalist targets could outperform specialists over short search durations due to the presence of poorly matched specialists. Over longer searches, the remaining well-matched specialists had greater success than generalists, leading to an overall benefit of specialization at longer search durations. Against four different backgrounds, the initial cost to specialization was greater, so specialists and generalists ultimately experienced similar survival. Generalists performed better when their patterning was a compromise between backgrounds that were more similar to each other than when backgrounds were more different, with similarity in luminance more relevant than pattern differences. Time dependence in the relative success of these strategies suggests that predator search behavior may affect optimal camouflage in real-world situations.

Correction: Foraging as sampling without replacement: A Bayesian statistical model for estimating biases in target selection.

[This corrects the article DOI: 10.1371/journal.pcbi.1009813.].

A Bayesian Statistical Model Is Able to Predict Target-by-Target Selection Behaviour in a Human Foraging Task.

Foraging refers to search involving multiple targets or multiple types of targets, and as a model task has a long history in animal behaviour and human cognition research. Foraging behaviour is usually operationalized using summary statistics, such as average distance covered during target collection (the path length) and the frequency of switching between target types. We recently introduced an alternative approach, which is to model each instance of target selection as random selection without replacement. Our model produces estimates of a set of foraging biases, such as a bias to select closer targets or targets of a particular category. Here we apply this model to predict individual target selection events. We add a new start position bias to the model, and generate foraging paths using the parameters estimated from individual participants' pre-existing data. The model predicts which target the participant will select next with a range of accuracy from 43% to 69% across participants (chance is 11%). The model therefore explains a substantial proportion of foraging behaviour in this paradigm. The situations where the model makes errors reveal useful information to guide future research on those aspects of foraging that we have not yet explained.

Nest sanitation as an effective defence against brood parasitism.

Egg rejection is a crucial defence strategy against brood parasitism that requires the host to correctly recognise the foreign egg. Rejection behaviour has, thus, evolved in many hosts, facilitated by the visual differences between the parasitic and host eggs, and driving hosts to rely on colour and pattern cues. On the other hand, the need to recognise non-egg-shaped objects to carry out nest sanitation led birds to evolve the ability to discriminate and eject objects using mainly shape cues. However, little is known regarding the evolutionary significance of rejection behaviour in general and the cognitive processes underlying it. Here, we investigated the response of the barn swallow (Hirundo rustica) during pre-laying and laying stages to four objects types that differed in shape (eggs vs stars) and colour/pattern (mimetic vs non-mimetic) to investigate (1) what cognitive mechanisms are involved in object discrimination and (2) whether egg rejection is a direct defence against brood parasitism, or simply a product of nest sanitation. We found that swallows ejected stars more often than eggs in both stages, indicating that swallows possess a template for the shape of their eggs. Since the effect of colour/pattern on ejection decisions was minor, we suggest that barn swallows have not evolved a direct defence against brood parasitism but instead, egg ejection might be a product of their well-developed nest sanitation behaviour. Nonetheless, the fact that mimetic eggs were ejected especially in the pre-laying stage shows that nest sanitation could be an effective defence against poorly timed brood parasitism.

Foraging as sampling without replacement: A Bayesian statistical model for estimating biases in target selection.

Foraging entails finding multiple targets sequentially. In humans and other animals, a key observation has been a tendency to forage in 'runs' of the same target type. This tendency is context-sensitive, and in humans, it is strongest when the targets are difficult to distinguish from the distractors. Many important questions have yet to be addressed about this and other tendencies in human foraging, and a key limitation is a lack of precise measures of foraging behaviour. The standard measures tend to be run statistics, such as the maximum run length and the number of runs. But these measures are not only interdependent, they are also constrained by the number and distribution of targets, making it difficult to make inferences about the effects of these aspects of the environment on foraging. Moreover, run statistics are underspecified about the underlying cognitive processes determining foraging behaviour. We present an alternative approach: modelling foraging as a procedure of generative sampling without replacement, implemented in a Bayesian multilevel model. This allows us to break behaviour down into a number of biases that influence target selection, such as the proximity of targets and a bias for selecting targets in runs, in a way that is not dependent on the number of targets present. Our method thereby facilitates direct comparison of specific foraging tendencies between search environments that differ in theoretically important dimensions. We demonstrate the use of our model with simulation examples and re-analysis of existing data. We believe our model will provide deeper insights into visual foraging and provide a foundation for further modelling work in this area.

