ABSTRACT. First, I introduce a smart and mobile system that achieves the measurements of the human spectral sensitivity function in a short period of time, and show the age-related changes. Second, we introduce an experiment where young and aged subjects responded to the appearance of color patches using an elemental color scaling method so as to compare and relate the color appearance as seen by elderly and young people, and a color vision simulator which enables young observers to perceive color appearance as seen by the elderly in real time by using our color conversion method and dynamic image processing techniques. Third, I introduce a contrast discrimination experiment using achromatic Gabor patches (3.24 x 3.24 degrees, average luminance: 47.5 cd/m2) with several kinds of spatial frequency (1, 2, 4, 8, 16 cpd) and contrast intensity C (0, 5, 10, 15, 20, 40, 60, 80 %) in young (23-27 yrs) and elderly (65-75 yrs) participants and expressions for contrast sensitivity and contrast response as functions of age, spatial frequency and contrast intensity. Finally, I introduce age-related changes of visibility of characters.

ABSTRACT. Learning is known to facilitate performance in a wide range of perceptual skills and promote flexible behaviour. However, the brain mechanisms that support our ability for flexible perceptual decisions in older age remain largely unknown. Here we review perceptual and statistical learning studies suggesting that different brain plasticity mechanisms mediate behavioural improvements in young and older adults. First, we test for brain mechanisms that support learning to discriminate targets in cluttered scenes. Our behavioural studies demonstrate that visual selection rather than global feature processing provides a fundamental limit for learning-dependent plasticity in the ageing brain. Complementary fMRI studies suggest that visual shape learning in older adults engages primarily parietal regions, suggesting a key role for attentionally-guided learning in older age. Second, we test whether learning facilitates our ability to predict upcoming events based on implicit knowledge about the past. Our fMRI studies show that while predictive learning engages both implicit (striatum) and explicit (medial fronto-parietal) learning circuits in young adults, it recruits mostly implicit (striatum) learning circuits in older adults. Thus, our findings suggest that flexible behaviour is maintained across the lifespan by brain plasticity mechanisms that may compensate for decline in attentional and explicit memory circuits.

ABSTRACT. Patients with juvenile (JMD) and age-related macular degeneration (AMD) develop central scotomata in both eyes. We trained JMD/AMD patients to perform a texture-discrimination task (TDT) at their PRL. Six training sessions of approximately one hour duration were conducted over a period of approximately three weeks. Before, during and after training twelve patients and twelve age-matched controls took part in three fMRI sessions. Training led to a significant decrease in the stimulus onset asynchrony (SOA) between target and mask. We found a training-induced increase in the BOLD response in the early visual cortex and this increase was more pronounced for the PRL (trained) location compared to the opposite-PRL, untrained location. Our results suggest that perceptual learning can enhance eccentric vision and increase cortical activation in patients with central vision loss.

ABSTRACT. Age-related eye diseases such as glaucoma and age-related macular degeneration (AMD) are associated with retinal defects and vision loss. Because of the retinotopic organization of the connections of the visual pathways this may affect specific parts of the visual pathways and cortex. For this reason, over the past several years, structural and functional MRI studies have examined patients with age-related eye diseases and found structural changes –most often degeneration– in the visual pathways and brain, while most studies find that functional maps in visual cortex are remarkably stable. The most parsimonious explanation is that brain changes are caused by deprivation of the visual pathways and brain regions, rather than by transsynaptic degeneration. Of clinical relevance, for some of the diseases - in particular glaucoma and AMD – present results are compatible with the view that the eye disease is part of a more general neurological or neurodegenerative disorder that also affects the brain. Finally, establishing the degree of structural and functional stability and plasticity of the visual pathways has been relevant in the context of new therapeutic strategies to restore retinal function: the presence of degeneration implies that restoring retinal function may not suffice to also effectively restore vision.

ABSTRACT. High IOP is a major POAG risk factor for most forms of glaucoma contributing to retinal ganglion cell loss, cupping of the optic nerve head and loss of visual field which are the characteristic features of the condition. As we age resistance increases due to age changes but IOP does not rise because aqueous production and flow rate also decreases. In POAG the changes in the trabecular meshwork (TM) are so marked that IOP is often elevated to pathologically high levels that compromise retinal function. There is a marked loss of TM cells such that we have about 800,000 at birth which falls to around 400,000 in 80 year olds. In POAG the cell population falls to about half of that of age-matched normals. Understanding of drainage through the outflow system in ageing and glaucoma is essential as IOP lowering is the only means of halting glaucomatous progress. POAG patients are not aware of either their elevated IOP or their loss of visual field so early diagnosis remains a challenge and repeated population prevalence studies show that almost half of glaucoma patients are undetected.

ABSTRACT. Age-related Macular Degeneration (AMD) patients with central field loss make saccadic eye movements whose average horizontal component (in the forward direction) correlates with their reading speed. This is usually interpreted as evidence that patients' perceptual span is the major determinant of their reading speed (Calabrèse et al., 2014). Despite this consensus, very little is known about the characteristics of other oculomotor factors and their influence on reading speed in AMD. In this study, we describe and quantify an important oculomotor pattern which seems specific to AMD and which, to our knowledge, has never been investigated: the presence in many sentences of regions with very high densities of fixations (fixation clusters). We measured eye movements (with a video eye-tracker - 500 Hz) of 39 low vision patients who monocularly read aloud 14 single-line French sentences (print size: 3X acuity threshold). All patients had a dense scotoma covering the fovea, as assessed with MP1 microperimetry, and therefore used eccentric viewing. Only correctly read sentences were kept in the analysis. Our main result is that fixation clusters are strong determinants of reading speed independently of other oculomotor factors. This finding should help constrain the design of low vision reading models.

ABSTRACT. Processing fluency has been proven to be an important driver in forming judgments, especially in the domain of empirical aesthetics. For instance, Hedonic Fluency Model assumes that we like objects more when we can perceive and process them more fluently. The model has repetitively been verified but still faces clear limitations and also conceptual problems. One issue was addressed only recently: whether objective or subjective fluency is mainly causing the documented effects (Forster and colleagues)—actually, it was shown that perceptual fluency is not only implicitly processed, but can also be explicitly reported and is as such a subjective variable still an important determinant of liking. Another limitation is that most research used neutral or positive stimuli with low complexity, thus fluency effects on valence could not be effectively tested. However, when integrating negatively affected stimuli into a stimulus set, typical increases of liking could not be demonstrated but instead amplified evaluations. As a result, extended models have been established quite recently, for instance the Fluency Amplification Model (FAM) which assumes that fluency does not increase affective value, but amplifies the initial affective value of an object (Albrecht & Carbon).

ABSTRACT. Much evidence shows that perceptual and conceptual fluency influences diverse cognitive judgments and affective reactions. This evidence usually comes from studies which manipulate fluency by (i) objective features of the current target stimulus (e.g., averageness), or (ii) contextual stimuli that change fluency of the current target (e.g., identity priming). This evidence seems to suggest a straightforward recipe for increasing fluency and predicting its impact on judgments (e.g., enhance averageness, or use stronger priming). However, fluency is very much a subjective property - it is in the processing experience, not in the stimulus. As such, any stimulus could be potentially made "fluent" or "disfluent", depending on the current task context. My talk will show how easy it is to "flip" fluency of a stimulus with contextual and task manipulations, and, as a consequence, reverse cognitive judgments and hedonic reactions to the same target stimulus. I will illustrate this theoretical perspective with several empirical phenomena, including the "over-priming effect” , “cross-modal fluency" effect, "ugliness-in-averageness" effect, and "distrust-in-smiles" effect.

ABSTRACT. Current theories assume that perception and affect are separate realms of the mind. In contrast, we argue that affect is a genuine online-component of perception instantaneously mirroring the success of different perceptual stages. Consequently, we predicted that the success (failure) of even very early and cognitively encapsulated basic visual processing steps triggers immediate positive (negative) affective responses. To test this assumption, simple visual stimuli that either allowed or obstructed early visual processing stages without participants being aware of this were presented briefly. Across three series of experiments, we found more positive affective responses to stimuli allowing than obstructing Gestalt completion at very early visual stages, contour interpolation in illusory contours of Kanizsa stimuli, and visual disambiguation in possible versus impossible Necker cubes. This effect was observed on both verbal preference ratings as well as subtle facial muscle responses occurring within 2-4 seconds after stimulus-onset. For instance, we found affective discrimination between possible and impossible Necker cubes although a conscious discrimination between the stimuli was rendered impossible by very brief presentation timings. These effects demonstrate that early perceptual processes work remarkably fast and accurate and that there are distinct intrinsic affective signatures of success and failure of these processes.

ABSTRACT. Processing fluency provides a gratifying explanation for elevations of judgment in regard of various constructs, including truth, confidence, familiarity, fame, typicality, and especially liking. It is assumed that fluency is linked to a positive affect (e.g., Reber, Schwarz, & Winkielman, 2004). However, there are several findings which do not seem to fit into the theory (e.g., Albrecht & Carbon, 2014, Brinol, Petty, & Tormala, 2006, Olds & Westerman, 2012). Drawing on recent outcomes, we suggest a re-interpretation of processing fluency as an amplifying factor on affective reactions: The Fluency Amplification Model (FAM, Albrecht & Carbon, 2014) assumes an intensified affective judgment depending on the initial stimulus valence. FAM can serve as a framework for integrating seemingly contradictory findings. We attempt to compile several factors (including stimulus, person and context variables) under which fluency can affect behaviour and how these influences can be interpreted. Moreover, we address the risk of attributing findings to fluency effects whilst there might be shared variance of other origin, e.g. effects due to reward.

ABSTRACT. Different approaches in psychological aesthetics stress the importance of different factors of aesthetic preference: Affective approaches stress Pleasure (preference for stimuli with positive valence), Motivational approaches stress Interestingness (preference for arousing or supernormal stimuli), Perceptual approaches stress Harmony (preference for the good Gestalt), and Cognitive approaches stress Clarity (preference for fluently processed stimuli). In the present study we investigated the relationship between these factors and two main forms of aesthetic preference, one based on experience of Beauty and the other based on aesthetic Fascination (i.e. aesthetic experience, Markovic, 2012). Fourty-six participants judged seventy-two photographs of artificial objects and scenes (buildings, interiors, tools) on six scales: Beautiful, Fascinating, Pleasant, Harmonious, Interesting and Clear. Analyses indicated that all scales were significantly positively inter-correlated except Harmony (not correlated with Fascination) and Clarity (not correlated with Beauty and negatively correlated with Fascination). Multiple regression showed that Beauty was most strongly positively correlated with Pleasure, while Fascination was most strongly negatively correlated with Clarity. These results are in line with predictions of affective approaches, whereas predictions of cognitive approaches (e.g. processing fluency theory) failed: both clear and vague stimuli can be equally beautiful, while clear stimuli are less fascinating than vague stimuli.

ABSTRACT. A key prediction of processing fluency theory is that the ease of processing a stimulus determines the aesthetic response towards the stimulus. In particular, visual symmetry, simplicity, prototypicality, and contrast/clarity have been proposed to facilitate processing and hence to increase aesthetic liking. The existing literature provides a rich set of empirical studies in support of this basic notion. The majority of these studies either manipulates visual aspects of the stimulus according to one of the four mentioned dimensions or they measure subjective perceptions of these dimensions to predict aesthetic liking judgments. However, algorithmic and hence objective measures of these visual stimulus properties have not been established thus far. The key aim of the present research is to propose such objective measures and to validate their predictive strength with respect to a subjective fluency experience and aesthetic liking. The major advantage of these measures is that they can be applied in research that examines fluency effects based on large stimulus databases. Furthermore, several stimulus properties and their interactive effects can be considered simultaneously. We provide successful applications of the proposed measures across a range of stimulus categories (e.g., visual art, car designs) and discuss theoretical implications of our findings.

ABSTRACT. The EEG frequency-tagging approach has been recently expanded to high-level vision by periodically modulating diagnostic information (e.g., a periodic change of identity in a rapid train of identical face stimuli, Liu-Shuang et al., 2014). Here we designed a paradigm to identify an objective signature of visual discrimination (from nonface objects) and generalization (across widely variable face exemplars) with natural images. Scalp EEG is recorded while observers view natural images of objects at a rate of 5.88 images per second for 60 seconds. Natural images of faces are interleaved every 5 stimuli, i.e., at 1.18 Hz (5.88 Hz / 5). Face categorization is indexed by a high signal-to-noise ratio response specifically at the 1.18 Hz face stimulation frequency and its harmonics, localized over the right occipito-temporal cortex. Despite the use of natural images, this face-selective periodic EEG response is free of low-level visual confounds and is highly significant for single participants, even for one stimulation sequence. This approach reveals face-selective responses in the right hemisphere of 4-6 month old infants, and reveal a sequence of novel differential face-selective EEG components between 120 and 450 ms after natural face image onset, progressing from right lateral occipital to occipito-temporal and anterior temporal regions.

ABSTRACT. We investigated neural temporal dynamics in early visual cortex in feature-based attention when subjects shifted attention either within a feature dimension (i.e. color to color) or between feature dimensions (i.e. color to orientation) by means of frequency tagged random dot kinematograms (RDKs) that elicited steady state visual evoked potentials. Based on the Dimensional Weighting Account (DWA) that predicts a time consuming shift of weights between but not within feature dimensions, we expected longer shifting latencies from color to orientation compared to shifts from color to color. Neither Biased Competition nor the Feature Similarity Gain model would predict such latency differences. In line with DWA we found longer shifting times between compared to within feature dimensions. Interestingly, shifts to color were always faster compared to orientation, indicating top-down modulation in reversed hierarchy as previously found in animal and human studies. The following experiment manipulated shifts in color and/or location. To this end subjects either attended to the same location but shifted attention to another color, shifted location but remained at the same color or shifted color and location. We were interested in the temporal dynamics of such isolated or combined shifts. During the time of abstract submision, data is still analyzed.

ABSTRACT. The ability to parse scenes into objects and their surrounding backgrounds is a fundamental aspect of visual function that provides the building blocks for perception and higher-level cognition. In the current talk, I will discuss our research using frequency-tagged EEG, combined with an fMRI-based source imaging approach, to investigate the cortical networks responsible for figure-ground processing. In this approach, different spatial texture cues modulate at periodic frequencies causing the displays to alternate between uniform and segmented states. In different conditions and tasks these scenes create surrounded figure-ground arrangements or ambiguous figure-ground arrangements that are either the focus of attention or are task-irrelevant. Through these studies we identified separate brain networks responsible for figure and background processing that are invariant with respect to low-level stimulus texture attributes. We demonstrate that the Gestalt property of “surroundedness” is necessary to selectively activate the figure network, while processing of the border regions between figures and backgrounds elicits additional activity in frontal and occipital areas. In addition, we find that attention to the scene selectively enhances figure and border, but not background, responses. Collectively, these studies illustrate how frequency-tagged EEG can be used to investigate the mechanisms supporting the early stages of visual scene processing.

ABSTRACT. Frequency-tagging of stimuli in different sensory modalities at the same time can be a versatile means to study the dynamics of multisensory processing. Nevertheless, studies employing respective approaches have remained sparse. Here, we briefly review findings from previous multisensory frequency-tagging experiments and summarize data of two recent studies on the interplay of selective attention and audio-visual (AV) synchrony. Using frequency-tagged visual and auditory stimulation, we investigated influences of AV synchrony on spatial- (study 1) and color-selective gain effects on early visual processing (study 2). Steady-state responses (SSRs) driven by frequency-tagged stimuli allowed examining individual stimulus processing under conditions of synchronous vs. asynchronous tone presentation and when respective stimuli were attended vs. unattended. A prominent assumption holds that single AV-synchronous stimuli stand out and act as strong attractors for attention in unisensory contexts. In line with this prediction we found that, both, attending to the spatial location or color of a visual stimulus and its synchrony with a simultaneously presented tone, enhanced its processing. Interestingly, both gain effects combined additively, indicating that they relied on independent pools of neural resources.

ABSTRACT. A plethora of studies suggest that the biological relevance of visual stimuli is associated with facilitated sensory processing. Beyond the phenomenological demonstration of this effect however little is known about the neurophysiological mechanisms that mediate selective facilitation of biologically relevant cues. In this presentation, we discuss conceptual and methodological issues regarding the characterization and induction of threat and safety in the laboratory. We then turn to experimental explorations of behavioral and neurophysiological dynamics as observers learn that a novel, initially unengaging, stimulus is behaviorally relevant. Combining the steady-state visual evoked potential and other electrophysiological signals with hemodynamic neuroimaging techniques, we find that activity in lower-tier visual cortex changes as a function of motivational relevance acquired through Pavlovian fear conditioning. The temporal rate of these changes, their extent, and the brain regions involved reflect the specific behavioral and environmental contingencies to which observers adapt. Findings will be discussed in a theoretical framework that views emotions as action dispositions.

ABSTRACT. It remains hotly debated whether semantic information can be processed under continuous flash suppression (CFS), and one factor that affects previous results is the stimuli used. We adopted a priming paradigm and manipulated three types of primes (pictures, Chinese two-character words, and English words). The prime was shown in one eye and made invisible by continuously flashed masks presented in the other eye. Participants were instructed to classify whether a subsequently presented target picture was a living thing or non-living thing. The prime-target relationship was either congruent or incongruent in terms of the living or non-living categories. A visibility test at the end of each trial was conducted to ensure that all the trials included in the analysis contained only invisible primes. Results revealed an interaction of prime types and congruency: only Chinese words showed a robust priming effect (i.e., RTs were shorter for congruent trials than for incongruent ones). The null results for prime pictures may have been caused by the cancellation of positive and negative effects. We conclude that there indeed exists unconscious semantic processing under CFS, and the time course of semantic processing of different stimuli may account for the results we observed here.

ABSTRACT. Studies have shown that the motion aftereffect (MAE) is modulated by attention (e.g. Chaudhuri, 1990; Rees, Frith & Lavie, 1997; Taya, Adams, Graf & Lavie, 2009). When a moving stimulus is presented in the periphery the resultant MAE is reduced if, during adaptation, attention is diverted to a centrally presented stimulus. However, recent studies have failed to replicate this effect, attributing previous findings to response bias (Morgan, 2011, 2012, 2013). We explore whether differences in adaptation duration, within or across trials, might account for these apparent inconsistencies. Observers fixated a stream of letters. A surrounding annulus contained a random dot motion stimulus. Participants detected either speed changes in the adaptation stimulus (attention-focused), or white vowels within the letter stream (attention-diverted). Adaptation duration was manipulated by (i) varying the duration of single adaptation periods, or (ii) allowing adaptation to build up over multiple, short adaptation periods. MAEs were quantified by the reported motion direction of a test stimulus, whose true motion direction was orthogonal to the adapting motion. Observers’ MAEs were larger in the attention-focussed than the attention-diverted conditions. However, this effect was more apparent following short adaptation periods. Previous failures to find attentional effects may be due to MAE saturation.

ABSTRACT. Vehicle Head-Up Displays display information on the windscreen for use whilst driving. In the near future, these displays could cover the entire windscreen. This study asked whether discriminating visual stimuli in the periphery impaired performance on a central continuous tracking task. Participants performed a continuous motion tracking task in central vision whilst also performing a Landolt C detection and orientation discrimination task. The Landolt C targets were presented at four different contrasts in various positions in the periphery, 10o and 20o above, and 60o either side of the participant. In the motion tracking task participants attempted to keep a moving target inside a moveable box (similar to a lane-keeping task in driving). The difficulty of the tracking task was manipulated by varying the size of the box. Participants made more errors in the tracking task when the peripheral task was at low contrast, but eccentricity of the Landolt C did not affect number of errors made. Performance in the Landolt C task was better when the tracking task was harder. Low contrast peripheral targets may reduce attention available to a central task. Allocation of attention to the periphery may depend on the difficultly of a central task.

ABSTRACT. We review the evidence, from different data sets collected under different viewing conditions, illumination sources, and measurement protocols, for intra- and inter-observer variability in “generic subjective white-point” settings along the daylight locus. By "generic subjective white-point" we mean the subjective white-point independent of the specific context. We examine the evidence for a "blue" bias in subjective white-points (i.e. increased variability or reduced sensitivity in the “bluish” direction) by comparing the extent of variability along the daylight-locus generally and specifically in the "bluish" direction across all data sets. The variability in subjective white-point may correspond to subjective priors on illumination chromaticity. In turn, individual differences in assumptions about the specific illumination chromaticity on "the dress" (the recent internet phenomenon) is widely thought to explain the individual differences in reported dress colours. We therefore compare the variability in generic white-point settings collated across these datasets with the variability in white-point settings made in the specific context of the dress (Witzel and O'Regan, ECVP 2015). Our preliminary analysis suggests that (1) there is an overall "blue" bias in generic subjective white-point settings and (2) the variability in generic subjective white-point settings is insufficient to explain the variability in reported dress colours.

ABSTRACT. Of the hundreds of camouflage uniform patterns only a handful were given official names (e.g. the US Army’s M81 Woodland), therefore camouflage collectors have introduced their own terms of classification. Based on visual similarity, patterns are grouped into ‘families’, such as ‘Lizard’ or ‘Splinter’. While there is considerable consensus on the group identity of most patterns, classificatory features of groups are often loosely defined and with circular arguments arising from prior historical knowledge. While previous attempts to classify features of human textiles were based on manual selection of reoccurring shapes and structures by researchers (e.g. Tehrani & Collard, 2009), camouflage patterns are composed of abstract, hardly interpretable elements. Here we present a computational method describing the most exclusive features of groups and simulating the assortment process of camouflage experts. The method is composed of applying a log-Gabor filter bank to each colour segment of a pattern, followed by nonlinear dimensionality reduction (Tenenbaum et al. 2000) of the sum of filter responses. After applying a Gaussian mixture model to the low-dimension data, similar segments are grouped together and posterior probabilities of each cluster can be interpreted as the prevalence of a particular feature in the pattern.

ABSTRACT. Micromotions of the eyes while fixating a target are called fixation eye movements. Little is known about effects of fixation eye movements on visual perception. In this study, we performed a simulation experiments using a mathematical model of a wide-scale retinal network considering the receptive fields' properties in parafovea. Vertical sinusoidal grating patterns and random dot patterns, which consist of particular spacial frequencies, were given as input patterns. We computed the retinal ganglion cells responses which were composed of either a high or a low spacial frequency. Simulation results of grating pattern input suggest that microsaccades enhance the membrane potential of ganglion cells for the low-frequency pattern, and as for the high-frequency patten, it is influenced by drifts and tremor, particularly on the responses of M-type cells under both conditions. On the other hand, they do not influence the activity of M- or P-type cells when any random dot pattern inputs was given. These results suggest that the strength of the effects derived by fixation eye movements on the retinal responses depend on the nature of the input patterns.

ABSTRACT. Female carriers of anomalous trichromacy are known to express four, rather than three, types of retinal cone photopigment, but evidence for behavioural tetrachromacy is limited (Jordan et al., 2010). Colour matching functions are the definitive way of specifying a colour vision phenotype. We set out to measure such functions between 550 and 700nm in candidate tetrachromats and in control participants. Primary lights (550nm, 610nm, 690nm) were chosen by modelling the tetrachromatic dimension in an analogue of the MacLeod-Boynton colour space. In this space the spectrum locus between 550 and 690nm is convex with 610nm at the apex, but forms a straight line for trichromats. On each trial, two stimuli, each subtending 2 deg, were presented in succession. One field contained a mixture of two of the primaries and the other contained a mixture of the test wavelength and the remaining primary. For test wavelengths < 610nm, the test was combined with 690nm, and for test wavelengths > 610nm, it was combined with 550nm. Subjects adjusted the luminance ratios of two fields so as to achieve a colour match. In preliminary settings, subjects equated the sensation luminances of all primaries, and of the test, according to a minimum flicker criterion.

ABSTRACT. Recent studies demonstrate that alpha frequency photic driving modulates visual detection thresholds for subsequent target stimuli compared with baseline conditions (Mathewson et al., 2009, 2012: Spaak et al., 2014). However, apparently opposing results regarding the effects of varying the ISI between the onset of the final entrainer and the target have been reported. Mathewson et al. (2012) found peak sensitivity for a target onset in-phase with the entrainer (e.g. occurring when an entrainer would have appeared) and minimum sensitivity at time points that would lie mid-way between entertainers. In contrast, Spaak et al., (2014) report peak sensitivity when the target and entrainer were out-of-phase and minimum sensitivity at in-phase time points. Using a novel variant of Mathewson’ paradigm we tested whether this was due to entrainment at mask (Mathewson) rather than target (Spaak) locations. Eleven participants judged whether a black target dot (1 degree) on a grey background appeared before a black annulus mask (80% target present).Prior to the target either the annulus or target area was flickered in white (12Hz). The results showed peak sensitivity in-phase with both entrainers: the mask in one condition and the target in the second; a pattern that is unreported in previous research.

ABSTRACT. It has been demonstrated that exposure to a particular colour may affect physiological values (heart rate, blood pressure). Results are nevertheless controversial, being related to coloured lights and with a general agreement only on the relaxing effect of blue compared to red. A more ecological approach would consider the interaction between different shapes and colours. We presented 24 participants with three different objects (sphere, cube, mask) suspended inside a rectangular dome and viewed separately in each condition of coloured illumination (blue, red, or white light). If blue has a relaxing effect, it is interesting to evaluate its strength in association with different objects. Participants were asked to rate each object for each condition of illumination using 10 bipolar adjectives on 7 points Semantic Differential (SMD) Likert scales. Heart rate variability (HRV) and Skin Conductance Response (SCR) were recorded for each participant in each condition. SMD data reveal different attitudes towards both different colours and different objects. SCR data reveal an effect of object shape on physiological activation regardless of lighting colour. In sum, the effect of object shape is stronger than that of colour, suggesting at least some caution before generalizing the effect of colour on relaxing/activating states.

ABSTRACT. Eye movements play multiple roles - small stabilizing movements keep the scene steady during locomotion, whist relevant parts of the scene attract large directed movements. We investigated the effect of these movements on motion information available. Observers navigated through an office and a woodland environment, wearing a head mounted device that recorded the scene ahead and tracked eye movements, allowing us to determine the gaze direction in the scene. A target was present in the scene for observers to fixate or walked towards naturally. Local motion direction and magnitude was calculated for each frame of the sequences using a biologically plausible motion detection model. By realigning the frames relative to eye fixation location, we could reconstruct the input in retinal coordinates and compare this to the uncorrected input received by the camera. We used a method for calculating the error in heading direction estimation to evaluate the effect of eye movements on the available motion information as a function of time. We assessed whether the eyes were pointing in the direction of heading or scanning the scene. We found eye movements allowed heading information to be extracted accurately, even when large saccades away from the heading direction were made.

ABSTRACT. The 1/fβ fractal model can be used to generate textures resembling natural surfaces by varying its parameters (high cut-off frequency andβ). However, it is difficult to choose proper parameter values so that the model can produce textures described using perceptual features, e.g. roughness and directionality. This work focuses on automatically finding such parameter values. We generated 16000 textures by sampling in the parameter space at a linear scale. Twelve perceptual features for each texture were predicted by using the PCANet classification model, which were initially trained based on samples with subjective ratings (Liu & Dong, 2013). The sparse autoencoder was then used to construct a mapping between the twelve scales and the two parameters. It consisted of one input layer with 12 neurons, one hidden layer with 200 neurons and one output layer with 2 neurons. The autoencoder was learned by optimizing a cost function defined as the mean square error between the predicted and real parameters. We randomly chose 14000 textures for training and the remaining 2000 for testing. The result showed that our method can accurately predict parameter values, and with these values the model can produce texture in accordance with perceptual description.

ABSTRACT. Individual differences in response to visual contrast are typically observed in the amount of (i) saturation of responses at high contrast levels and (ii) suppression by cross-channel masking. A potential mechanism that underlies these processes is GABA-ergic inhibition, as GABA antagonists eliminate both effects in single neurons (e.g. Morrone, Burr & Speed, 1987). However, these two properties of contrast transduction have not been directly compared within the human population. We measured steady-state EEG responses to patches of sine-wave grating s (0.5c/deg) flickering at 7Hz in 100 subjects at seven contrast levels (0-64%). In some conditions a high contrast (32%) orthogonal mask flickering at 5Hz was superimposed on the target stimuli. We calculated a saturation index, defined as the ratio of amplitudes at the two highest target contrast levels. To quantify masking, we took the ratio of amplitudes when the mask was present versus absent, averaged across intermediate contrast levels. We found a highly significant correlation between masking and saturation across observers (r = 0.50, p < .001), such that individuals who exhibited substantial masking also displayed strong saturation. This suggests a common underlying mechanism, most likely GABA-ergic inhibition, that varies widely in the population and therefore may have clinical relevance.

ABSTRACT. Skin colour measurements were obtained for 539 subjects at four different body locations (forehead, cheek, inner arm, back of hand) in four countries with a view of establishing a new skin colour database for different ethnicities (Caucasians, Chinese, Kurdish, Thai). Skin colour means, variation and skin colour boundaries were calculated. We report three main results (1) When plotted in a standard colour appearance space (CIE LAB), there is a large overlap between the skin distributions and boundaries for all four ethnicities. (2) The largest variability and gamut in skin tones was found in the Caucasian group, the smallest one in the Chinese group. (3) The largest ethnicity differences were found in yellowness (b*) with the Thai group showing the largest and the caucausian group the smallest b* values. Vice versa, the caucausian group is the group with the largest lightness values and differs significantly from all other groups. The ethnic differences along the redness dimension (a*) are small and less systematic. Consequences for visual processing of skin colour will be discussed.

ABSTRACT. According to Posner and Peterson (1990), neural systems of attention can be divided into three networks. These networks execute such functions as alerting, orienting, and executive control. In 2002 The Attentional Network Test (ANT) was developed by Fan et. al. as a tool for testing not only all three systems simultaneously, but interactions between them. Callejas and associates (2004) added an auditory trigger for alerting to further improve the ANT. We present data on the Callejas ANT test while participants gaze was monitored in order to gather information on how eye movements may or may not affect reaction times for each mode of attention. Participants were instructed to remain fixated at centre throughout the experiment, but no feedback was given if they moved their eyes during a trial. Overall RTs were faster in congruent trials versus incongruent and in cued versus uncued. Initial results showed that there was no significant difference in RT between trials with and without alerting sound, which does not replicate previous findings (Callejas et. al., 2004) A significant interaction was found between congruency and cueing with larger spatial cuing in incongruent trials (50 ms) than in congruent trials (13 ms).

ABSTRACT. Does visual awareness correlate with early activity in visual cortex or with later wide-spread neural activation? This question was studied by presenting liminal targets in one of the four quadrants of the screen and asking the participants to make forced-choice localization responses to them and to rate their visual awareness of each target, while electroencephalography was measured. The analyses of event-related potential (ERP) correlates of awareness kept the correctness of the localization response constant so that only awareness varied. Aware-correct trials were associated with enhanced contralateral N200 as compared with unaware-correct trials (= visual awareness negativity, VAN). This effect also correlated with the subjective ability to discriminate between stimulus-present and -absent trials (d’). In addition, aware-correct trials were associated with later enhanced P3 amplitudes, but this effect correlated only with the response criterion (c). ERPs to unaware-correct trials elicited larger contralateral N200 than ERPs to unaware-incorrect trials, and this effect correlated with conservative response bias, suggesting that it reflected weak awareness rather than unconscious processing. The results suggest that the enhanced N200 correlates with graded awareness. The results support theories of visual awareness in which early activity in the visual cortex gives rise for the first subjective visual experiences.

ABSTRACT. Top-down voluntary attention modulates the amplitude of SSVEPs in such a way that cued shifts of feature-selective attention lead to an enhancement of the attended stimulus. It is unclear, if feature-based target selection guided by internal cues, such as a previously learned rank-order rule, leads to a modulation of attentional facilitation in HVC. We hypothesized that SSVEP amplitudes of attended stimuli are positively correlated with stimulus priority. Initially, subjects learned a color rank-rule consisting of three secondary colors of different priority (magenta > cyan > orange), that was changed in a second condition (orange > magenta > cyan). In each experimental trial, two superimposed random dot kinematograms (RDKs) were presented, flickering at different frequencies while we recorded the SSVEP. Subjects had to attend to one of the RDKs based on the learned color rank-rule, while ignoring the competing RDK. We found that SSVEP amplitudes were modulated by stimulus rank: higher amplitudes were found for the attended stimulus of highest priority (magenta) as compared to the identical attended stimulus of medium priority (magenta). Our results indicate that, when a rank rule has to be applied in order to select targets from distractors, activity in HVC scales following the stimulus rank.

ABSTRACT. Glare of ambient light degrades the visibility of a display. Optical properties such as reflectance and haze factor are commonly used to evaluate the quality of any anti-glare treatment on the display surface. In practical situations, however, these properties often disagree with actual appearance. In this study, using a visibility matching technique and degradation category rating (DCR), we evaluated the quality of display images on which a glare reflection image was superimposed. In the matching experiment observers adjusted the luminance and contrast in an original display image to replicate the visibility of the degraded image. In DCR experiment observers evaluated the quality of image by seven rating categories. The display image with glare was simulated by superimposing a reflection image on a display image. The results showed that all observers set higher luminance and lower contrast to replicate the visibility of the images degraded by the reflection of images with higher luminance. However, the observers’ settings were lower in luminance and contrast than those calculated from superimposed images. This indicates that observers might segregate the appearance of the original display image perceptually from the reflection image. In DCR, the observer’s ratings monotonically declined as the image contrast decreased.

ABSTRACT. When two visual events happen in very rapid succession, they may not be separable to the visual system and are thus processed as one event. The probability for the events to be temporally segregated increases with longer ISIs. Given suitable timing, identical paired-stimulus presentations may result in different outcomes (segregation or integration) on a trial-by-trial basis. The neuronal correlates of these differences may provide insight into the temporal processing hierarchy of the visual system. We presented two stimuli at different ISIs (33, 66 and 100ms) while recording MEG. The first stimulus was presented at near threshold intensity, while the second was presented at above-threshold intensity. Stimuli were Gaussian blobs presented on a neutral gray background. Subjects reported the number of stimuli seen (1 or 2). Behavioral performance was most veridical at 100ms, while at 33ms most trials were perceived as only one stimulation instead of two. Recorded difference ERFs between trials with one and two reported stimuli show a field strength reduction around and slightly before the time of stimulus presentation. This suggests the probability of detection of the near-threshold stimulus to be modulated by ongoing processes around the time of stimulus presentation, consistent with a theory of perceptual moments.

ABSTRACT. Background: Associations between global- and local-shape perception and colour discrimination were investigated. The shape tasks were completed on different background colours, although background colour was irrelevant to either task. Little research has been conducted in this area and none has used colours tailored to the cone-opponent pathways early in the visual system. Method: Participants were presented briefly with one of four shapes on different trials: a global square made up of small square (congruent) or diamond (incongruent) local elements, or a global diamond made up of small diamond (congruent) or square (incongruent) local elements. Each display was presented on five coloured backgrounds (tritan/S-cone: purple, neutral, yellow; L(M)-cone: red, neutral, green). Luminance and contrast were equated across conditions. Participants also completed the Farnsworth-Munsell 100-hue test. Results: Participants were more accurate at the global task than local and responded more quickly for congruent than incongruent trials, as expected. There were no significant differences between performance on any of the coloured backgrounds for either accuracy or reaction times. There were, however, significant correlations between colour discrimination as assessed by Farnsworth-Munsell scores and some of the shape perception tasks. Participants with poorer colour discrimination were less accurate despite colour being irrelevant to the tasks.

ABSTRACT. Most of the routine tasks that people perform on a daily basis are processed automatically. For example, a person can detect obvious grammatical errors in the text in her native language without much effort. Knowing that automated eye movements can serve as predictors for decision making process, we developed a computational model that can identify the stimulus material containing errors based on the subject’s fixations. In our experiment, the subjects were asked to check if the presented text contained errors. They provided correct answers in more than 92% of cases, and their average fixation duration was not statistically different in cases with errors and without them. To design the model that predicts the presence of errors in the stimulus material, we relied on the duration and the average value of the fixation and used only the correct answers of the subjects for training the model. After comparing several algorithms, we chose the DecisionTree with learning outcome equal to 79.7 % (kappa 0.533) (ZeroR 60.51 %). We then tested it on the new dataset and got accurate predictions for errors in the text in 68.6% of the cases.

ABSTRACT. Changes in the surface reflectance of an object and changes in the spectral content of the illumination both change the spectral composition of light reaching our eyes, but usually we do not have difficulty determining which change occurred. We investigated how specular reflections might allow an observer to distinguish changes in lighting and reflectance in scenes containing only one surface. Specular reflections make a surface look glossy and contain reflected light that has not been spectrally modified by the surface, so could provide information about the illuminant. Additionally, there are spatio-chromatic changes that differ systematically for changes in lighting and reflectance, and we investigate how observers extract information that is not available at any given instant. We presented hyperspectrally raytraced movies showing isolated objects undergoing gradual illuminant and reflectance changes. Observers were able to distinguish the two types of change at very low specularities (~1%). Performance was reduced, but not extinguished, when we distorted the images to weaken some of the spatial or chromatic relationships. The highlights on low-gloss surfaces are heavily contaminated by surface colour so they do not allow direct estimation of the illuminant. Yet the chromatic transitions under gradual changes in lighting and reflectance are perceptually discriminable.

ABSTRACT. Although the use of daylight is one strategy for reducing lighting energy, the effect of daylight on space brightness perception is unclear. In the few studies that have examined the effects of daylight, the brightness produced by daylight is not as efficiently enhanced as expected by horizontal illuminance. Moreover, those studies did not consider a scenic view through a window. In the present study, we further investigated the effects of daylight on space brightness by manipulating scenic views outside a window. In the experiment, we used two scale models simulating an office: one was a room with a window (test room) and the other was a room without a window (reference room). There were two types of scenic views (natural or urban landscape) with or without a human-shaped board covered with a full-length photograph of an adult male. In each trial, participants viewed the two models and rated the space brightness of the test room relative to that of the reference room. The results clearly showed that the efficiency of brightness enhancement provided by daylight was lower with a scenic view than without a scenic view, and the pattern of results did not change with the human-shaped board.

ABSTRACT. Attentional biases play an important role in guiding visual behaviour. Persistent, long-term attentional biases to arbitrary, non-emotional stimuli (such as a particular colour) can be induced via a single information sheet (Knight et al. 2015). This induced attentional bias causes green objects to capture and hold visual attention even when such items are behaviourally irrelevant distractors and should be ignored. Neurobiologically, the left dorsolateral prefrontal cortex (DLPFC) is thought to be involved in such attentional suppression which we investigated by increasing or decreasing its cortical excitability via transcranial direct current stimulation (tDCS). Anodal tDCS of the left DLPFC reduced distractions caused by irrelevant green items suggesting that increasing the cortical excitability here can improve cognitive control over behaviourally-irrelevant attentional bias-related items. Decreasing the cortical excitability of the left DLPFC via cathodal tDCS on the other hand reduced the overall effect of the induced attentional bias, potentially by preventing the attentional bias fully developing in the first place. These results suggest a causal role of the left DLPFC in the establishment of attentional biases, and the control we have over distractions caused by these biases.