Generalist camouflage can be more successful than microhabitat specialisation in natural environments.

BACKGROUND: Crypsis by background-matching is a critical form of anti-predator defence for animals exposed to visual predators, but achieving effective camouflage in patchy and variable natural environments is not straightforward. To cope with heterogeneous backgrounds, animals could either specialise on particular microhabitat patches, appearing cryptic in some areas but mismatching others, or adopt a compromise strategy, providing partial matching across different patch types. Existing studies have tested the effectiveness of compromise strategies in only a limited set of circumstances, primarily with small targets varying in pattern, and usually in screen-based tasks. Here, we measured the detection risk associated with different background-matching strategies for relatively large targets, with human observers searching for them in natural scenes, and focusing on colour. Model prey were designed to either 'specialise' on the colour of common microhabitat patches, or 'generalise' by matching the average colour of the whole visual scenes. RESULTS: In both the field and an equivalent online computer-based search task, targets adopting the generalist strategy were more successful in evading detection than those matching microhabitat patches. This advantage occurred because, across all possible locations in these experiments, targets were typically viewed against a patchwork of different microhabitat areas; the putatively generalist targets were thus more similar on average to their various immediate surroundings than were the specialists. CONCLUSIONS: Demonstrating close agreement between the results of field and online search experiments provides useful validation of online citizen science methods commonly used to test principles of camouflage, at least for human observers. In finding a survival benefit to matching the average colour of the visual scenes in our chosen environment, our results highlight the importance of relative scales in determining optimal camouflage strategies, and suggest how compromise coloration can succeed in nature.

The evolution of patterning during movement in a large-scale citizen science game.

The motion dazzle hypothesis posits that high contrast geometric patterns can cause difficulties in tracking a moving target and has been argued to explain the patterning of animals such as zebras. Research to date has only tested a small number of patterns, offering equivocal support for the hypothesis. Here, we take a genetic programming approach to allow patterns to evolve based on their fitness (time taken to capture) and thus find the optimal strategy for providing protection when moving. Our 'Dazzle Bug' citizen science game tested over 1.5 million targets in a touch screen game at a popular visitor attraction. Surprisingly, we found that targets lost pattern elements during evolution and became closely background matching. Modelling results suggested that targets with lower motion energy were harder to catch. Our results indicate that low contrast, featureless targets offer the greatest protection against capture when in motion, challenging the motion dazzle hypothesis.

Population receptive field estimates for motion-defined stimuli.

The processing of motion changes throughout the visual hierarchy, from spatially restricted 'local motion' in early visual cortex to more complex large-field 'global motion' at later stages. Here we used functional magnetic resonance imaging (fMRI) to examine spatially selective responses in these areas related to the processing of random-dot stimuli defined by differences in motion. We used population receptive field (pRF) analyses to map retinotopic cortex using bar stimuli comprising coherently moving dots. In the first experiment, we used three separate background conditions: no background dots (dot-defined bar-only), dots moving coherently in the opposite direction to the bar (kinetic boundary) and dots moving incoherently in random directions (global motion). Clear retinotopic maps were obtained for the bar-only and kinetic-boundary conditions across visual areas V1-V3 and in higher dorsal areas. For the global-motion condition, retinotopic maps were much weaker in early areas and became clear only in higher areas, consistent with the emergence of global-motion processing throughout the visual hierarchy. However, in a second experiment we demonstrate that this pattern is not specific to motion-defined stimuli, with very similar results for a transparent-motion stimulus and a bar defined by a static low-level property (dot size) that should have driven responses particularly in V1. We further exclude explanations based on stimulus visibility by demonstrating that the observed differences in pRF properties do not follow the ability of observers to localise or attend to these bar elements. Rather, our findings indicate that dorsal extrastriate retinotopic maps may primarily be determined by the visibility of the neural responses to the bar relative to the background response (i.e. neural signal-to-noise ratios) and suggests that claims about stimulus selectivity from pRF experiments must be interpreted with caution.