ABSTRACT. Contrast is an important feature for visual performance on many tasks such as object identification, speed or motion detection (Kilpeläinen, Nurminen, & Donner, 2011). Perceived contrast of a grating depends on the luminance of its background. On the other hand, context often causes a large difference between luminance and its perceived correspondence, brightness (e.g. simultaneous brightness contrast). Thus, characterizing different effects of luminance- and context-dependent brightness on contrast is critical. In this study we investigate how context-dependent brightness affects contrast judgements using a variant of Adelson's checkerboard illusion stimulus (Adelson, 1995). Two series of behavioral experiments were conducted using different psychophysical procedures. First, brightness of equiluminant target regions on the illusory stimulus was measured. Significant differences were observed in participants' brightness judgements (N=8). Second, we superimposed rectified gratings with incremental and decremental contrast on the target regions, and measured their perceived contrast for various levels of frequency, luminance and contrast. Results show that context-dependent brightness of the target region, not only its luminance, influences the perceived contrast of incremental gratings (N=6). This finding indicates that perceived contrast depends on background brightness, as well as luminance. In addition, results of both experimental series showed that, experimental procedures significantly affected the measurements.

ABSTRACT. A wide range of dichoptic visual phenomena can be displayed and controlled using a smart phone. As part of “Project LITE: Light Inquiry Through Experiments”, we developed a suite of about 50 apps (found at http://lite5.bu.edu/v4.1/) that include: binocular rivalry; dichoptic red and green color mixing to produce “cortical yellow”; three dimensional Kanizsa square; controllable texture random dot stereograms; and a variety of stereoscopic images. I have now developed a new binocular viewer that can conveniently assist people (who cannot “free view”) in seeing and using the displayed image pairs. The “Dichopter” was designed using rapid prototyping software and 3D printing for several of its parts. It consists of a pair of lenses with 10 cm focal length; a newly designed bracket (which obviates the need for a septum) that holds both the lenses and the phone mount; a spring loaded mount that can accommodate a wide range of cell phones or other image display devices; and a screw mount to allow the device to be either handheld or supported by a standard tripod. The Dichopter is intended for the educational display of visual perception phenomena, and for clinical uses, either as a diagnostic or as a training tool.

ABSTRACT. Visual response tasks are interfered by task irrelevant features in some situations. In this study we chose two popular interfering effects of Stroop and Simon, to investigate temporal characteristics of the interferences. Stimuli were painted with color of red or green: Participants were instructed to respond to the color with their left or right hands. For examining Stroop effect in Experiment 1, the task irrelevant feature was the stimulus shape of color-word of “RED” or “GREEN” (in Japanese). For Simon effect in Experiment 2, the stimulus was a circle, and the irrelevant feature was its location of the left or right of a fixation cross. SOA between the color and the irrelevant feature appearances was set from -300 to +300 ms. The irrelevant feature appeared first in negative SOAs, and the color did in positive SOAs. As results, Stroop effect was obtained at 0 ms and negative SOAs, whereas Simon effect was at 0 ms and positive SOAs. This agrees with that Stroop effect occurs around data processing stages in visual response processing, whereas Simon effect does at response selection stage or later. We further discuss Simon effect with an arrow that is a symbolic figure representing spatial orientation.

ABSTRACT. Perfect colour constancy entails no change in surface appearance under changing illumination; therefore the ability to discriminate an illumination change on a fixed scene may provide a probe for measuring constancy. Previously, we found thresholds for global illumination discrimination higher for ‘bluer’ illuminations compared to ‘yellower’, ‘redder’ or ‘greener’ illuminations, at neutral adaptation (D67), indicating better colour constancy for ‘bluer’ illuminations (Pearce et al. 2014). Here we examine whether this asymmetry in constancy thresholds depends on the adaptation point. Participants performed the illumination discrimination task in an immersive setting, viewing a 3D Mondrian-papered scene illuminated by spectrally tuneable 10-primary LED lamps. Five adaptation conditions were tested in separate blocks, corresponding to five target lights (neutral, ‘red’, ‘green’, and extreme ‘blue’ and ‘yellow’ daylights). On each trial, after viewing the target light, participants indicated which of two successive lights (the target and a comparison) matched it; comparison lights (‘bluer’, ‘greener’, ‘redder’, ‘yellower’) varied systematically away from each adaptation point. Thresholds were determined using a transformed, weighted, 1-up, 3-down staircase procedure. The results reveal an interaction between the direction of illumination change and adaptation point, yet on average, the asymmetry in thresholds prevails across conditions, confirming the blue bias in constancy.

ABSTRACT. The aim of this study was to estimate discrimination thresholds for calibrated skin images in 3-D colour space. The stimuli were generated by adding test vectors to the original polychromatic image in CIELAB space. Thresholds along 14 directions were estimated using a 4-AFC task and the lengths of the test vectors were controlled by the QUEST adaptive procedure. Ellipsoids were fitted to the estimated thresholds in u’v’Y space.

16 subjects (11 female, 5 male; 15 Caucasian, 1 Oriental; Mean age 24.75) were tested in an anechoic chamber under three illumination conditions - dark, daylight (D65) and cool white fluorescent (TL84). Two skin-patch images were used – Caucasian and Oriental. A subset of these observers was also tested using uniform patches with the same average CIELAB luminance as the images. The stimuli were displayed using the CRS ViSaGe system on a calibrated EIZO ColorEdge CG243W monitor.

Our two main findings were: 1. On average, thresholds are lower for Caucasian skin patches as compared to Oriental skin patches. 2. The thresholds in all directions are smaller for uniform patches compared to polychromatic skin patches.

ABSTRACT. Eyes and gaze direction play an important role in social and non-verbal communication. This study investigated the time course of brain processes involved in the visual perception of different gaze interactions in woman–child dyads and the association between attachment dimensions and brain activation during the presentation of gaze interactions by using ERP and sLORETA methodology. Electroencephalogram (EEG) data about 44 female subjects (24 ± 2 years old) were recorded using a 256-channel HydroCel Geodesic Sensor Net. The findings showed a main role of limbic and primary somatosensory brain areas in response to the gaze interaction. Moreover the woman who avoids the infant while the infant looks at her produces an spontaneous and early alert response probably due to the un-responsiveness of the caregiver toward the infant. In response to the convergence gaze, was found a negative association between the confidence attachment dimension and the intensities of the temporal and limbic areas, which seems to be an early spontaneous and emotional response to the intimacy; moreover was found a positive association between the relationships as secondary attachment dimension and the intensities of the frontal and parietal areas, which seems to be a more controlled and defensive response to the intimacy.

ABSTRACT. We see shadows everywhere; however, it is not understood how shadows affect our perception of brightness in a space. To investigate this, we used multidimensional scaling to analyze the dissimilarity structure produced by different intensities of illuminance and shadows. Four miniature rooms were set up, each containing a parallelepiped object covered by black felt with an accompanying shadow. Each array was illuminated independently with one of three lighting intensities. Additionally, each shadow was projected by a ceiling digital projector with one of four intensities. Six observers evaluated the apparent brightness of the space in Experiment 1 and the intensity of the shadows in Experiment 2 by ranking. In Experiment 1, the apparent brightness of the space was proportional to the intensity of illuminance and the shades of the shadow. In Experiment 2, the perceived intensity of the shadow was proportional to the intensity of illuminance and the luminance contrast of the shadow and its background, but not to the intensity of the shadow itself. These findings show that shadows affect the perception of brightness in a space.

ABSTRACT. Humans employ two distinct visual exploration modes, i.e., local scan and global scan, while they freely view natural scene images. Here we show that these modes are also observed in macaques and affect neuronal activities in the primary visual cortex (V1). We trained a macaque monkey to freely view a set of images constructed by embedding 5 object images (2 deg in diameter) at random positions in natural scene images (40 x 30 deg2). Eye movements and V1 neuronal activities were concurrently recorded with a scleral search coil and extracellular electrodes, respectively. We found that 70% of the fixations were on the embedded objects. Saccades into an object followed by a saccade to another object (i.e. across-object saccades, representing global scan), or saccades made successively within an object (i.e. within-object saccades, representing local scan) occurred at significantly higher rates than expected by chance (chi-square test, p < 0.001). V1 neuronal activities showed stronger responses following fixations after across-object saccades than after within-object saccades, even when both saccades brought the gaze to the same embedded images. These results suggest the existence of distinct scan modes and a top-down modulation of V1 activity affecting the response magnitude depending on the current mode.

ABSTRACT. Absorption and scattering are the two most fundamental processes through which light is modified by a material, giving rise to all visual surface properties. While each is associated with a perceptual quality (colour and texture, respectively), both have been shown to affect judgements of surface properties in opaque objects. Our objective was to determine whether these fundamental processes produce separable perceptual qualities in translucent liquids, or whether they interact – and if they interact, the nature of this interaction. Perceptual scaling and conjoint measurement techniques (Ho, Landy & Maloney, 2008) were used with physically realistic rendered stimuli, and models of best fit determined. The model that allowed for complex interaction was the best fit for both qualities , showing that our perceptions of light absorption and scattering cannot be easily separated. The between-participant variation for judgements of scatter implied that observers may rely on slightly different cues in the visual scene. We will discuss the relationship between these results and simulated results obtained using low-level image statistics. An additional experiment was carried out using photographs of real stimuli. The models of best fit for real and artificial stimuli differ slightly, suggesting greater separability in the case of physically realisable manipulations.

ABSTRACT. Developing from previous work distinguishing attention to features and objects (Eimer and Grubert, 2014) and work showing early top-down selection for categories, we measured the N2pc component as an electrophysiological marker of attentional object selection. Participants searched for a target that was defined by a conjunction of color and category, using alphanumeric stimuli. Some search displays contained the target or a partially matching non-target that either shared a color or category with the target among three non-matching distractors. In other displays, the target and a partially matching non-target were simultaneously present. N2pc results demonstrated that attention was allocated rapidly and in parallel to target color and target category. This shows that object categories can act as a feature for the guidance of attention. Following this initial stage of object selection where attention was allocated independently and additively to all objects with target-matching features, global competitive interactions between these objects resulted in the selective processing of target objects. These findings demonstrate that category-based attentional selection can be tracked in real time, Category-based and feature-based selection operate in parallel during the gradual emergence of goal-selective processing biases in ventral visual cortex.

ABSTRACT. The contextual cueing effect shows that invariant visual contexts can guide deployment of attention and facilitate visual search (Chun, 2000). Previous studies only focused on normal contrast search displays. However, it is still not clear whether the contextual cueing effect also occurs for low contrast displays. Therefore, the present study used the standard contextual paradigm to investigate this issue. In contrast to the fixed display contrast used in the standard paradigm, here the item-background luminance contrast was set to high level in training and to low level in the transfer session in Experiment 1. Luminance settings were reversed in Experiment 2. All participants attended both experiments with at least two weeks break in between, and the experimental order of their participation was counterbalanced. During the training session, participants were significantly faster in responding to repeated displays than non-repeated displays in Experiment 1 (i.e. contextual cueing effect) but no differences were found between repeated and non-repeated displays in Experiment 2. Interestingly, no contextual facilitation was observed in the transfer session of either experiment, suggesting low display contrast impedes both contextual learning and retrieval. We discuss the findings in relation to the perceptual overload and task difficulty.

ABSTRACT. For maximum visual acuity we have to fixate an object of interest and keep it on the fovea because visual acuity declines very fast outside the fovea. Eye movements are therefore vital for accurate visual perception since they allow us to maintain gaze on the object of interest. Latency, peak velocity, duration and accuracy are the most studied parameters of eye movements and their characteristics are rather well known. Less is known about how stimulus properties, such as size and saliency, might influence these parameters. We tested saccades to targets of varied size (ranging from 0.2° to 1.4° in seven steps), and saliency (brightness varied from almost white on a white background,to black, also in seven steps). The targets appeared either left or right of central fixation (at 8° eccentricity). Our results show that latencies were shorter towards larger than smaller stimuli, but size did not affect peak velocity. Similarly, latencies became shorter with increased salience, but peak velocities were again unaffected. Our results emphasize the importance of stimulus properties for investigations of saccadic parameters.

ABSTRACT. The dead zone of attention (DZA) refers to an exaggerated failure of noticing changes in close proximity to a focus of interest of a visual scene [Utochkin, 2011, Visual Cognition, 19, 1063-1088]. In our study, we tested if the DZA pattern is reproduced in gaze behavior. We ran two change blindness experiments with eye-tracking using DZA-inducing scenes. In Experiment 1, observers looked for one change that could be central, near or far from the focal point, in Experiment 2, they looked for either near, or far changes in the presence of a once noticed central one. Behavioral data replicated the DZA pattern from earlier experiments. However, there was no DZA signature in eye movements. The average fixation hit rate was about the same for all three change types. Moreover, measuring the mean fixation distance to each change, the shortest mean distances were around central changes, the longest were around far changes and intermediate distances around near changes. Eye movements show a gradient pattern around the focus of interest rather than a “Mexican hat” one predicted by DZA. In general, our results suggest that there is some dissociation between eye movements and the distribution of attention around the center of interest.

ABSTRACT. How do we estimate visibility of perceptual stimuli? Is this only the effect of perceptual evidence accumulation or do we use other information to judge visibility? We present four studies aiming to investigate those questions by analysing the effects of decision in identification task on subsequent subjective visibility ratings. Perceptual awareness scale was applied in the context of a Gabor patches orientation identification task either before, or after the identification responses. Results of the first two studies show that visibility ratings predict accuracy, but the effect is weaker, when visibility ratings precede identification responses. This could be a result of either longer time participants required to estimate visibility, or the influence of the identification decision itself. Thus, we run two additional studies to separate the influence of decision time and identification/awareness rating order. First, we manipulated the amount of time participants had to rate visibility, while the identification decision was always made after visibility ratings. Then, we manipulated tasks order but controlled the time participants had to rate visibility. The results show the effects of both manipulations. Thus, we conclude that visibility ratings are based on low-level visual processing, but also influenced by post-decision processing.

ABSTRACT. Humans tend to orient attention in response to gaze direction of conspecifics, a phenomenon which can be shaped by several social variables. Here, two experiments are reported aimed to investigate whether voluntary saccadic eye movements are modulated by centrally-placed task-irrelevant White and Black faces with averted gaze. A symbolic central cue informed White participants to perform a saccade rightwards or leftwards. Saccade direction could be either spatially congruent or incongruent with gaze direction of the facial stimulus. In the first experiment, more directional errors emerged in response to White vs. Black faces, but only when the averted-gaze face and cue onsets were synchronous (i.e., 0 ms) rather than separated by a 900-ms asynchrony. In the second experiment, two temporal intervals (50 ms vs. 1000 ms) occurred between the direct-gaze face and the averted-gaze face onsets, whereas the averted-gaze face and cue onset were synchronous (i.e., 0 ms). Also in this case, a greater interference emerged for White vs. Black faces, and irrespective of the temporal interval. Overall, these findings suggest that saccadic generation system seems to be sensitive to features of facial stimuli conveying eye gaze.

ABSTRACT. Evidence for sensory processing in an object-based frame of reference has previously only been indirect, through object-based attentional deficits in the neglect syndrome. However, no direct physiological evidence exists of sensory processing differentiating, for instance, the left and right parts of an object in a non-retinotopic object-based coordinate system.

In the present study we investigate electroencephalographic correlates of different coding schemes in 27 healthy participants. To differentiate cortical responses to the left and right side of faces, independently of retinotopic position, a steady-state paradigm was employed, tagging the two face-halves with distinct frequencies.

To separate retinotopic- and object-based frames of reference, we used a general linear model to explain the pattern of induced response. This analysis revealed significant electrode-clusters for both explanatory variables, indicating the co-existence of both types of coding schemes. The object-based processing was further supported by increased within-frequency coherence between lateralized electrodes pairs. In contrast, the integration of the two face halves was supported by between-frequency power envelope correlations between bilateral electrode pairs.

Taken together, our results provide direct evidence of object-based coding in healthy humans, which is based on the canonical representation of objects’ left and right parts, and relying on intra- and inter-hemispheric communication.

ABSTRACT. Visual attention can be directed by visual and linguistic information. It is not well understood how attention is directed when linguistic information conflicts with the visual scene. Knoeferle and Crocker (2006) established the coordinated interplay account model of sentence comprehension and linguistically mediated visual attention, but it did not address such conflicts. To inform the model, we created a decision making experiment where auditory and visual information provided conflicting information. Participants’ eye and hand movements were tracked during a sorting task. Each trial, a square appeared onscreen without a category. The participant selected the square with a mouse and moved it to a ‘sensor’ to reveal its category, which dictated its sorting location. Once placed on the sensor, the participant would hear a linguistic category utterance and briefly see the visual category. Crucially, on some trials, visual and auditory information mismatched. Additionally, the conflict was manipulated such that visual information preceded auditory by 200ms, followed it by 200ms or they occurred simultaneously. After being presented with this information, participants sorted the block while under a time limit. We present preliminary data examining eye and mouse movements, with a focus on predictive eye movements made following auditory and visual information presentation.

ABSTRACT. The predictive coding framework (Friston, 2009) conceptualizes attention as an error in prediction. This theorization implies automatic processing of stimuli that do not fit with the prediction generated by the system. Recently, we showed that action-effect captures attention (Kumar, Manjaly & Sunny, 2015) through intentional binding between action and a feature singleton. Prediction is a critical component in binding. In the present study, we explore how predictions about the space or time and the probability of action-effect occurrence interact with each other to modulate attention capture. In three experiments we showed that a feature singleton does not capture attention when all of these three variables are fully predictable throughout the experiment. That is when there is no error in prediction capture does not occur. However, when there is an unpredictability associated with any one of these variables, the feature singleton captures attention. That is, only in the case of a prediction error do action and effect bind together leading to capture. These results provide conclusive evidence in support that bottom-up attention is an outcome of error in prediction. In case of complete predictability, attention is allocated in a top-down manner as the system requires minimal information from the environment.

ABSTRACT. Autism Spectrum Disorders (ASD) are associated with atypical visuo-cognitive processing. Weak central coherence (Happé & Frith, 2006) – the tendency for individuals with autism to employ locally-oriented cognition, with a reduced ability to extract global meaning – may help explain perceptual-cognitive differences in ASD. We present two experiments investigating colour perception and cognition in individuals with and without ASD. In Experiment 1, adults with ASD (n=16) performed just as well as typical adults (n=16) at averaging the colour of a rapidly-presented, colourful ensemble, despite this task requiring distributed attention to the holistic characteristics of the ensemble. However, the ASD group outperformed the typical group when evaluating whether a single colour was part of the ensemble – a task requiring more local attention. In Experiment 2, unexpectedly, there were no group differences in whole-scene colour afterimages (“Spanish Castle Illusion”, Sadowski, 2006). For both groups (total n=22), afterimages were stronger for scenes with diagnostic colour (natural scenes) than without diagnostic colour (man-made objects) for both up-right and upside-down images. Together, these findings suggest that the abilities to extract summary statistics from scenes and apply prior knowledge to current percepts is not disrupted in autism, contrary to prominent theoretical accounts.

ABSTRACT. Attentional selection in visual search and perceptual selection in bistable perception show similar characteristics. Both in visual search and bistable perception observers are biased toward previously selected targets or interpretations. We measured whether these priming effects are related by presenting two interleaved trial types that contained either an ambiguous sphere that could be perceived as rotating in one of two directions around a central axis, or a search display with a target sphere and distractor spheres that resembled the two possible interpretations of the ambiguous stimulus. The two trial types were interleaved to examine whether priming elicited by target selection during visual search could affect the outcome of perceptual conflict during bistable perception and, conversely, whether perceptual priming during bistable perception could affect search response times during visual search. We established that our experiments elicited search priming and perceptual priming, but found no evidence that search priming affected bistable perception, or that perceptual priming affected search response times. This shows that while each paradigm elicited priming, there was no interaction between the two forms of priming.

ABSTRACT. Cued locations result in short lived response facilitation and a longer lasting disadvantage – a result termed ‘Inhibition of return’ (IOR). Krüger et al, (2014) explored the possible contributions of attention and perceptual merging on facilitation with reaction time responses as well as the perceived onset time of the probe using a non-speeded clock judgement task. Cues presented both before and after the target probe influenced results, suggesting that the perceptual merging of cue and target was caused by an attentional account which included re-entrant processing. Here we present data across a range of cue target intervals using both speeded responses and clock judgements to test the timing and transition of IOR, facilitation and re-entrant processing. Although we observe robust IOR at late cue-target onset asynchronies for reaction times, we do not observe an effect of IOR on the perceptual onset times suggesting that IOR is not perceptual. In contrast to Krüger et al. (2004) we do not observe an effect of the cue when it appears immediately before or after the probe neither in RTs nor in perceptual judgements. We conclude that cuing effects are not measured with a spatio-temporal gradient or are susceptible to continuous probability manipulations.

ABSTRACT. We investigated how the mean and variance of the background element size distribution affect the perceived size of a target. The target was a disk (diameter 240 arcmin) on a frontoparallel plane. The background texture consisted of 5000 randomly distributed disks whose diameter was drawn from distributions with various mean (60 to 600 arcmin) and standard deviation (0 to 0.27 fold of the mean). We used a two-interval forced-choice paradigm to measure the perceived target size at various background textures. In each trial, the target with a background was presented in one interval while a reference disk on a blank in the other interval. The observers' task was to determine which interval contained a larger disk. We measured the point of subjective equality (PSE) for the perceived target size with a staircase procedure. Regardless of the variance, the perceived target size decreased with mean background disk size until it reached 360 arcmin. After that the perceived target size changed little with further increase of mean background disk size. Our result shows that only the first order statistics, but not the second order statistics of the background can modulate the perceived target size and this modulation is nonlinear in nature.

ABSTRACT. There are different opinions about the roles of local interactions and central processing capacity in visual search. This study attempts to clarify the problem using a new version of relevant set cueing. A central pre-cue indicates two symmetrical segments (that may contain a target object) within a circular array of objects presented briefly around the fixation point. The number of objects in the relevant segments, and density of objects in the array were varied independently. Three types of search experiments were run: (1) search for a simple visual feature (color, size, and orientation), (2) conjunctions of simple features, and (3) spatial configuration of simple features (rotated Ts). For spatial configuration stimuli, the results were consistent with a fixed global processing capacity and standard crowding zones. For simple features and their conjunctions, the results were different, dependent on the features involved. While color search exhibits virtually no capacity limits or crowding, search for an orientation target was limited by both. Results for conjunctions of features can be largely explained by the results from the respective features. This study shows that visual search is limited by both local interference and global capacity, and the limitations are different for different visual features.

ABSTRACT. In real-world environments, relevant objects can appear rapidly and unpredictably. Research by Eimer and Grubert (2014) used event-related brain potential (ERP) markers to show that attention can be allocated rapidly and in parallel to new target objects defined by colour. The current research aimed to determine whether this phenomenon also exists for targets defined by alphanumeric category (letters versus digits). Two displays were presented in rapid succession, with a stimulus-onset asynchrony (SOA) varying between 10, 20, 50 and 100ms. Each display contained one item in the current target category and another item in the nontarget category on opposite sides. Participants had to compare the identity of the target-category items in the two displays (e.g., vowels vs. consonants) and to ignore the nontarget items. The speed of category-based attentional selection was measured using the N2pc component. N2pc onset latencies to targets in the first and second display closely matched the objective SOA between the two displays, demonstrating that attention can be allocated rapidly and in parallel to new category-defined targets. These results show that rapid parallel target selection processes are not confined to simple target-defining features such as colour, but can also be elicited when selection is based on category.

ABSTRACT. After making an action, visual sensitivity modulates rhythmically, within the theta range of frequencies. At low luminances, the frequency of endogenous brain-rhythms increases, reflecting increased alpha power. On the other hand, visual processing slows at low luminances (Pulfrich effect). We measured oscillatory perceptual effects at high and low luminances to test whether their frequency is governed by sensory mechanisms or endogenous brain rhythms. In separate sessions at photopic (38 cd/m2) and mesopic/scotopic (1 cd/m2) luminances, gabor-patches were briefly displayed to subjects after a random delay after a self-initiated button-press: they reported in 2AFC whether the contrast increment was in upper or lower half. In a third control condition, the stimulus was delivered by the computer at random intervals. The results revealed significant theta-oscillations in both button-initiated conditions, but none in the control condition. Importantly, the oscillations at low luminance were significantly higher in frequency, 5.0 Hz compared with 4.5 Hz (p<0.05). However, at low-luminance the phase led by about 15 ms ahead of that at high-luminance. These results show that while low luminance retards visual processing, causing a phase-lead, the frequency of visual oscillations seems to be determined by endogenous brain rhythms, which increase at lower luminance.

ABSTRACT. It has been proposed that the visual system uses background distortions as a cue to estimate the refractive index of thick transparent objects. We have previously argued that it is theoretically implausible that such estimates can be made and that approximately correct refractive index matches that putatively support this claim can alternatively be explained more parsimoniously by assuming direct matches of image attributes. In the present investigation the subjects had to match the material of two transparent objects, but, in contrast to previous studies, we used different background texture densities for standard and test and manipulated the visibility of the undistorted part of the backgrounds. The results suggest that the subjects simply maximized the similarity of the image patterns inside the object boundaries and thus referred to information that is insufficient to estimate the objects’ refractive indices. Subjects can also take the undistorted part of the background into account if this is suggested by experimental layout or instruction, but they do this in a rather simplistic way that also seems inappropriate to estimate refractive indices of transparent objects. These results provide additional evidence against the assumption that the visual system estimates the refractive index of transparent objects from background distortions.

ABSTRACT. At visual fixation, gaze cues strongly influence an observer’s attention and response preparation. Using an interference paradigm, we have shown that gaze cues presented away from fixation are much less effective. Instead, strongest response cueing is found for pointing hands and rotated heads (Burton, Bindemann, Langton, Schweinberger & Jenkins, 2009; Hermens & Bindemann, 2014). These hand and head cues have a clear shape outline and therefore the results suggest that the strength of extrafoveally presented cues is determined by the shape of the cue. Here, we present data using the same interference paradigm and show that extrafoveally presented arrow cues (symbolic cue; distinct shape) result in stronger response cueing than eye gaze cues and direction words (social and symbolic cue, respectively, no distinct shape). These results further support the notion that cue shape rather than social status determines the strength of extrafoveally presented cues.

ABSTRACT. Inflight correction of saccades remains a topic of debate, as saccades are too short for peri-saccadic visual signals to be processed and to subsequently manipulate saccade path. To test this, we measured saccades to targets randomly stepping or not after saccade onset, and compared the trajectories.

10 subjects’ eye movements were recorded (Eyelink2000, 1000 Hz). After central fixation (1.5 sec), a saccade target (135 deg, distance 8.9 deg from fixation) appeared. Saccade onset triggered target displacement (0.7 deg leftwards). Sample-by-sample Step/NoStep-trajectory distances were calculated by randomly pairing trials (bootstrapped) within each session (min. 150 trials). Using t-tests we estimated when trial types started to differ.

Saccade trajectories significantly differed between step and no step trials, with step-saccades landing closer to stepped targets. Trajectory differences arose shortly after the velocity peak at ~20 msec after saccade onset, with average velocities of ~400 deg/sec.

Step information was not available before saccade onset. As visual signals reach cortex with a latency of ~50 msec (Schmolesky et al. 1998), our results indicate that subcortical pathways may modulate saccading performance at least to apply small deviations from an initial path, or to switch between two alternate, trained paths. We discuss possible substrates for this effect.

ABSTRACT. Classical models of exogenous attention suggest that attentional enhancement at the focus of attention degrades gradually with distance from the attended location. Here we present first evidence in support of a Mexican Hat shaped modulation in a visual acuity task. In two experiments participants had to decide the location of a small gap in a target circle that was preceded by a non-informative transient attention capturing cue. The distance between cue and target and the latency between their onsets were varied. Enhancement at the focus of attention was accompanied, at short latencies, by suppression at the surrounds, exhibiting a Mexican Hat shaped modulation of performance, whereas at latencies over 200ms performance decreased gradually with distance. Our results suggest that a rapid Mexican Hat modulation is an inherent property of the mechanism underlying exogenous attention. They further suggest that this modulation is rapid and that a monotonically degrading trend, such as advocated by classical models, develops only at later stages of processing.

ABSTRACT. Colour has variously been defined as (a) a real property of objects in the physical world, or a disposition to create certain spectral patterns of light; (b) a subjective experience, or category of experiences, with arbitrary quality; or (c) an emergent property of the interactions between the external world and the body or mind. But (b) and (c) can lead to paradoxes. Berkeley famously argued that 'secondary' qualities (e.g. colour, taste) do not differ metaphysically from 'primary' qualities (e.g. size, shape, hardness). Therefore if colour is not a real, mind-independent property then neither are size, shape, etc. Moreover since colour did not exist before there was consciousness then neither did size – nor rocks, the sun, or any object with primary qualities. Here, primary quality realism is defended by demonstrating Berkeley's reasoning was invalid. For example Berkeley believed only minds have causal powers; therefore external objects cannot cause sensory experiences; therefore belief in external objects (the bearers of primary qualities) is “a contradiction”. But Berkeley's argument includes his idealist axiom about the monopoly of mental causation as one of its premises, which we can reject. Thus the metaphysical status of one dimension of experience does not necessarily generalize to others.

ABSTRACT. Reaching toward a visual target displaced by laterally refracting prisms results in adaptation of limb and ocular proprioception, which individually and additively lead to prism adaptation (PA). The recalibration of oculomotor corollary discharge signals, from which orbito-ocular straight-ahead is derived, displaces perception of visual straight-ahead. The contribution of ocular proprioception to PA has not yet been directly measured. This research will be the first to utilise eye-tracking technology to measure eye position before, during, and after PA, and to compare those measures with a condition where eye muscle potentiation occurs in the absence of visual error feedback. Additionally, it will be the first study to correlate ocular proprioceptive changes with traditional PA after-effect measures of straight-ahead pointing (limb proprioception), open-loop pointing, and dual-limb passive proprioceptive pointing in both conditions. We found that visual adaptation does not occur in patients with lesions of intraparietal cortex. We therefore hypothesise that calibration of ocular proprioception with oculomotor corollary discharge in this brain region is used in computing visual straight ahead. With this novel work, we aim to better understand the role of ocular proprioception in the rehabilitative application of PA.

ABSTRACT. Dynamic stimuli capture attention, even if not in the focus of endogenous attention. Such a stimulus is apparent motion, given that it benefits perception of targets in the motion path. These benefits have been attributed to motion-induced ‘entrainment’ of attention to expected locations (spatial extrapolation) and/or expected time-points (temporal entrainment). Here, we studied the automatic nature of spatial extrapolation versus temporal entrainment with apparent motion stimuli, when motion was task-irrelevant. Participants performed an endogenously cued target detection task, in which symbolic cues prompted attention shifts to lateralized target positons (75% validity). Simultaneously, apparent motion cues flickered either rhythmically or arhythmically across the screen, such that targets appeared either in or out of motion trajectory. Although the motion cue can be considered a distractor (non-informative as to target location), motion direction influenced target detection, which is in line with automatic extrapolation of spatial positions during apparent motion. An effect that was independent and additive to the endogenous cueing benefit. Importantly, temporal cueing in the motion stream also influenced target detection. However, this effect was independent of reflexive motion-cueing to spatial positions. We conclude that spatial extrapolation and temporal entrainment of attention by apparent motion are governed by partially independent reflexive mechanisms.

ABSTRACT. Express saccade makers (ESMs) produce high proportions (>30%) of low latency (80–130ms) express saccades (ES) in prosaccade overlap tasks; a behaviour that is stable over time (Knox & Wolohan, in press). ESMs also produce higher proportions of errors on the antisaccade task, compared to non-ESMs (Knox et al. 2012); although this is not due to a deficit in oculomotor inhibitory control (Wolohan & Knox, 2014). The relationship between this saccade behaviour in ESMs and personality traits is unknown. We investigated self-reported impulsivity and the occurrence of ES. Healthy participants (N=138) classified as ESMs (n=43) or non-ESMs (n=95) based upon their performance on prosaccade overlap tasks, completed the UPPS questionnaire (Whiteside & Lynam, 2001) which distinguishes five independent dimensions of impulsivity. These dimensions as predictors of ES% were explored using a linear regression analysis. This model indicated no relationship between ES% and any of the dimensions (R2=0.03,F(5,137)=0.79,p =0.56; all dimensions p>0.05). Thus while associations between other aspects of oculomotor behaviour and impulsivity have been observed (Cirilli et al., 2011), and relationships between inhibitory control and impulsivity have been suggested (Roberts et al., 2012), we find no evidence to link impulsivity with the high levels of ES production in ESMs.

ABSTRACT. We investigated whether different spatial frequencies are equal in competition for attention. Digital face (184) and object (63) photographs were used. Images were equalized in mean luminance and size (7 angle degrees). They were spatially filtered by 5 filters with central frequencies 0.5, 1, 2, 4, 8 cpd and bandwidth 2 octaves. Two differentially filtered images of the same face or object were simultaneously presented on two screens during 1 s (the eccentricity was 7 ang.deg.). Then 4 unfiltered images were shown. The observer was to determine one that matches to the test stimulus and press the corresponding key. Prior to each trial the observer fixed gaze at the cross-mark located amidst two screens. Then he/she run test stimuli presentation. We determined the image that caused an attentional capture by electrooculogram.  It was revealed that the spatial frequencies differ in ability to capture attention and form the significant order. There is a hierarchy of priorities headed by 2 cpd. It was shown that this order does not change when the categories of identifying objects vary. We assume that this phenomenon is based on distinct activation of different spatial-frequency pathways and this mechanism takes part in formation of saliency maps.

ABSTRACT. There has been considerable interest in understanding the capacity of colour to shape our functioning and behaviour (Elliot & Maier, 2014). For example, performance on intelligence tests is found to be poorer after exposure to the colour red, compared to green (Elliot et al., 2007). One possibility is that such effects arise through colour-concept associations (i.e., red - failure). Furthermore, the symbolism of a colour may vary across cultures. However, this has not been systematically investigated. The present investigation explores the range of abstract colour associations that English and Chinese speaking adults have, using hues from the World Colour Survey (WCS: Berlin & Kay, 1969, 1991). In a series of two experiments, we explored: the associations that people have with each of the 11 Basic Colour Terms (black, white, red, yellow, green, blue, brown, purple, pink, orange, grey; Experiment 1); and which specific physical colours are associated with the concepts generated in Experiment 1 (Experiment 2). The findings have implications for better understanding effects of colour on behaviour and enable us to make and systematically test predictions about the direction of colour-behaviour effects across colour space.

ABSTRACT. Dissent on perceived colour and colour combinations can often be found in everyday life, particularly with regard to fashion. In the case of the low-quality photograph of an actually blue-black lace dress posted on the internet, this dissent became a viral phenomenon fascinating people around the globe: While some perceived what is now known as #TheDress as being blue-black, others saw it as white-gold, and others again can switch between both impressions. Soon, a huge debate on the underlying reasons developed and various hypotheses were put forth in the media, some of which we tested in a series of experiments. Our results show that about 50% of the people interpret #TheDress as blue-black, the others as white-gold, while both groups show high internal consistencies across different presentation conditions. We further revealed that blue-black perceivers show a slightly higher sensitivity for blue and are stronger affected by visual cues for shades and the source of illumination. This was indicated by strong effects of scrambling the depiction. Changing the background colourisation/light scene was further related to small modulations of the effect, which indicates that anchoring might play its role as well. In sum, #TheDress nicely demonstrates the cognitive nature of (colour) perception.

ABSTRACT. Fixation-aligned pupillary response averaging uses gaze data to identify events of interest which occur at unpredictable times during a task, such as the detection of a target during visual search. Proof of principle for this technique was given by Klingner (2010) though its validity remains questionable due to the lack of statistical comparisons between subjects, the use of a remote eye tracker and the lack of consideration for the gaze-dependent measurement error (GDME) in video-based eye trackers. Using a tower-mounted eye tracker and a routine described by Brisson et al. (2010) to correct the GDME, we tested the ability of fixation-aligned pupillometry to resolve the transient pupil dilations that are known to accompany target detection during visual search. Participants searched for target “C”s among heterogeneously oriented distractors and reported the number of targets present at the end of each search (0, 1, 2 or 3). Interest areas assisted the identification and matching of target and control fixations and the extraction of the subsequent 3000ms pupil time course. Preliminary analysis has revealed that target fixations consistently elicit a pupil dilation that peaks around 800ms after fixation onset and withstands correction for the gaze-dependent measurement error.

ABSTRACT. On the basis of measurements of the perceived coherence of superposed drifting gratings, Krauskopf and Farell (1990) proposed that motion is analysed independently in different chromatic channels. They found that two gratings appeared to slip if each modulated one of the two ‘cardinal’ colour mechanisms S/(L+M) and L/(L+M). If the gratings were defined along intermediate colour directions, observers reported a plaid, moving coherently. Since achromatic gratings are less likely to cohere if they differ in speed, we hypothesised that slippage might occur in chromatic gratings if the motion signal from the S/(L+M) channel is weak and equivalent to a lower speed. We asked observers to judge coherence in two conditions. In one, S/(L+M) and L/(L+M) gratings were physically the same speed. In the other, the two gratings had perceptually matched speeds. We found that the relative incoherence of cardinal gratings is the same whether gratings are physically or perceptually matched in speed. Our hypothesis was firmly contradicted. In a control condition, participants judged the ‘coherence’ of stationary gratings. Surprisingly, the difference in judged coherence between cardinal and intermediate gratings remained as strong as it was when the gratings moved. Krauskopf and Farell's result may not arise from motion processes.

ABSTRACT. Oversight is a common cause of ship accidents (Marine Accident Tribunal 2013). We assumed that visual attention played a role in these oversights and examined the effects of a dual-task, body roll, and body pitch on the visual attention of 31 participants viewing a video depicting a scene from the Ship Handling Simulator. The participants were asked to perform a simple-task and a dual-task. The simple-task involved pressing a button as soon as the target presents in the picture was detected. The dual-task involved carrying out the simple-task while performing mental calculations. We used the participants’ reaction time as an index of visual attention. Body roll and body pitch were controlled by means of the motion base, which moved sinusoidally with an amplitude of 5 or 10 degrees of arc and a frequency of 0.07, 0.10, or 0.20 Hz. The results showed that the body pitch had an effect on the reaction time only when the motion base was vibrated at a frequency of 0.10 Hz in the dual-task, and that body roll had no effect on the reaction time in either task. These results suggest that body pitch at a medium frequency affects visual attention in the dual-task.

ABSTRACT. According to the predictive coding framework, brain generates a predictive model of statistical regularities in the environment. Attention is conceptualized as a measure of uncertainty in prediction error. (Clark, 2013). We employed a novel paradigm to empirically study selective attention under predictive coding framework. In two experiments, participants completed a visual search task (display size 8) where they searched through either a spatially fixed configuration (FC) or a configuration in which one of the items were displaced from the fixed configuration (DC), making the displacement a prediction error. In DC, the displaced item was also the target in half of the trials (Exp1) or only 1/8th of the trials (Exp2). If attention is indeed a measure of uncertainty in prediction error then, compared with the fixed configuration, search should be faster when the moved item was the target and slower when it was the distractor. The result showed that the search times were fastest when the shifted element was the target and least when it was distractor, in both the experiments providing empirical support for predictive coding hypothesis. Thus, it seems that attention is indeed the precision weighted on the discrepancy from the predicted model.

ABSTRACT. How does the human brain code shape? Theories abound, but a data-driven method may allow us to discover principal dimensions of this encoding. The classification image method has been applied effectively to contrast discrimination – can it also be applied to shape? A 2D shape can be represented as a cycle of points in the plane or alternatively as a vector of complex Fourier Descriptor (FD) coefficients (Zahn & Roskies, 1972). Observing that natural shapes are low-pass in this representation, we hypothesize that it may prove an efficient basis for classification image analysis. In a yes/no task, three observers were asked to distinguish animal shapes from an ellipse, both corrupted in the FD domain with additive Gaussian noise matched to the low-pass spectral density of the shapes, thus rendering each frequency equally informative for the task. Despite this noise equalization, the resulting classification templates are tuned primarily to lower frequencies. The templates estimated for each observer were found to be significantly more predictive of their responses than ideal templates (p=.001). Estimated templates were also more predictive than ideal templates when predicting across observers (p=.03). These results attest to the utility of the classification image methodology for shape discrimination.