AccessLab: Workshops to broaden access to scientific research.

AccessLabs are workshops with two simultaneous motivations, achieved through direct citizen-scientist pairings: (1) to decentralise research skills so that a broader range of people are able to access/use scientific research, and (2) to expose science researchers to the difficulties of using their research as an outsider, creating new open access advocates. Five trial AccessLabs have taken place for policy makers, media/journalists, marine sector participants, community groups, and artists. The act of pairing science academics with local community members helps build understanding and trust between groups at a time when this relationship appears to be under increasing threat from different political and economic currents in society. Here, we outline the workshop motivations, format, and evaluation, with the aim that others can build on the methods developed.

Dissociation between perception and smooth pursuit eye movements in speed judgments of moving Gabor targets.

The relationship between eye movements and subjective perception is still relatively poorly understood. In this study, participants tracked the movement of a Gabor patch and made perceptual judgments of its speed using a two-interval forced choice task. The Gabor patch could either have a static carrier or a carrier moving in the same or opposite direction as the overall envelope motion. We found that smooth pursuit speed was strongly affected by the internal motion of the Gabor carrier, with faster smooth pursuit being made to targets with internal motion in the same direction as overall motion compared to targets with internal motion in the opposite direction. However, we found that there were only small and highly variable differences in the perceptual speed judgments made simultaneously, and that these perceptual and smooth pursuit measures did not significantly correlate with each other. This contrasts with the number of catch-up saccades (saccades made in the direction of overall target motion), which was significantly correlated with the simultaneous perceptual judgments. There was also a significant correlation between perceptual judgments and the difference between the target and eye position immediately before a saccade. These results suggest that it is possible to see dissociations between vision and action in this task, and that the specific type of visual action studied may determine the relationship with perception.

Diverted by dazzle: perceived movement direction is biased by target pattern orientation.

'Motion dazzle' is the hypothesis that predators may misjudge the speed or direction of moving prey which have high-contrast patterning, such as stripes. However, there is currently little experimental evidence that such patterns cause visual illusions. Here, observers binocularly tracked a Gabor target, moving with a linear trajectory randomly chosen within 18° of the horizontal. This target then became occluded, and observers were asked to judge where they thought it would later cross a vertical line to the side. We found that internal motion of the stripes within the Gabor biased judgements as expected: Gabors with upwards internal stripe motion relative to the overall direction of motion were perceived to be crossing above Gabors with downwards internal stripe movement. However, surprisingly, we found a much stronger effect of the rigid pattern orientation. Patches with oblique stripes pointing upwards relative to the direction of motion were perceived to cross above patches with downward-pointing stripes. This effect occurred only at high speeds, suggesting that it may reflect an orientation-dependent effect in which spatial signals are used in direction judgements. These findings have implications for our understanding of motion dazzle mechanisms and how human motion and form processing interact.

Quantifying peripheral and foveal perceived differences in natural image patches to predict visual search performance.

Duncan and Humphreys (1989) identified two key factors that affected performance in a visual search task for a target among distractors. The first was the similarity of the target to distractors (TD), and the second was the similarity of distractors to each other (DD). Here we investigate if it is the perceived similarity in foveal or peripheral vision that determines performance. We studied search using stimuli made from patches cut from colored images of natural objects; differences between targets and their modified distractors were estimated using a ratings task peripherally and foveally. We used search conditions in which the targets and distractors were easy to distinguish both foveally and peripherally ("high" stimuli), in which they were difficult to distinguish both foveally and peripherally ("low"), and in which they were easy to distinguish foveally but difficult to distinguish peripherally ("metamers"). In the critical metameric condition, search slopes (change of search time with number of distractors) were similar to the "low" condition, indicating a key role for peripheral information in visual search as both conditions have low perceived similarity peripherally. Furthermore, in all conditions, search slope was well described quantitatively from peripheral TD and DD but not foveal. However, some features of search, such as error rates, do indicate roles for foveal vision too.

The role of stripe orientation in target capture success.