ABSTRACT. Saccades are used to reallocate attention to new targets. Because attentional processing and saccade behavior change in the presence of a distractor (e.g., RTs are slower and saccades exhibit greater curvature), these two systems may be fundamentally linked. Although it is known that attentional processing unfolds within surface-based representations, such that search for a target is unaffected by distractors on a different surface (e.g., Atchley et al., 1997), it is not known whether the saccade motor plan – executed in 2 dimensions – is also sensitive to 3D surface information. We ask whether saccades to targets are sensitive to surface structure. Exps. 1-3 manipulated the presence and absence of a distractor and different kinds of surface structure. Although saccades curved more in the presence of distractors, there was no main effect of surface structure even when surfaces were task-relevant, nor did surfaces modulate the effect of the distractor. In Exp. 4, however, oculomotor capture by a distractor singleton was significantly reduced when participants knew in advance on which surface (rather than which locations) a target would appear. We tentatively conclude that both saccade programming and attentional selection benefit from surface structure only under high selection demands.

ABSTRACT. How do people use the structure of items when storing them in visual memory? Experiment 1 asked what format visual working memory uses to encode objects and their structure. Subjects saw objects arranged in different spatial clustering structures and recalled their positions. Objects in the same cluster were misreported in similar directions, indicating that memory errors were shared within clusters. Additionally, the shared errors for clusters decreased when clusters were closer. These results are captured by a model that encodes object positions relative to an inferred grouping structure and recalls relative positions with Weber noise. Experiment 2 adopted an iterated learning paradigm to amplify biases due to people’s prior expectations about spatial structure. Each subject saw 15 items and reported their positions; critically, the positions one subject reported served as the stimulus for the next subject. People converged to reporting items in few groups that are either tight clusters or lines, and multiple lines in a display with similar orientations and lengths. This effectively recovers visual memory’s use of Gestalt principles to encode objects. Together, these results show how people use environmental structure to remember displays: what structures they expect and exploit, and what format encodes objects and their structure.

ABSTRACT. Early stages of information processing are thought to be carried out by static feature-detectors that operate independently of high-level representations of a stimulus. This notion is challenged by new perspectives that emphasise the role of feedback connections in shaping the properties of early visual function. Any experimental design trying to test this hypothesis has to ensure that sensory stimulation of early detectors remains identical while high-level representations of the stimulus are manipulated. Here, we used two-tone images of objects in a psychophysical procedure to achieve this and demonstrate a significant influence of high-level object knowledge on response properties of early edge-detectors. Two-tone images feature physically homogenous regions that are experienced as an object contour (somewhat similar to illusory contours in Kanizsa figures) once relevant image information is provided. We embedded edge-elements in these areas that were consistent or inconsistent with the high-level contour representation and measured absolute contrast detection-thresholds before and after providing relevant object knowledge. Results indicate that prior knowledge facilitates absolute contrast-threshold for edge elements that are consistent with the high-level representation of a stimulus. Supported by computational simulations, these findings indicate that response properties of early information-processing units are shaped by top-down modulation from high-level image representations.

ABSTRACT. Local visual elements are grouped into global objects seemingly without effort. Using a contour integration task and EEG source level analyses, we tested the hypothesis that visual perceptual grouping requires a top-down selection, rather than a passive pooling, of neural activity coding the local information. The participants were presented visual displays with or without a hidden contour. Two tasks were performed: a central luminance-change detection task and a peripheral contour detection task. Only in the contour-detection task could we find differential brain activity between contour and non-contour conditions, within a distributed brain network including parietal, lateral occipital and primary visual areas. Contour processing was associated with an inflow of information from lateral occipital into primary visual regions, as revealed from the slope of phase differences between source level oscillations within these areas. The findings suggest that contour integration results from a selection of neural information from lower visual areas, and that this selection is driven by the lateral occipital cortex.

ABSTRACT. Neurons at early stages of the visual cortex signal elemental features such as pieces of contour, but how these signals are organized into perceptual objects is unclear. Theories have proposed that spiking synchrony between these neurons encodes how features are grouped (binding-by-synchrony), but recent studies did not find the predicted increase in synchrony with binding. We propose that features are grouped to “proto-objects” by intrinsic feedback circuits that enhance the responses of the participating feature neurons. This hypothesis predicts synchrony exclusively between feature neurons that receive feedback from the same grouping circuit. We recorded from neurons in macaque visual cortex and used border ownership selectivity, an intrinsic property of the neurons, to infer if two neurons are part of the same grouping circuit or not. We found that binding produced synchrony between same-circuit neurons, but not between other pairs of neurons, as predicted by the grouping hypothesis. In a selective attention task, synchrony emerged with ignored as well as attended objects, and higher synchrony was associated with faster behavioral responses, as would be expected from early grouping mechanisms that provide the structure for object-based processing. Our results suggest a novel coding mechanism that might underlie the proto-objects of perception.

ABSTRACT. For the first time we report that adaptation to optic flow induces robust crossmodal self-motion aftereffects in the vestibular domain. We assessed the dependence of these aftereffects on adaptor duration and their relationship with visual motion aftereffects. A motion nulling procedure was employed to measure the inertial (i.e. vestibular) stimulus needed to cancel the aftereffect induced by a visual-only optic flow. Experiments were conducted using a motion platform, equipped with a 3D monitor. On each trial, subjects experienced a visual-only adaptor consisting of expanding or contracting optic flow, followed by a 2-seconds physical movement on the fore-aft axis. Then they indicated the perceived direction of the physical movement. Exposure to optic flow stimulation lasting 15-seconds induces a shift of the point of perceived zero-motion (i.e. an aftereffect) whereas shorter adaptor durations were ineffective, suggesting that sustained perception of self-motion is required to elicit these aftereffects. Magnitudes of crossmodal and visual aftereffects were not correlated, suggesting separate underlying mechanisms. These crossmodal aftereffects likely result from adaptation of multimodal neural mechanisms specialized for self-motion processing. Transfer of adaptation across modalities provides behavioral evidence for neural systems that constantly function to calibrate sensory modalities with each other as well as with the environment.

ABSTRACT. Purpose. Optic flow does not allow to distinguish ego-motion from object depth and thus leads to erroneous ego-motion estimates in tunnels with varying diameters (Festl, Recktenwald, Yuan, Mallot 2012). Mathematical optic flow theories predict that additional feature- or object-based stereo cues might support veridical ego-motion perception. We tested this prediction with human subjects. Methods. Three experiments are reported: (i) yes-no judgments for the presence of ego-acceleration in straight, narrowing, and widening corridors defined by limited lifetime random dots; (ii) maintenance of constant speed in tubular tunnels with varying diameter (adjustment task); and (iii) yes-no judgments for the presence of ego-acceleration in straight, narrowing, or widening alleys passing through rows of 3D blocks. Results. In all experiments, ego-acceleration judgments were affected by the change of corridor diameter, even though depth could be assessed independently by stereoscopic cues. Results are in quantitative agreement with the assumption that ego-motion is judged from overall radial flow, i.e. from matching retinal flow to a radial template. Conclusion. Independent depth information that would be useful in technical optic-flow algorithms is not used in human ego-acceleration judgments. We suggest that vectional ego-acceleration and environmental depth structure are independent perceptions.

ABSTRACT. Research on biological motion perception has focused on 2D motion and form cues, while shading cues were largely neglected. Using point-light stimuli with volumetric elements, we discovered a new visual illusion, where the change of the light source direction flips the perceived walking direction. We extended a neural model for biological motion perception (Giese, 2003) to account for this illusion. METHOD: Walker stimuli consisted of 3D conic shapes that were centered on the limbs and illuminated from different light-source positions. We studied the dependence of the perceived walking direction on the position of the light source. We extended the model by a new pathway, which processes shading cues by analysis of inner brightness gradients. RESULTS: The light source position has a strong influence on the perceived walking directions (F(16,176)>178, p< 0.01), where illumination from below results in a flip of the perceived walking direction by 180 deg compared to the veridical direction. The model reproduces this illusion. A control experiment reveals the critical shading features that cause the illusion, consistent with predictions from the model. CONCLUSION: Biological motion perception is influenced by a lighting-from-above prior, similar to the perception of static shapes (Brewster, 1874; Ramachandran, 1988).

ABSTRACT. Many biologically important motions are described by periodic trajectories. Previously we introduced radial frequency (RF) trajectories in which the motion of a difference of Gaussians (DOG) stimulus differed from a circular trajectory by having the radius vary sinusoidally as a function of polar coordinate angle. Here we explore the hypothesis that discrimination of these trajectories involves global spatio-temporal processes that depend on motion continuity. To test this hypothesis RF3 and RF4 trajectories were presented such that there was smooth motion during each single cycle, but the moving DOG jumped discontinuously to a different position at the end of each cycle. For example, if the cycles in an RF3 trajectory were traversed in the sequence 1-2-3 for continuous motion, discontinuous motion would be defined by the temporal sequence 1-3-2. Significantly, both continuous and discontinuous stimuli traverse the same spatial points in the same period of time. Thresholds for discrimination of discontinuous RF motion from discontinuous circular motion were compared to thresholds for continuous motion. The discontinuous trajectory motion increased thresholds by a factor of approximately 2.0 across conditions. These results provide evidence for global processing of periodic RF motion trajectories.

ABSTRACT. Would escaping from the Earth’s gravity alter human visual perception? On the Earth, people are less sensitive to upside-down biological motions (BM) whose motion violates the effect of gravity compared with the upright ones whose motion carries natural gravitational acceleration. Such inversion effect has been taken as evidence for the visual system’s exquisite sensitivity to gravitational motion cues. Here we reported that the inversion effect of BM perception declined after prolonged exposure to space microgravity environment (2-week spaceflight) or simulated microgravity condition (45-day -6° head-down bed rest, HDBR). Moreover, HDBR resulted in a substantial reduction in neural response to upright versus inverted BM in the STS, a brain region dedicated to BM processing, as well as a perception-related increase in the functional connectivity between the STS and the key vestibular cortex that is involved in multisensory gravitational computation. Together, these results show that altered vestibular input due to microgravity leads to a re-adaptation of the brain’s response to visual gravitational motion cues. The Earth’s gravitational environment that we undergo may serve to maintain our brain’s selective sensitivity to visual gravitational motion cues and enforce the internal law of gravity on visual perception.

ABSTRACT. We effortlessly interact with moving objects despite being hopeless at visually judging acceleration. Moving objects that we encounter often accelerate due to gravity and friction, or to other people’s actions such as handing us the object. How do we deal with these accelerations? Misjudging or ignoring acceleration can be compensated for by continuously adjusting movements on the basis of the latest visual information, but this mechanism is limited by visuomotor delays. Our study was designed to verify that people ignore acceleration, and examine whether it is enough to observe the motion to learn to fully compensate for this, or whether it is essential to interact with the object. We asked subjects to try to tap on targets moving at various randomly interleaved speeds. In alternating blocks of trials, the targets either accelerated or decelerated (at 1m/s2). On the first trial of each block the tapping error was consistent with the new acceleration (or deceleration) being ignored. Within a few taps the error was negligible. Only observing the target’s motion hardly influenced subsequent performance. Thus, people do not learn the object’s acceleration, but they learn how to adjust their movements to compensate for the error that arises from ignoring the acceleration.

ABSTRACT. Spatial crowding refers to impaired target identification when it is surrounded by flankers in space. In this study we demonstrate, for the first time with normal observers, temporal crowding: impaired target identification when it is surrounded by other stimuli in time. We presented a sequence of 3 letters to the same location, right or left of fixation, separated by varying ISI (100-450 ms). One of these letters was an oriented T. Observers indicated the T's orientation. We found that target identification was impaired by preceding and succeeding stimuli up to an ISI of 300 ms. Interestingly, we did not find an interaction between spatial and temporal crowding. In the second part of this study we examined whether spatial attention can alleviate temporal crowding, as it does with spatial crowding. Attentional precues (auditory or visual) that attract attention to the letters’ location were added to the basic paradigm. As expected, with both types of cues precueing improved overall performance when the target appeared in the first display. Moreover, although with the auditory cue there was no cueing x ISI interaction, such an interaction emerged with the visual cue, suggesting that as with spatial crowding, spatial attention can reduce temporal crowding.

ABSTRACT. Peripheral vision can be severely impaired by nearby clutter. Decades of research using sparse displays have established that this phenomenon, known as visual crowding, follows Bouma’s law: Interference occurs for target-distractor separations up to half the target’s eccentricity. Although considered a fundamental constraint on human vision, Bouma’s law has never been tested in more dense visual environments. Using a genetic algorithm we investigated crowding in densely cluttered displays. Participants were instructed to identify the orientation of a target line (6° eccentricity) among 284 distractor lines. Displays supporting highest accuracy were selected (“survival of the fittest”) and combined to create new displays. Performance improved over generations, predominantly driven by the emergence of horizontal flankers within 1° of the near-vertical target, but with no evidence of interference beyond this radius. We conclude that Bouma’s law does not necessarily hold in dense displays. Instead, a nearest-neighbour segmentation rule provides a better account.

ABSTRACT. Visual crowding impairs recognition of flanked objects in the periphery. Further, it is known that low-level feature differences between such objects (e.g., colour) weakens crowding. Recently, higher-level differences have also been posited to similarly modulate crowding: crowding has been shown to be weaker if objects differ in category membership or if the objects are familiar (meaningful). However, these effects of category and familiarity can be deemed to be higher-level influences only if all low-level differences between stimuli are controlled for, which has not been the case in previous research. In a series of experiments, we investigated the effects of a range of low-level attributes on higher-level crowding interactions. We found that the ostensible category effect was fully accounted for by differences in features and overlap between target and flanker stimulus sets. However, the effect of familiarity was persistent even when low-level attributes were controlled for. Thus, at least some of the so-called higher-level crowding interactions are a result of inadequate stimulus control, while others seem robust. Therefore, we strongly encourage caution and a thorough control of stimulus-level properties before higher-level effects are claimed to influence crowding.

ABSTRACT. While the reduced ability to identify crowded objects in the periphery continues to be thoroughly studied, the specific nature of the perceptual degradation remains elusive. We utilized a novel drawing paradigm to probe the phenomenology of peripheral vision, using the Rey-Osterrieth Complex Figure. The figure was presented at 12 or 6 degrees in the right visual field. Eye tracking ensured that the figure was only presented when observers maintained stable fixation. Observers were asked to draw the figure with free viewing, capturing its peripheral appearance. To assess the drawings, two raters used a standard scoring system that evaluated feature positions, spatial distortions, and omission errors. A linear model revealed several characteristics of the errors. Error rates were higher at 12 compared to 6 degrees. Internal features which were adjacent to similar shapes were more likely to be omitted, especially at 12 degrees. Highly salient components that differed from surrounding shapes were generally depicted correctly at both eccentricities. We discuss methodological advantages of the drawing method over traditional forced-choice procedures, and present our empirical results in relation to appearance and error distributions in crowded peripheral vision.

ABSTRACT. An object in the visual field will activate populations of neurons tuned to its various features. Theories of 'local sign' predict that its position is perceived via the centroid of this activity. In contrast, the perception of other features (motion, orientation) has been shown to depend on the task. We sought to measure similar task dependencies in position perception. First, following adaptation, the detection of a small target dot (at 8deg. eccentricity) was most impaired by adaptors that overlapped the target. Discrimination of the same dot in a three-dot Vernier stimulus ('was it left or right of the references?') was instead most impaired by adaptors displaced away from the target. Second, in a continuous-estimation paradigm, the perceived position of a target dot was largely veridical during detection judgements. During discrimination, the same dot appeared to be repulsed away from the decision boundary. Both results can be attributed to observers relying on the most informative channels for each task: those tuned to the stimulus (for detection) and those tuned away from the decision boundary (for discrimination). We conclude that perceived position is not simply given by the sensory distribution but rather reflects the read-out of these signals in a task-driven context.

ABSTRACT. Li and Atick (1994) presented a theory of efficient binocular encoding in which the two eyes’ signals are combined using separately adaptable binocular summation and differencing channels. We designed a dichoptic test stimulus for which the summation channel sees a grating tilted in one direction (clockwise or anticlockwise of horizontal), and the differencing channel sees a grating tilted in the opposite direction. The observer’s perceived direction of tilt (summation or difference direction) should depend on the relative sensitivities of the two channels. We manipulated channel sensitivity using adaptation. In correlated adaptation, each eye received the same image, which selectively adapted the summation channel; in anticorrelated adaptation, each eye received the photographic negative of the other eye’s image, which selectively adapted the differencing channel. These adaptation stimuli had equal energy at all orientations. Despite being isotropic, the adaptors influenced perceived tilt: The test stimulus usually appeared tilted in the difference direction after correlated adaptation, and usually appeared tilted in the summation direction after anticorrelated adaptation. This counterintuitive finding of a tilt aftereffect from isotropic adaptors is analogous to May, Zhaoping and Hibbard’s (2012) finding of a motion aftereffect from static adaptors. These two results strongly support Li and Atick’s theory.

ABSTRACT. The degree of efficiency with which observers find a target object in a visual scene depends both on the features of the target and on the structure of the background elements. Dissimilarity between target object and surrounding context elements usually increases search efficiency, and so does similarity between the individual context elements. In several experiments, we investigated the role of context element homogeneity in the deployment of visual attention. We hypothesized that context homogeneity modulates attention guidance in addition to other attentional control mechanisms such as the observer’s current intention. We used texture stimuli composed of simple line elements that were arranged in such a way as to form homogeneous or heterogeneous contexts. Observers were to search for and report an orientation target. Behavioral search efficiency (accuracy and response times) as well as the neural processes (such as the N2pc component, an attention-related marker in the event-related brain potential) involved in target and background processing were measured. Results showed that (pre-attentive) context grouping preceded target detection and had a strong impact on the way visual attention was deployed in the scene. The relative contribution of top-down and bottom-up processing on attention deployment depended on the homogeneity of the context elements.

ABSTRACT. Eye movements, attention and adaptation change and adjust visual sensitivity continuously. In particular, during the execution of saccades, visual sensitivity is reduced by an active process suppressing low-frequency luminance stimuli, but not isoluminant color stimuli (Burr, Morrone & Ross, 1994). During smooth pursuit, we found a small sensitivity decrease for the detection of low-frequency luminance stimuli, but a 16% sensitivity increase for color stimuli (Schütz, Braun, Kerzel & Gegenfurtner, 2008). We investigated whether these sensitivity adjustments interact for saccades to moving targets. As baseline, contrast thresholds were determined for the detection of red isoluminant lines flashed for 8 ms on a gray screen 2 degs above or below a central fixation point. For six observers detection performance for the horizontal line was measured during pursuit with step-ramps and during saccades with steps to stationary and moving targets at two fixed contrasts at different times relative to eye target onset. Contrary to our expectations observers showed for saccades to static and moving targets a marked saccadic suppression for the detection of isoluminant and high spatial frequency stimuli which was not present during pursuit. Whatever sensitivity improvement there is during pursuit, it is suspended during the initial saccade to a moving target.

ABSTRACT. The hypothesis that different oscillatory frequencies in Visual Evoked Cortical Potentials (VECP) should have different physiological roles, led us to investigate differences in the power of alpha, beta and gamma bands for achromatic and chromatic stimuli. Eight subjects underwent two four-minute recordings of 1Hz Pattern onset/offset VECP of the occipital pole for both achromatic and chromatic (red-green equiluminant) 2cpd sine-wave gratings. Separate spectrograms were built with Discrete Fourier Transforms for alpha (8-12Hz), beta (12Hz-30Hz) and gamma (30-60Hz) bands. For both color and luminance, two amplitude peaks were identified in all frequency bands: an early peak (~100-150ms) and a late peak (~450ms). The late peak is associated with the stimulus offset and does not differ between color and luminance signals for any of the frequency bands. The early peak differed between luminance and color in both amplitude and latency. Latencies were ~50ms earlier for color on average. Amplitude of the early peak was twice as high for color than luminance in the alpha and beta bands, and four times higher for gamma. Results support a different role of different frequency bands in color and luminance processing. Future work must approach the differential role of oscillatory potentials in luminance and color processing.

ABSTRACT. We measured suprathreshold brightness in different levels of edges hardness. Steven’s Power Law for circles modulating in luminance were estimated for 10 subjects (mean age 25 SD 3). Stimuli consisted of two circles of 3 degrees of visual angle, separated by 10 degrees. We tested 7 levels of Michelson contrast: 7, 8, 10, 15, 26, 50, 100. Three edge filtering was tested (0.3, 0.8 and 1.5 degrees of smoothing). The subject task were to judge the brightness of the edge filtered circle compared with the circle of hard edge which was considered the Modulus with an arbitrary level of 50, representing the amount of brightness perception. In each trial, the same contrast level was presented in both circles. Five judgments were performed for each contrast level in each edge filtering. We found an increase in the Power Law exponent as the increase of filtering (for sigma of 0.3= 0.37, sigma of 0.8= 0.55, and sigma 1.5= 1.03). All power function fitting had a correlation coefficient higher than 0.95. We conclude that there is a progressively increase in brightness perception as increase the edge filtering and a full performance in suprathreshold contrast perception is achieved with a half size stimulus smoothing.

ABSTRACT. Pupil diameter in human eyes plays an important role for clear vision like aperture of a camera. Pupil diameter, however, fluctuates around an optimal value even if a subject fixes one’s eyes on a target (Stark 1968). A part of pupil fluctuations is originated form changes in autonomic nervous activity (Yoshida, et.al., 1991). Recently, Hoeks, et.al. (1993) showed pupillary dilation can be a measure of attention and Wierda, et.al. (2012) also showed that high-temporal-resolution tracking of attention and cognitive processes can be obtained from the slow pupillary response using dilation deconvolution method. EEG analysis is effective for these research but there is no results simultaneously comparing both the pupil dilation and EEG studies. In order to study pupillary dilation during a fast-paced cognitive and attention task, we simultaneously recorded pupil diameter and electroencephalography (EEG) during the AB task as described in the literature (Wierda, et.al., 2012). Pupil size and EEGs were measured using the HS Eye Tracker (Cambridge Res. Syst.) and EEG-1200 (Nihon Koden Co.), respectively. Event-related potential (ERP) was extracted from the EEGs using averaging method. Finally, we discuss both the results of pupillary dilation and ERP of the response for the AB task.

ABSTRACT. Previous research has demonstrated that emotional pictures, compared to neutral pictures, can bias attention towards non-emotional targets in a different modality (Van Damme, Gallace, Spence, Crombez, & Moseley, 2009). The current experiment used a modified spatial cueing paradigm to test whether affective visual stimuli modulated auditory spatial attention, and which components of attention contributed to the modulation. Participants indicated whether an auditory target appeared either on the left or right, after brief (250ms) presentation of a spatially non-predictive peripheral visual cue (pleasant, unpleasant, or neutral). Responses to targets were faster when presented in the location of the preceding visual cue compared to in the opposite location (cue validity effect). The cue validity effect was larger for targets following emotional compared to neutral cues, but only for right-sided targets. There were no differences between emotional and neutral conditions on valid trials (targets in the same location as the cue) but on invalid trials (targets in the opposite location), participants showed delayed attentional disengagement when the target was on the right. We conclude that modulation of auditory attention by briefly presented visual emotional cues can be attributed to difficulties in moving attention away from visual affective cues presented in the left hemifield.

ABSTRACT. In “the dress” phenomenon of internet fame, striking individual differences in reported dress colours suggest individual variability in colour constancy. We assessed whether the phenomenon is peculiar to the photograph by investigating colour naming for the real dress. Participants viewed the real dress (blue/black) and another version (white/black) displayed singly under five illumination conditions, either two-source ( metameric ‘blueish’ daylight and candle light) or single-source (equal energy white, red or blue). Illuminations were generated with 10-channel LED luminaires. Participants (n=12) viewed the chest area of each item and freely named all colours. Naming variability was highest for the blue/black dress in the two-source condition. Colour name constancy was poor for both dresses; for the blue/black dress, all observers changed colour names for body, lace or both under at least one light change. Colour name constancy was best for the change from white to blue light for both dresses, but better for the blue/black than white/black dress. The white/black dress showed less constancy under the change from white to red light than from white to blue light. Thus, the phenomenon is likely linked to the “blue” bias for colour constancy (Pearce et al. 2014), but not restricted to the photograph.

ABSTRACT. Schelske & Ghose (VSS, 2015) reported an automatic user-identification system based on eye-movements and how it is affected by different grouping cues. Here we investigate whether the grouping cue present in the stimulus can in turn be predicted from eye-movements. Eye-movements of 8 participants were recorded while inspecting a Gabor-field for 10s in 120 trials. Their task was to indicate by key-press when a red dot appeared anywhere on the screen (randomly after 10-13s after stimulus onset). The stimuli contained either no-groups, weak-groups (detection-rate 65%) or strong-groups (detection-rate 82%) induced by either similarity of luminance or orientation of a subset of the Gabor-elements. A support vector machine (SVM) was trained to identify users based on twenty-five fixation, saccade and scanpath features computed from the eye-movement data. The identification accuracy was 71% for no-grouping, 78% for weak-grouping and 73% for strong-grouping (chance-level = 12.5%). Another SVM was trained to classify grouping cues. The accuracy for distinguishing grouping present vs. absent was ~ 65%, for luminance vs. orientation was 56% and was at chance-level (50%) for weak vs. strong cue. We conclude that grouping cues per se have strong but inherently different effects on eye-movements that can be utilized to train classifiers.

ABSTRACT. Sinusoidally-modulated flickering stimuli are typically used to draw temporal contrast-sensitivity functions (TCSF). However, due to the nature of a sinusoid, these functions don’t capture the differences in the perception of flickers with luminance increments or decrements. We generated flickering stimuli using impulses of transient increases or decreases in luminance at different levels of frequency to study distinct influences of contrast polarity on TCSF. Tree shrews were trained and human subjects were instructed to discriminate a flickering stimulus among two other iso-luminant non-flickering stimuli in an automated 3-alternative forced-choice task. We found that both tree shrews and humans exhibited higher sensitivity to contrast decrements compared to contrast increments. We show that the peak sensitivity for human subjects is at 15Hz. While human subjects showed less difference in the performance, they were faster in perceiving decrements than increments. In tree shrews, we report the largest gap between sensitivity to contrast increments and decrements at higher frequencies. Tree shrews exhibited peak sensitivities to contrast increments and decrements at 24-40Hz and 60Hz respectively. Taken together, our results support a higher perceptual acuity for luminance decrements compared to luminance increments and suggest inter-species differences in temporal characteristics of visual perception accounting for their ecological needs.

ABSTRACT. Inhibition of visual distractors usually follows the presentation of task-irrelevant stimuli. Recent claims suggest that inhibition can also be initiated through purely endogenous processes. We investigated this claim using a central cueing paradigm.

The cue indicated the likely location of either a target or a distractor. Targets appeared in only half of the trials. We examined how cue validity and distractors affected manual and saccadic response latencies to targets following the central cue.

We found: a) faster responses when the target appeared at the cued location, even when the cue was predictive of the distractor rather than target's location; b) response latencies were delayed in the presence of a distractor, except when the target was invalidly cued; c) orienting was faster, even when the target appeared alone, when one of the previous eight trials contained a target.

We conclude: 1) distractor suppression does not underlay the benefits of distractor cueing. 2) Reorienting to invalidly cued is remarkably robust in the face of distractor interference suggesting that during reorienting there is filtering of task–irrelevant stimuli, in keeping with fMRI findings. 3). Orienting to targets is maintained by a tonic, slowly decaying signal, most likely generated by previous target evoked responses.

ABSTRACT. Visual evoked cortical potentials (VECP) are useful for functional investigation of the chromatic visual system and to diagnose color vision deficits. Usually chromatic sinusoidal gratings are used as stimulus to elicit VECP, but it requires a long psychophysical protocol to match the perceptual brightness between their stripes. An alternative is to substitute them by pseudoisochromatic stimulus, which has a luminance noise inserted on its design and the perception of the target is dependent of the chromatic contrast. We compared VECP generated by sinusoidal and pseudoisochromatic gratings. Seven trichromat subjects were stimulated by chromatic sinusoidal gratings and illusory gratings in a pseudoisochromatic design shown at 0.33, 0.66, 1, 1.33, 1.66, and 2 cpd presented in pattern reversal mode (1 Hz) and pattern onset (300 ms) – offset (700 ms). The signals were recorded using surface electrodes, amplified x30000, digitized at 1000 Hz, and filtered between 0.1-100 Hz. Results. Pattern reversal: VECPs elicited by pseudoisochromatic grating had similar amplitude and latency compared to those elicited by sinusoidal gratings. Onset-offset: VECPs elicited by sinusoidal gratings had higher amplitude and earlier latency than those obtained with pseudoisochromatic stimuli. Different visual mechanism possibly underlying the cortical responses for illusory stimuli when presented in different modes.

ABSTRACT. Whenever people actively ignore stimuli in one half of the visual field, their alpha-band amplitudes measured over the contralateral occipital and parietal cortices increase. Since high alpha-band amplitudes have been associated with low cortical excitability, it is assumed that these stimuli are actively blocked from further processing. Only very salient stimuli might sometimes overcome this active inhibition and attract attention in a bottom-up fashion. This means that increasing distractor salience at an ignored location should impair target detection sensitivity at an attended location. To test this prediction, we asked participants to detect an at-threshold Gabor at one of two possible cued target locations. A neutral cue condition ensured that participants made use of the cues. When the target was accompanied by a distractor of one of three different contrast levels at the uncued location, detection sensitivity was unaffected by distractor contrast. Instead, the participants' response bias was affected by distractor presence. These results suggest that ignored distractors are not completely blocked from early processing stages, but rather contribute to response selection at later stages of visual processing.

ABSTRACT. By showing that verbal interference cancels the categorical perception of color (CPC), previous research essentially suggested that the facilitation of between-category color comparison is based on verbal coding. Those results were mostly obtained on memory demanding tasks, which favors verbal coding. In this study we investigated CPC in a task with no memory load and we combined that task with the simultaneous verbal interference, i.e. the Stroop paradigm. In two experiments with simultaneous color discrimination task (16 participants each) we measured participants’ RT. Color stimuli around the blue-green boundary formed 6 within-category and 6 between-category pairs. Stroop (congruent, incongruent or control-fixation crosses) was applied on the top square in Experiment 1, and on the bottom square (test or distractor) in Experiment 2. Despite the interference, CPC was significant both in Experiment 1 (F (1, 15) = 9.03; p <.05) and Experiment 2 (F (1,15) = 20.56; p<.001). Participants were slower when Stroop was presented on the bottom test square in relation to the distractor (F (1, 15) = 4.79, p <.05). The Stroop effect was not significant – both words and fixation crosses affected RTs. This result suggests that in simultaneous discrimination participants rely more on visual than verbal coding.

ABSTRACT. Light scatter in the eye can substantially reduce the dynamic range (DR) of the retinal signal. We quantify this effect by convolving high DR images with a point spread function that models eye scattering. We find that the resulting retinal DR can be described as the original DR raised to a power p. For images viewed in a dark background the exponent p is 0.68, while for image viewed with a dim background p is 0.45; this implies that a high DR image spanning seven orders of magnitude will span five in dark background conditions, but only three in dim background conditions. We also investigate the perceived quality of high DR images presented on an OLED monitor. We find that the highest quality is perceived when a dark background is used, but also note that the difference is most apparent when low key images are used. We investigate whether this effect can be explained by current image quality models.

ABSTRACT. Procedural texture models are widely used to generate textures. Previous studies proved that for near-regular textures, twelve perceptual features were sufficient to describe corresponding procedural texture models (Liu & Dong, 2013), i.e. a particular model can be found to generate textures with given perceptual descriptions. However, it is unclear whether these perceptual features are independent or correlated with each other. We analyzed the rating scales of the twelve features obtained from previous psychophysical experiments, and based on these scales, Pearson’s correlation coefficients are obtained with a range from 0 to 0.9670, meaning that there existed significant linear correlations between certain features. Meanwhile, 16.5% of the correlation coefficients had magnitudes less than 0.1, indicating that certain features were independent with the others. The most strongly correlated features were “locally oriented” vs “directional”, followed by “regular” vs “repetitive”. Linear regression models between significantly correlated features were constructed; the results showed that sample perceptual scales fit the regression line well. With these regression models, corresponding features can be predicted with few input perceptual scales. The predicted features can be used to accurately find procedural texture generation models, which were able to produce corresponding textures.

ABSTRACT. The gain and speed of smooth pursuit is known to degrade after the disappearance of a tracked target behind a mask (Madelain & Krauzlis, 2003). Here, we measure the effects of mask characteristics on the speed and gain of smooth pursuit after target disappearance. Participants (N=8) tracked a target moving along horizontal axis (13°/s). After 900ms, the target disappeared for 700 ms behind a mask consisting either in a uniform, visible or invisible, mask or in a texture of static or flickering (10 Hz) disks. The results show that smooth pursuit gain and speed drop after target disappearance for uniform and textured masks, but are maintained with a flickering mask, indicating that temporally modulated backgrounds facilitate voluntary smooth pursuit maintenance in the absence of a driving target. We suggest that with a flickering mask, the retinal slip elicits reverse-phi motion responses in MT neurons (Krekelberg & Albright, 2005) and interpret this novel effect as reflecting the on-going contribution of motion responses of MT neurons to pursuit. This questions models of pursuit relying on the retinal slip. Instead pursuit generation continuously integrates cortical motion responses, which with the flickering mask used herein provides a positive feedback to the pursuit system.

ABSTRACT. Some studies find evidence that multiple features can be simultaneously selected (e.g. Adamo et al., 2008), but Lo and Holcombe (2013) found that multiple-feature selection was achieved only with sufficient exposure duration. Here, participants monitored four simultaneous streams of letters, two to the left of fixation and two to the right, presented at a rate of 5.3 items/second. One stream on each side was a target stream of letters in a pre-cued colour, while the other was a stream of distractors presented in a different colour. At a variable time, circles appeared around the streams, and participants were asked to report the letters of the pre-cued colour for each side. We used mixture modelling to estimate efficacy, the proportion of trials where the participants reported a letter around the time of the circles. When observers monitored the same colour for both targets, efficacy was approximately 60%. When participants monitored different colours for each target, efficacy for the left was approximately 50%, whereas that for the right was as low as 26%. The results suggest that at least for these brief-exposure conditions, simultaneously selecting multiple colours is difficult and the colour on the left side is usually selected first.

ABSTRACT. To investigate the role of attention in the time course of post-iconic memory stores in visual short-term memory, a memory scanning experiment was conducted. On each trial, a memory array of 4 items was presented, followed by, after a varying SOA (500 to 4000 ms), the probe stimulus; observers were asked to report whether or not the probe differed from an item in the memory array. In cued conditions, either a neutral or informative spatial cue was presented 300 ms before probe onset. Similar to previous findings (Jacob, Breitmeyer & Treviño, 2013), comparison effects (CEs) in the control condition varied systematically across SOAs, likely to reflect fluctuations in attention to working memory (WM) content. In the location-cued condition the amplitude of these fluctuations was dampened, and the CEs diminished by 4000 ms. This is likely to reflect spatial attention overriding the effects of attention to WM content. In contrast, the CEs in the neutral-cued condition showed fluctuations with higher amplitudes for later SOAs. Thus the neutral cue enhanced the impact of working memory content, especially for the long SOAs. Taken together, our results suggest that attention plays a role in determining stages of information processing in working memory.

ABSTRACT. Covert endogenous attention, voluntarily allocating attention to a particular location without eye movements, increases accuracy and cortical activity in striate and extrastriate areas. We used attention to modulate cortical excitation and occipital Transcranial Magnetic Stimulation (TMS) to investigate how they jointly modulate early visual processing as a function of cortical excitability. First, observers performed an orientation discrimination task, in which performance is contingent upon contrast sensitivity, across a range of stimulus contrasts under valid and invalid attention cues. A valid cue induced response gain, a higher asymptotic performance for the attended than unattended location. Then, during the TMS sessions, observers performed the same task at their own asymptotic contrast level. On each trial, double-pulses of TMS were applied at various delays, before or after stimulus presentation, to retinotopic early visual areas (V1/V2) to modulate the processing of either the cued or the uncued stimulus. TMS modulated the magnitude of the attentional cueing effect across delays, especially when applied to the retinotopic location corresponding to the unattended stimulus. These findings support the idea that these effects are activity dependent, and show that occipital TMS modulates the effect of attention on contrast sensitivity.

ABSTRACT. Recent buzz in the social media has brought up an example of a photo of a dress in which different observers have surprisingly different judgments about the colour of the dress. The most obvious explanation from colour science should surely be that observers have different implicit assumptions about the illumination in the photo. The reason this particular photo produces so diverse a variety of judgments presumably is that this photo allows a variety of interpretations about the illumination of the dress. To test this hypothesis we measured the perceived colours of the dress, the subjective grey point of the dress, and the estimated colour of the illumination: we did this by using (1) a colour naming and (2) a colour adjustment procedure. The results show that the perceived colour of the dress is negatively correlated with the adjusted subjective grey point and the illumination along the daylight axis. This was also confirmed in the naming data. The phenomenon nicely illustrates the power of prior assumptions in perception. It is exactly what would be predicted from classical colour science, and no additional mechanism need be invoked to account for the surprising diversity in judgments of the dress’s colour.

ABSTRACT. Continuous Flash Suppression (CFS) (Tsuchiya & Koch, 2005) has become a popular tool for investigating visual processing outside of conscious awareness. During CFS, masks are flashed to one eye, suppressing awareness of stimuli presented to the other eye. To date, little is known about the precise relationship between masking effectiveness and temporal masking frequency. We investigated the suppression effectiveness of a wide range of masking frequencies (0-32Hz), using both complex (faces/houses) and simple (closed/open geometric shapes) stimuli in two conditions: different frequencies occurred in random order; trials with the same masking frequency were grouped into blocks.

Break-through contrast differed dramatically between masking frequencies (F(9,117)=17.539, p<0.001), with mask effectiveness following a log-normal curve peaking around 6Hz. This suggests that the 10 Hz or 20 Hz frequencies used in most CFS studies may not be optimally effective. In terms of type of stimulus, we found no significant difference in peak frequency between the stimulus groups (complex/simple, face/house, closed/open). Condition 1, frequency randomized within blocks, resulted in higher break-through contrast (F(1, 13)=6.886, p=0.021). More generally, these findings support the idea that temporal factors play a critical role in perceptual awareness, perhaps by interacting with the ongoing or stimulus-evoked oscillations involved in conscious perception.

ABSTRACT. The cone photoreceptor mosaic is the cellular entry point for photopic vision in humans. Its discrete packing arrangement and inner retinal connectivity also sets the first limit on spatial sampling. Due to constant eye motion, retinal images are being translated across multiple receptors at all times, yielding spatio-temporal information necessarily to be included in the sampling process. By arresting visual stimuli on a set of targeted photoreceptors, we here explore to what degree this information is employed during high acuity and hyperacuity tasks. We used an adaptive optics scanning laser ophthalmoscope to image the retina and deliver aberration corrected visual stimuli (acuity: 4-AFC tumbling E, hyperacuity: 2-AFC vertical 3 dot) at high levels of retinal contingency. During sustained fixation, subjects were randomly presented with stimuli shown in either retinally stabilized or normal viewing conditions at specific retinal locations for 750 msec, with stimulus feature size below the Nyquist sampling limit. In acuity experiments, all subjects performed significantly worse in the stabilized condition, with a mean performance drop of ~23%. Performance was unchanged during hyperacuity tasks. We conclude that fixational eye movements can facilitate discrimination in acuity tasks, but do not in tasks where integration processes are operating at maximum already.