INTRODUCTION: 'Motion dazzle' refers to the hypothesis that high contrast patterns such as stripes and zigzags may have evolved in a wide range of animals as they make it difficult to judge the trajectory of an animal in motion. Despite recent research into this idea, it is still unclear to what extent stripes interfere with motion judgement and if effects are seen, what visual processes might underlie them. We use human participants performing a touch screen task in which they attempt to 'catch' moving targets in order to determine whether stripe orientation affects capture success, as previous research has suggested that different stripe orientations may be processed differently by the visual system. We also ask whether increasing the number of targets presented in a trial can affect capture success, as previous research has suggested that motion dazzle effects may be larger in groups. RESULTS: When single targets were presented sequentially within each trial, we find that perpendicular and oblique striped targets are captured at a similar rate to uniform grey targets, but parallel striped targets are significantly easier to capture. However, when multiple targets are present simultaneously during a trial we find that striped targets are captured in fewer attempts and more quickly than grey targets. CONCLUSIONS: Our results suggest that there may be differences in capture success based on target pattern orientation, perhaps suggesting that different visual mechanisms are involved in processing of parallel stripes compared to perpendicular/oblique stripes. However, these results do not seem to generalise to trials with multiple targets, and contrary to previous predictions, striped targets appear to be easier to capture when multiple targets are present compared to being presented individually. These results suggest that the different orientations of stripes seen on animals in nature (such as in fish and snakes) may serve different purposes, and that it is unclear whether motion dazzle effects may have greater benefits for animals living in groups.

Motion dazzle and the effects of target patterning on capture success.

BACKGROUND: Stripes and other high contrast patterns found on animals have been hypothesised to cause "motion dazzle", a type of defensive coloration that operates when in motion, causing predators to misjudge the speed and direction of object movement. Several recent studies have found some support for this idea, but little is currently understood about the mechanisms underlying this effect. Using humans as model 'predators' in a touch screen experiment we investigated further the effectiveness of striped targets in preventing capture, and considered how stripes compare to other types of patterning in order to understand what aspects of target patterning are important in making a target difficult to capture. RESULTS: We find that striped targets are among the most difficult to capture, but that other patterning types are also highly effective at preventing capture in this task. Several target types, including background sampled targets and targets with a 'spot' on were significantly easier to capture than striped targets. We also show differences in capture attempt rates between different target types, but we find no differences in learning rates between target types. CONCLUSIONS: We conclude that striped targets are effective in preventing capture, but are not uniquely difficult to catch, with luminance matched grey targets also showing a similar capture rate. We show that key factors in making capture easier are a lack of average background luminance matching and having trackable 'features' on the target body. We also find that striped patterns are attempted relatively quickly, despite being difficult to catch. We discuss these findings in relation to the motion dazzle hypothesis and how capture rates may be affected more generally by pattern type.

Defeating crypsis: detection and learning of camouflage strategies.

Camouflage is perhaps the most widespread defence against predators in nature and an active area of interdisciplinary research. Recent work has aimed to understand what camouflage types exist (e.g. background matching, disruptive, and distractive patterns) and their effectiveness. However, work has almost exclusively focused on the efficacy of these strategies in preventing initial detection, despite the fact that predators often encounter the same prey phenotype repeatedly, affording them opportunities to learn to find those prey more effectively. The overall value of a camouflage strategy may, therefore, reflect both its ability to prevent detection by predators and resist predator learning. We conducted four experiments with humans searching for hidden targets of different camouflage types (disruptive, distractive, and background matching of various contrast levels) over a series of touch screen trials. As with previous work, disruptive coloration was the most successful method of concealment overall, especially with relatively high contrast patterns, whereas potentially distractive markings were either neutral or costly. However, high contrast patterns incurred faster decreases in detection times over trials compared to other stimuli. In addition, potentially distractive markings were sometimes learnt more slowly than background matching markings, despite being found more readily overall. Finally, learning effects were highly dependent upon the experimental paradigm, including the number of prey types seen and whether subjects encountered targets simultaneously or sequentially. Our results show that the survival advantage of camouflage strategies reflects both their ability to avoid initial detection (sensory mechanisms) and predator learning (perceptual mechanisms).