ABSTRACT. A strong light or ophthalmic disorder causes the glare. However, we can perceive the glare except for the environment, for example, the glare illusion makes us perceive the glare. Thus, the neural mechanism of the glare perception has not been clarified. In this study, we used the SSVEPs to observe the neural activity in early visual cortex. We manipulated the luminance of the glare illusion and the dots flickering on its center region at 7.5 Hz. The participants were requested to pay attention to the dots and responded to the random distractor on center of the stimulus. The grandaveraged SSVEPs amplitudes (9 participants) of 7.5 Hz were compared among each gradient and luminance condition. The result of the SSVEP amplitude for the control condition (no glare illusion) showed that it increased in proportion to the contrast ratio. On the other hand, the results for glare showed that the amplitude to high contrast ratio was significantly low compared with the control condition. These results suggested that the perceived contrast was decreased by glare perception whereas the perceived luminance increased in proportion to contrast. Hence, bright enhancement perception with glare has been suggested to be processed in the primary visual cortex.

ABSTRACT. The intrinsically photosensitive (melanopsin-containing) retinal ganglion cells are proposed to provide signals to the visual as well as the non-visual pathway, with evidence suggesting melanopic input to both full-field brightness perception (Brown et al. 2012) and peripheral chromatic detection (Horiguchi et al. 2012). Here, in order to separate peripheral from central contributions to brightness perception, we examined brightness matching between stimuli restricted to different retinal locations. Observers performed a dichoptic matching task by adjusting a 1° test field, presented to the fovea of the left eye, to appear equal in brightness to a 3° standard field presented at ~15° eccentricity in the right eye. Each field’s spectrum was controlled independently using 13-primary, tuneable LED luminaires. The standard fields were visual metamers with different amounts of cone-silent spectral modulation targeting the melanopic spectral sensitivity function. Observers controlled the test field’s luminance only. Preliminary data from 5 observers demonstrate that test field matches require significantly higher luminance to match the higher melanopic-activation standard fields. These results suggest that the brightness of peripheral stimuli is influenced by cone-silent modulation most likely driven by a fourth photopigment with melanopic spectral sensitivity.

ABSTRACT. The goal of a color management system (CMS) is to provide consistent colors across different devices or displays. In standard CMSs, equivalent colors are attained using colorimetric values such as CIEXYZ or CIELAB; however, the resulting colors do not always correspond to human perception. In addition, standard CMSs cannot manage color appearance changes due to the observer’s adaptation to illumination. To resolve these problems, we have recently developed a new CMS based on color constancy. In our CMS, we derive a matrix to convert the RGB values of a display from one illumination condition to another condition through a color matching experiment. In the present study, we applied our CMS to a smartphone display. In the experiment, a color chip and a smartphone display were placed in separate rooms with different illumination conditions. Observers moved between the rooms and adjusted the color on the smartphone display until its color appearance was equal to the color appearance of the chip. The result showed that, although the RGB values significantly varied depending on the illumination conditions, the conversion matrices obtained from the experiment could accurately predict the color appearance under different illuminants.

ABSTRACT. Apparent spatial frequency of a grating is affected by a surrounding grating (Klein et al., 1974). Also, a fast flickering grating appears finer than it is (e.g. Virsu, Nyman & Lehtiö, 1974). We combined these two effects to locate their responsible processing sites and their relative locations. We used a center-surround configuration to see how various surrounding gratings affect apparent test spatial frequencies. Subjects matched the spatial frequencies of the vertical test grating with horizontal comparison grating presented beside it. The test grating was 2 cpd, and the surrounding vertical grating was either the same (2 cpd) or coarser (1.2 cpd). The surround grating was either flickering (25 Hz, counterphase) or static. All gratings were sinusoidal. The results showed that the test appeared finer when surrounded by the static coarse grating, but not when surrounded by the flickering coarse grating. Additionally, the test appeared coarser when surrounded by a flickering same-frequency grating compared with when surrounded by a static same-frequency grating. Subsequent experiment confirmed that flicker made the surround finer by 14 to 62 %. The results suggest that the site of simultaneous contrast of spatial frequency is located after the site of the flicker-induced apparent spatial frequency change.

ABSTRACT. Many previous studies reported qualitatively different characteristics of feature and spatial attentions. We investigated the orienting strategy of feature and spatial attentions for visual processing by using bimodal functional brain imaging methods with functional near-infrared spectroscopy and electroencephalogram. By manipulating the feature dimension toward which attention was oriented and the task difficulty in visual search paradigm we examined the mechanism of assigning attentional resource. Two visual search patterns with different colors were displayed in both visual hemifields, which were updated repeatedly at random timing asynchronously. Participants conducted visual search task only for one hemifield and neglected the other. Attended side was indicated by spatial or color cue. Task difficulty level was manipulated by set size, one or eight. OxyHb activation and alpha-band oscillation were recorded mainly on the posterior part of head surface. OxyHb increased as task difficulty in task execution period. In spatial cueing condition, this elevation was observed also in preparation period. Alpha oscillation during preparation period was also affected by the difficulty only in feature cue condition. These results seem suggest that even expecting higher attentional resource consumption would change strategy of orienting spatial attention and that spatial and feature attentions would have qualitatively different anticipatory systems.

ABSTRACT. Attention to color operates beyond the spatial focus of attention – a property referred to as global color-based selection (GCBS). In a recent EEG/MEG study, we observed a neural correlate of GCBS reflecting the selection of any color that is task-relevant, even when it is absent from the focus of attention (template matching effect, Bartsch et al., 2014), suggesting that this GCBS modulation reflects an explicit bias for target colors. Here we asked whether implicit biasing effects, like color-priming, could contribute to the template-matching-related GCBS modulation. To assess the effect of explicit biasing, we compared GCBS modulations of colors that matched the set of target-defining colors on a given experimental block with task-irrelevant colors that were target-defining on other experimental blocks. The effect of implicit biasing was assessed by comparing GCBS responses of those task-irrelevant colors to that of a color that was never target-defining throughout the entire experiment. We observed that explicit and implicit color biasing both gave rise to GCBS modulations, and that they were dissociable in ventral extrastriate visual cortex. This extends our previous demonstration that GCBS does not refer to a single selection process but instead involves multiple separable selection operations in ventral extrastriate visual cortex.

ABSTRACT. Glare may can be seen in two ways: The quantity of the light entering the eye may be to high or the light contrast of the light spot on the retina may be too high in relation to neighbouring areas. The anatomical and physiological properties of the eye, which are in relation with diminishing of the excess light to access useful vision in difficult light distribution in the eye are discussed. The localisation of retinal area which is the target of the light rays are important. Also the direction of the photoreceptor cells in the retina contributes to avoid glare. The use of the physiological glare avoiding mechanisms in everyday life which end up in a changed visual perception will be discussed.

ABSTRACT. Exogenous allocation of attentional resources allows the visual system to encode and maintain representations of stimuli in visual working memory. However, limits in the capacity to allocate resources can prevent unexpected visual stimuli from gaining access to consciousness. Using a novel approach to create unbiased stimuli of increasing saliency, we investigated early visual processing in humans that show a high (Inattentionally Blind, IB) or low propensity (Non-IB) to neglect unexpected stimuli during a visual search task. Electroencephalographic recordings resulted in diminished amplification during the latency of the N1 and poor enhancement of targets in the N2 latency (N2pc), both of which predicted a decrement in the likelihood to detect unexpected visual stimuli; a later modulation in the P3 latency was also found. Our data suggest that a lack of visual attention resources may be responsible for the failure to “ignite” a state of high-level activity necessary for unexpected stimuli to access awareness.

ABSTRACT. The purpose of this study is to measure color appearance during color adaptation as a function of time. We conducted an experiment using a 1/6 scale model room in which LED lights and white walls were installed. Five kinds of lighting condition, 2000K, 2800K, 4200K, 5800K and 10000K, that combined of six kinds LED chips were prepared. Four participants observed the center of the front wall and evaluated the color appearance with a color scaling method at ten seconds, twenty seconds, thirty seconds, one minute, two minutes, three minutes and four minutes after changing the lighting. They answered the ratio of an achromatic and chromatic color, and the ratio of the selected two colors from four psychological primary colors. Participants were females in their twenties with normal color vision. Participants evaluated each condition three times. As a result, the ratio of chromatic response was large in the case of large differences of CCT between before and after the lighting conditions. The mean ratio of chromatic response was 0.33 when changing the lighting from 2800K to 5800K but was 3.00 from 5800K to 2800K. We analyzed these results in terms of XYZ values and LMS cone activities.

ABSTRACT. Visual perception of attention to multiple moving objects, such as Multiple Object Tracking (MOT) is one of the essential abilities for humans. Some previous studies showed that MOT task performance is influenced by the number of the tracking object, moving speed, and hemifield limit. However, few studies have investigated the attentional state of multiple objects moving across hemifields. Therefore, we examined the attentional state of multiple object crossing hemifields in this study. In the experiment, two pairs of two dots were presented in two positions of each quadrant. In one condition, the two pairs moved up and down, or left and right from the same sides and in the other condition, the pairs moved from the diagonal positions. Participants were instructed to track two moving objects among four. The results showed that the accuracy of tracking objects crossing hemifields (horizontal movement condition) is lower than that of tracking objects moving vertically. The same trend was observed in the diagonal condition. This result showed that attention was suppressed by object crossing hemifields horizontally. Our finding suggests the importance of the object crossing hemifields in MOT.

ABSTRACT. Holding an item in working memory guides attention to the feature-matching item in visual environment. Recent studies have shown that this guidance of attention by memory item can be modulated strategically (e.g., Kiyonaga, Egner, & Soto, 2012). The present study investigated the underlying mechanisms by measuring electrophysiological index of attention to and working memory maintenance of memory item (i.e., colored oriented bars). Participants went through memory task and dual task (memory task and visual search task). In visual search task, we manipulated the probability of memory-match trials (the color of memory item and search target were the same; 0%, 50%, and 100%), which were pre-instructed to participants. Behavioral results showed that target detection became faster in 100% than in 50% condition. P3, which reflects attention allocation, was the largest in 100% condition. In contrast, the contralateral delay activity (CDA), an index of working memory load, was equal across tasks and probability conditions. These results suggest that allocating attention to the feature of memory item, not loading working memory, leads to the improvement of visual search performance.

ABSTRACT. In the study we modelled a situation of foreign language vocabulary learning: 31 Russian students memorized visually presented word pairs (native language word – pseudoword). Each pair was presented for 6 seconds, recall check was after each trial of 4 pairs. Monocular eye movements and quantitative recall data were recorded. We also analyzed post hoc reports about mnemonic techniques used and kinds of recall mistakes, which allowed us to distinguish 3 main strategies of paired associate vocabulary learning: “graphical” (attention to graphic details), “phonological” (rehearsing the words) and “semantic” (creating associations). These strategies correspond to the levels of processing as proposed by Craik & Lockhart (1972) and Velichkovsky (2006). Recall score was connected with the strategies used F(3;755)=11,2 (p<0,01), escalating with the increasing “depth” of processing. Eye movement characteristics also reflected types of strategies: "graphical” strategy was associated with more fixations; “phonological” strategy was characterized by more switches between Areas Of Interest (native and pseudo- word) and less fixations on the AOIs; “semantic" strategy was distinguished by fewer switches between AOIs and longer fixation duration. The findings indicate that subjects use different strategies of foreign words memorization, which are reflected in their eye movement characteristics and result in recall efficiency.

ABSTRACT. Visual texture is plausibly one of the more important features determining where we look, often being highly diagnostic of the presence of many important objects. Unfortunately even characterising a texture is non trivial: some modern computational models require over one thousand parameters. In contrast multi-dimensional scaling methods based on "perceptual similarity" are well accounted for by only two or three. To find the dimensionality and natural representation of texture underlying eye movements, we measured symmetrised confusion matrices in an eye movement task where performance was based on making eye movements to a pre-cued texture. This measure is a legitimate random partition kernel, but only sensitive over short distances. We therefore analysed the data using the non-linear manifold learning technique tSNE. We present estimates of the dimensionality of eye movement "texture space", and illustrate the nature of the two dimensional solution that captures much of the variance.

ABSTRACT. Visual search is facilitated if distractors are separated in time, allowing old information to be inhibited and new information to be prioritized (the preview benefit; Watson & Humphreys, 1997). In the present study, we investigated whether stimuli that required spatial grouping can be inhibited effectively. Experiments 1 and 2 showed that with Kanizsa stimuli, or with non-Kanizsa stimuli that required local grouping, a preview benefit was only obtained at small set-sizes. In Experiment 3, we found that local changes to the elements of grouped stimuli totally abolished the preview benefit leading to no search advantage for new stimuli. Overall, the results suggest that: i) the need to group stimuli reduces the capacity to prioritize multiple new items, ii) grouping local elements interferes with the maintenance of inhibitory visual marking irrespective of whether illusory contours can contribute to stimulus formation, iii) changes to locally grouped elements feed back to the inhibitory attentional set to abolish any inhibition at those locations. The findings are discussed in terms of the limits of time-based attentional selection, interference between grouping and inhibitory mechanisms, and the effectiveness of illusory contours in guiding search.

ABSTRACT. Feature contrasts, such as differences in luminance, orientation and movement, are salient and capture visual attention. What occurs when an item is salient by virtue of more than one feature, for example if it is both bright and moving? Some evidence suggests that additional salient features are redundant (Nothdurft, 2000), whereas other studies have found that salience in one feature enhances the apparent salience of other features of the same object (Kerzel et al., 2011). We investigated whether these conflicting results might be due to a change in the way salience is processed at different stages in the visual system. 16 participants completed 1080 trials in a visual search task, with distractors of varying luminance, orientation and movement. We measured the latencies of saccades to target and the deviation of these saccades towards distractors. We found interactive effects of these three features on saccadic deviation. Specifically, at short reaction times, salient features combined overadditively, whereas at longer reaction times, they combined underadditively. These two findings suggest that early processing in the visual system is characterised by a sensitive search, in which salience in one feature may trigger increased sensitivity to other features. Later processing then discards redundant sources of salience.

ABSTRACT. Value-associated distractors slow visual search performance by engaging attention. Here, we ask if value-associated items can boost preparatory cognitive control and aid attentional suppression. In a two-stage experiment, 24 participants first learned to associate a unique combination of colour (purple or orange) and point-value (win or zero) with each of four symbols. They then completed a cued visual search task without point rewards. Each trial consisted of two successive cues (symbols from the learning task) followed by a two-item search task (one orange and one purple letter). The task was to accurately locate one of two pre-specified target letters as quickly as possible. The colour associated with the first cue (present on 80% of trials) always predicted the target colour; the second cue was irrelevant and its associated colour was equally likely to be congruent or incongruent with the target colour. Value-associations were irrelevant. When the first cue was win versus zero-associated, response times were faster, indicating value-based benefits to preparatory control. When the first cue was absent, second cues associated with zero value produced large colour congruency effects, an effect not seen with win-associated items (p < .05). This latter finding suggests the presence of a value-based suppression mechanism.

ABSTRACT. When participants react to a target stimulus, reaction times (RTs) are shorter when the moment of stimulus occurrence can be predicted. In three experiments, we investigated whether temporal attention based on such predictions boosts the selection of targets among distractors. We employed a flanker task, in which a central target letter required a speeded left- or right-hand reaction. The target was flanked by lateral distractor letters, requiring either a response with the same hand as the target (compatible), no response (neutral), or a response with the other hand (incompatible). Temporal attention was manipulated across blocks of trials by means of a warning signal preceding target onset by a short or long time interval, yielding conditions of high and low temporal attention, respectively. Moreover, target and flankers were presented simultaneously or in short succession (flanker to target interval = 32ms). As expected, RTs were longer in incompatible trials and shorter in compatible trials as compared to neutral ones, which replicate the usual flanker compatibility effect. Importantly, this compatibility effect was reduced by temporal attention in the simultaneous condition, whereas it was increased in the successive condition. These results demonstrate that temporal attention modulates target processing by boosting targets or strengthening distractors.

ABSTRACT. An intriguing paradox is created by the phenomena of lightness contrast, in which the difference in lightness between targets and inducers is over-estimated, and assimilation, in which this difference is under-estimated. Inducers’ spatial frequency is an important factor: inducers with low-spatial frequency seem to induce contrast whilst inducers with high-spatial frequency seem to induce assimilation. The interaction of this factor with depth is less clear. Wolff (1934) found that contrast disappears when low-spatial frequency inducers are non-coplanar with targets, while Soranzo at al. (2010) found that contrast persists even with high-spatial frequency inducers when they are non-coplanar with the targets. The difference between these two studies is that Wolff manipulated actual depth, whilst Soranzo et al. manipulated stereoscopic depth. This study manipulated the actual distance between target and inducers, as well as Inducers’ Spatial frequency and Intensity. Participants indicated the target lightness by choosing the correspondent chip in a Munsell scale. Results suggest that high spatial-frequency inducers generate assimilation effects only when they are coplanar with the targets, but contrast effects when they are non-coplanar. Low-spatial frequency inducers generate contrast effects both when they are coplanar and at distance. We conclude that assimilation is more affected by distance than contrast.

ABSTRACT. Contrast sensitivity can be substantially reduced by the presence of nearby stimuli, a phenomenon referred to as surround suppression. Previous studies have measured contrast sensitivity as a function of surround spatial frequency and revealed surround suppression is band-pass tuned around the target frequency (Petrov, Carandini & McKee, 2005; Serrano-Pedraza, Grady & Read, 2012). This pattern of tuning implies that surround suppression reflects inhibition from cells with similar tuning preferences as the target. To date the spatial frequency tuning of surround suppression has been investigated using a limited range of stimulus parameters. We sought to determine to what extent the band-pass tuning observed in previous research was dependent on the phase of the surround, surround contrast, target size and the spatial configuration of the surround. Manipulation of these parameters changed the selectivity of surround suppression and revealed under different conditions that surround suppression can be low-pass tuned, broadly tuned or show no tuning at all to surround spatial frequency. Our results suggest that the characteristic band-pass tuning previously reported is not a universal property of surround suppression, but may represent the tuning associated with just one of multiple components contributing to the inhibitory influence of surrounding stimuli.

ABSTRACT. A unique characteristic of light emitting diodes (LEDs) is that they respond very fast to changes in the driving current, which can be used to e.g. easily control the intensity of the light output. However, improper selection of dimming technology parameters or addition of mains voltage fluctuations can result in flickering of the LEDs’ light output. A number of objective measures have been developed in the past to quantify flicker perception (i.e. Flicker Index, Flicker Percent, FVM, Pst), but none of them can accurately predict both periodic and aperiodic flicker for general lighting conditions. A new measure is proposed, called FVMt. It is based on the human visual system model developed by Frederiksen and Hees (1997) and it consists of two impulse response functions. The parameters of the model were determined so that they fit the flicker sensitivity curve for general lighting conditions, developed in earlier experiments (Perz et al., 2013). A perception experiment was conducted, in which flicker visibility to realistic waveforms was evaluated. The results showed that FVMt predicts flicker visibility significantly better than the conventional measures, FI and MD and correlates well with Pst for aperiodic, and FVM for periodic flicker while predicting the visibility threshold better.

ABSTRACT. Inhibition of Saccadic Return (ISR) is an attentional mechanism which discourages the reorientation to previously inspected locations and thus guides search towards uninspected locations. In the current experiment we investigated how ISR in visual search is affected by the storage of visuo-spatial information in short-term memory (STM). In a dual-task paradigm, participants performed a visual search as primary task while maintaining none, two or four object locations in STM as secondary task. During the search, a visual probe was presented at a location that had either been visited previously (old probe) or not (new probe). The participants were instructed to saccade to this probe immediately and then to resume search. The results showed that saccade latencies to old probes were longer than to new probes. This was independent of the number of stored locations in STM. This suggests that ISR is not affected by this kind of visuo-spatial STM load.

ABSTRACT. Attentional state has been linked to both the Locus-Coeruleus-Norepinephrine (LC-NE) system and to prestimulus occipital alpha power. Recent work has also highlighted the relationship between LC-NE activity - measured using pupil diameter - and prestimulus occipital alpha power. Here, we investigated the relationship that attentional state has with pupil diameter and prestimulus occipital alpha power by comparing the temporal dynamics of these relationships. Participants completed an RSVP detection task in which they signalled the presence or absence of a target embedded within a stream of white-noise patterns; in addition, they provided subjective reports of attentional state. Simultaneously, EEG and pupillometric measures were collected. In a second experiment, we used the same paradigm and manipulated the timecourse of participant engagement by altering block length. The temporal dynamics of the relationship between subjective attentional state and 1) pupil diameter and 2) prestimulus alpha power show similarities: both show strong correlations with attentional state at the same frequency of trials. Alongside the similarities between these relationships, our data also show that prestimulus occipital alpha is related to attentional state at a wider range of frequencies of trials than pupillometric measures.

ABSTRACT. In recent studies, multiple-object tracking (MOT) tasks combined with probe detection tasks have been used to investigate the distribution of attention around moving objects (e.g., Atsma, Koning & van Lier, 2012). During these tasks, participants were allowed to move their eyes freely around the scene, allowing both overt and covert attention to be deployed. Here we focus on the contributions of these two forms of attention in MOT. An MOT task was used with one overtly and one covertly tracked target and an equal number of distractors. In addition, probes were flashed around all objects and in open space. Preliminary results show an anisotropic distribution of probe detection rates around covertly tracked targets (as in Atsma et al., 2012), but less so around overtly tracked targets. This suggests that attention takes into account movement direction to keep track of object locations, especially when an attended object is not foveated. Remarkably, probe detection rates for the distractors also show anisotropic distribution. We discuss the possibility that these objects were also automatically (covertly) attended because they occasionally interact with the targets by bouncing.

ABSTRACT. During real-world vision observers need to attend to and track several moving objects. Sometimes the identities of the objects are task-relevant (when doing multiple identity tracking with distinct objects; MIT), and sometimes task-irrelevant (when doing multiple object tracking with identical objects; MOT). Here we reveal cortical circuits supporting MIT and MOT while tracking moving objects. Twenty-four participants tracked identities and locations (MIT) or locations only (MOT) of 0, 2, or 4 moving targets while their eye movements were recorded and haemodynamic brain activation measured with functional magnetic resonance imaging (fMRI). Participants made more saccades during MIT versus MOT condition. Both the MIT and MOT task engaged similar components of the dorsal attention system, yet MIT resulted in increased activity in frontocortical areas involved in working memory and temporal areas involved in object recognition. Additional activations were observed in intraparietal regions involved in attention shifting. A control experiment with 14 subjects and conditions allowing or not allowing eye movements confirmed that brain activation differences between MIT and MOT were not confounded by gaze shifts. We conclude that the MIT task relies on interactions between the dorsal attention circuit, executive functions, and ventral visual areas.

ABSTRACT. Gymnasts use visual information during aerial movements, but how they use vision to achieve safe and correct landing in a jump with a turn is still unclear. The purpose of this study was to reveal the gaze shifting patterns in gymnasts during a jump with full turn by measuring horizontal eye-head movements. 10 participants were male skilled gymnasts (age [M ± SD], 23.90±3.28years; career length, 15.70±3.06years). They had to perform a jump with full turn (360°). To measure horizontal eye movement during the jump, electrooculography (EOG) was recorded using a wireless system. Simultaneously, the jump was recorded using two high-speed digital cameras. Gaze direction was determined by adding horizontal angle of eye and head movement. It was found that all participants performed perfect landing. The gaze shifting were classified roughly into two patterns. In the one pattern, gaze stabilization depended before take-off and during landing. This pattern include correcting gaze direction during landing. In the other pattern, additional gaze stabilization developed during flight phase. They might use visual information to optimal landing and precise rotation by stabilizing gaze during a jump with full turn. Further studies are needed to examine which pattern have an advantage in their best performance.

ABSTRACT. An attentional blink occurs when a target presented in an RSVP stream (T2) fails to be detected (only) when it occurs soon after a previous target stimulus (T1). Olivers and Nieuwenhuis (2006) proposed that the blink results from “attentional overinvestment” in the first target (T1) driven by a focused attentional state. Building on this proposal, we show here for the first time that across individuals blink-size is (1) positively related to switch-costs on a divided-attention task; (2) negatively related to interference - deriving from response-competition - on a selective-attention task; (3) positively related to more rightward biases in line bisection. More rightward biases have been linked to decreased alertness (Manly et al., 2005). We conclude that (a) blink-size is a marker of an individual's ability to spread or focus attentional resources and to excel, respectively, at tasks of divided or selective attention; (b) the latter ability may be explained by differences in alertness, with increased alertness favouring divided over selective attention. Taken together, our results are consistent with the idea that there is a trade-off between exploratory and exploitative behaviour (divided and selective attention) and that this trade-off is modulated by alertness (Aston-Jones et al., 1999; Linnell et al., 2014).

ABSTRACT. A number of studies have examined to what extent social or emotional stimuli generate inhibition of return (IOR) comparable to non-social objects, but results have been inconsistent. One possible explanation for this inconsistency is that previous studies have conflated location IOR and object IOR. The present experiments addressed this by comparing IOR to social and non-social stimuli in dynamic displays. As objects moved between the presentation of the cue and the probe, location and object IOR could be differentiated. Two conditions, one with non-social objects (ovals) and one with social objects (schematic faces), were tested in a within-participants design. When the non-social object condition was completed first, both object and location IOR were observed for ovals whereas object facilitation was observed in the face condition. Additionally, only object IOR significantly differed between the two conditions and not location IOR. When the social object condition was done first, there was no object-IOR effect for either type of object. These results suggest that IOR observed in stimulus arrays containing social objects such as faces is location-based and not object-based. These data are consistent with the view of Perez-Duenas, Acosta and Lupianez (2014) that biologically relevant stimuli override IOR.

ABSTRACT. We investigated the degree of transparent layer constancy when matching filter colors across different illuminations. Previous studies reported varying degrees of constancy and it was suggested that constancy is not influenced by whether or not the subjects have a transparency impression. We systematically examined the role of viewing context and phenomenal impression on constancy performance. We compared normal filter presentation that leads to phenomenal transparency with a flipped filter version that violates geometric transparency cues and appears opaque (X- vs T-junctions). Two different background structures were used and we also varied the presentation mode (static vs motion, monocular vs stereo, simultaneous vs alternating display of test and standard). We found that the geometric cues had the largest effect on transparent layer constancy: It was significantly higher with X-junctions than with T-junctions. Alternating presentation of standard and test enhanced constancy throughout. The effects of motion, stereo and background structure were smaller and more complex: While in the case of X-junctions constancy was enhanced when the objects were moving or presented in a more structured background pattern it was reduced with T-junctions. Together, this suggests that the more vivid the impression of a transparent layer is the higher the degree of constancy.

ABSTRACT. It is commonly argued that colour categories affect colour perception. However, some suggest ‘category’ effects are due to inequalities in uniform colour spaces or stimulus-specific issues (e.g., Witzel and Gegenfurtner, 2013). Using a visual oddball task, where attention is directed away from colours, we investigate whether colour categories affect colour processing in English speakers, who vary in whether they reliably name a stimulus blue or green. Event-related potentials elicited to this frequently presented ‘standard’ stimulus were compared to those from two ‘deviant’ stimuli that were presented infrequently and separated from the standard by 2.5 just noticeable differences. Participants then named each colour 25 times and deviants were classified as same- or different- category to the standard for each participant based on their naming. A category effect was found over occipital sites around 100ms after stimulus onset: the different-category deviant elicited a significantly lower mean amplitude than the deviant named the same as the standard. Further effects of naming were found over frontal sites around 200ms. These effects cannot be accounted for by stimulus issues since same- and different-category stimuli varied across participants. The findings support prior claims of colour categories being reflected in early sensory stages of colour processing.

ABSTRACT. Color constancy refers to a phenomenon that the color of an object appears to be constant to some extent when the object is viewed through a chromatic transparent filter or when it is illuminated by a colored light. There are two different types of color changes that induce color constancy: one being the multiplicative color change and the other being the additive one (Kitaoka, 2011). It was pointed out that veins appear to be bluish but are actually of yellow or orange hue (Kitaoka, 2014), though this fact had been reported in the literature on the study of skin color. Here I propose a hypothesis that the vein illusion is not only a color-contrast illusion but reflects an erroneous expression of color constancy based upon the inverse transformation of the additive color change.

ABSTRACT. Priming of visual search has been shown to strongly affect vision and action, releasing items from crowding and during free-choice primed targets are chosen over unprimed ones. Two accounts of priming have been proposed: attentional facilitation of primed features and postperceptual episodic memory retrieval, involving mapping responses to visual events. Here, well-known masking effects were used to decide between the two. Object substitution masking (OSM) reflects attentional processing: it doesn’t occur when a target is precued and is mitigated by distractors. Conversely, metacontrast masking (MCM), has been connected to lower level processing where attention exerts little effect. If priming facilitates attention shifts we expect priming to mitigate OSM, while MCM might not be similarly influenced. If priming allows objects to escape OSM, this would be particularly compelling, given the relation of OSM to attentional processing. Observers searched for an odd-color-out target among distractors. Unpredictably (on 20% of trials) OSMs or MCMs appeared around the target. We assessed whether search repetition affected masking. OSM effects were strongly mitigated for primed target colors, while MCM was mostly unaffected. This casts doubt upon episodic retrieval accounts of priming, placing the priming locus firmly within the realm of attentional processing of visual input.

ABSTRACT. We measured reaction times and Hit rates in a 2AFC orientation discrimination task of Gabor patches at different spatial frequencies and contrasts surrounded by red, blue, or grey isoluminant surrounds. Preliminary results revealed generally slower reaction times with red in comparison to blue or grey and a stronger effect on low spatial frequencies in comparison to middle or high. A direct within-participant comparison of low vs high spatial frequency and low vs high contrast demonstrated that a red background raises RT and lowers HR in comparison to a gray background only at low spatial frequency and constrast. These findings are relevant to the interpretation of studies that used red backgrounds to selectively isolate the magnocellular pathway, and may have implications for the interpretation of recent results of the effect of red contexts on diverse perceptual and cognitive tasks.

ABSTRACT. The rapid extraction of affective cues from visual environments is fundamental for flexible behavior. However, it is still unclear how the evaluation of emotional information is modulated by basic perceptual features. Recent behavioral studies revealed the existence of a brightness bias, according to which luminance variations of neutral pictures influence their affective evaluations. The present event-related potential (ERP) study investigated the neuronal chronometry of brightness effects in neutral and emotional scenes. Seventeen male participants rated original and bright versions of grayscale neutral and erotic IAPS pictures. Bright scenes were rated more positively compared to their original counterpart when neutral but less positively when erotic. Classical and mass univariate ERP analysis showed, starting at ~200 ms post-stimulus onset, lower N1 amplitude for bright vs. original erotic pictures but no brightness differences for neutral scenes. Complementary spatiotemporal analysis concurrently revealed the selective contribution of a centroparietal positivity following the presentation of original, but not bright, erotic scenes. This topography was also present at later time windows for erotic scenes only, indicating its involvement in emotional processing. These results indicate that neural networks subtending the extraction of emotional information are differentially recruited depending on low-level perceptual features, which ultimately modulate emotion evaluations.

ABSTRACT. The perceived final location of a moving target that is suddenly halted is usually displaced both in the direction of motion (RM) and in the direction of gravity (RG). Moreover, these errors evolve with time in a pattern that mimics an anticipated falling trajectory. The degree to which these errors reflect cognitive analogues of physical invariants or features of eye movements has been a major topic of contend. Separately assessing the role of eye movements for RM and RG is further complicated by the fact that for some target motions both are confounded. The present study addresses this issue by taking advantage of the logic of Fourier decomposition. Targets, moving along 16 possible directions, were shown to participants required to locate their vanishing positions after a variable retention interval from 0 to 1200 ms. Participants completed two tasks differing only on the eye movements instructions (smooth pursuit or fixation). The outcomes revealed that constraining eye movements drastically reduced the spectral signature of RM but not of RG. Moreover, RG increased with time, irrespective of the presence or absence of smooth pursuit eye movements. The results lend credence to the role of an internal model of gravity in spatial localization tasks.

ABSTRACT. Using subjective measures of conscious awareness raises multiple methodological issues, including what conscious contents participants should be asked to report, what the most apt recording technique is, and what statistical procedures should be applied. Concerning content, reports about the confidence in the accuracy of the preceding task response are associated with lower psychophysical thresholds, greater type-2 sensitivities, and earlier EEG correlates than reports about the visual experience of the stimulus. Concerning recording technique, more precise reports of visual experience can be obtained by visual analogue scales than by discrete labelled scales. Concerning statistics, logistic regression can be a powerful test to detect if there is a correlation between reports and task accuracy at all. However, logistic regression is less convenient to estimate type-2 sensitivity than measures derived from signal detection theory due to bias from subjective criteria. Overall, we suggest the study of consciousness increasingly benefits from subjective reports if methodological choices are tailored to the research questions.

ABSTRACT. A gap between fixation target offset and saccade target onset encourages the production of express saccades (ES; latency 80-130ms). Express saccade makers (ESMs) persist in producing unexpectedly high proportions (>30%) of ES in overlap trials (fixation target present when the saccade target appears), a pattern of behaviour suggested to be rare in the absence of pathology and/or extensive training. However, we have shown this behaviour is expressed in >20% of Chinese participants compared to 10% or less of Caucasian participants (Knox & Wolohan, 2014). What about other groups? We exposed 70 naive Egyptian/Arab participants to gap (duration 200ms) and overlap trials (2 x 200 trials for each condition, in separate blocks, targets 10° left/right randomised, randomised fixation time 1s-2s, order of blocks randomised between runs/participants). The percentage of saccades with latencies of 80ms to 130ms (%ES) was calculated for all saccades with latencies between 50ms and 500ms. 10/70 (14%) participants were ESMs, a proportion intermediate between Caucasian and Chinese participant groups tested previously. No Egyptian participant reached the very highest %ES in overlap conditions (ie >60%) occasionally observed with Chinese participants. These results confirm that ESMs may be more frequently encountered than expected based on the (mainly Western) literature.

ABSTRACT. Salient peripheral cues produce transient attentional facilitation which is superseded by a sustained inhibitory cueing effect (ICE). In static scenes cueing one part of an object produces an ICE that spreads throughout the entire object. In dynamic scenes the ICE is attached to objects as they move through space. We examined object centred attentional facilitation and inhibition in a patient with visual form agnosia. Consistent with previous studies, the patient showed a deficit of object-centred attentional facilitation. In contrast, object centred inhibitory cueing effects were observed in 3 out of 4 tasks. These effects were weakest in tasks where cues to objecthood were weak and strongest where cues to objecthood were highly salient. These data demonstrate a neuropsychological dissociation between the facilitatory and inhibitory effects of object cueing and are consistent with the view that object-based inhibitory cueing effects are mediated by the dorsal rather than ventral visual system. From a theoretical perspective we propose that attentional facilitation and inhibition of objects operate at different levels of visual processing, such that inhibitory cueing effects act on visual representations at the level of scene segmentation, whereas object attentional facilitation operates on categorical object representations

ABSTRACT. There is strong neurophysiological evidence showing a contralateral mapping of saccade-related activity in the primate superior colliculus (SC). However, functional imaging studies of the human SC have shown an increase in saccade-related activity (BOLD response) in both the contralateral and ipsilateral SC. The present study examined the BOLD response in the human SC using 3T fMRI during the saccade preparation and saccade execution phases. A univariate analysis (GLM) revealed a significant bilateral increase in BOLD response associated with saccade preparation, but was unable to identify a contralateral bias. A second analysis compared activity in each voxel with a contralaterality index (CI) and provided stronger evidence of a contralateral mapping of saccade preparation. Both the GLM and the CI analysis were sensitive enough to identify the presence of a contralateral bias for the saccade execution phase. The results show that the expected contralateral bias can be measured in the human SC using functional imaging. The bilateral increase in BOLD may reflect the sensitivity to neural activity and local field potentials associated with inhibition.

ABSTRACT. Activities such as team sports or driving a car require us to track multiple moving objects. Multifocal attentional enhancement of tracked objects in visual cortex is thought to subserve this ability. Interestingly, objects’ positions can be tracked without tracking their identities. This suggests at least partially separate mechanisms for tracking positions and identities. Direct gaze stimuli show a pronounced advantage in identity but not in position tracking. We used this to distinguish both in order to investigate the role of attention in identity tracking. Participants tracked three out of six circular objects having eyespots with different gaze directions and indicated their locations at the end of each trial. Recordings of frequency-tagged steady-state visual evoked potentials (SSVEPs) elicited by the flickering stimuli allowed us to concurrently measure the allocation of attention to each of the six moving objects. We found clear attentional enhancement of SSVEPs elicited by tracked targets and a behavioural advantage for tracking the identity of direct gaze targets. However, this direct gaze advantage was not reflected in the magnitude of attentional enhancement of SSVEPs. Multifocal attention was distributed equally to tracked targets, indicating that the direct gaze bias in identity tracking arises only at later processing stages.

ABSTRACT. Colour is a construction of our brains which plays an important role in our perception of the world (e.g. we tend to describe objects by their colour, among other attributes). Within the infinite possible combinations of wavelength spectra, an average subject can name about thirty colour categories, eleven of which are believed to be universal across languages (Berlin & Kay, 1969). A recent study (Horwitz & Hass, 2012) suggests many V1 and V2 neurons are isoresponsive to stimuli belonging to the surface of an ellipsoid in LGN-opponent colour space. Coincidently, our psychophysical measurements of colour frontiers resemble such shape. Hence, we have attempted to model human colour vision through a set of 3D ellipsoids in a biologically-plausible colour-opponent space. We dynamically adapt those ellipsoids – i.e. modifying their axes and centres – by considering the image statistics, mimicking what is known about signal normalisation and hue shift. Unlike the majority of the state-of-the-art colour naming algorithms which are pixel-based, our biologically-inspired model considers the surroundings stimulus, addressing many of the challenges that other algorithms cannot overcome. Overall, our new model shows promising results for studying the perception of colour, and in the future we will consider adding colour induction phenomena.

ABSTRACT. In daily life we readily identify liquids with different viscosities, such as water, honey, or tar. Many such liquids also have distinctive optical properties (e.g., colour, translucency). Optical properties are independent of viscosity but it seems plausible that we may associate specific appearances (e.g. water, caramel) with certain ranges of viscosity. Contrarily, the visual system may discount optical properties, enabling ‘viscosity constancy’ based primarily on the liquid’s mechanical properties. We investigated whether optical characteristics affect viscosity perception. We simulated liquids with viscosities varying over a wide range and rendered them with nine different optical characteristics. Observers performed two tasks: (1) adjusting the viscosity of a match stimulus until it appeared to be the same as a test stimulus at a different time point and with different optical appearance (2) rating six physical properties of the test stimulus. We tested both moving and static stimuli. Observers were good at matching viscosity and optical properties had almost no effect. Some ratings of liquid properties did show substantial interactions between viscosity and optical properties. These results suggest that although optical properties influence the perception of some characteristics of liquids, they have very limited influence on shape- and motion-based viscosity judgements.

ABSTRACT. Detection of target spatial displacement during a saccade is poor, an effect known as Saccadic Suppression of Target Displacement (SSTD). Sensitivity to displacement can be restored, however, if the target is absent for the brief period after a saccade or if there is change of the target's properties during the saccade. According to (Deubel, Schneider, and Bridgeman, 1996), the visual system is biased to assume that the world remains stable during a saccade and the introduction of a blank or property change breaks this assumption resulting in a more accurate comparison of pre and post saccadic location. The goal of this study is to test if task irrelevant, continuous target motion would also break the stability assumption and thus increase the ratio of the correct answers. We modify a classic SSTD experiment design for moving targets. In addition to the distance of the displacement and the presence of temporal gap, in half of the trials the targets are moving orthogonal to the displacement direction smoothly prior to and after the saccade. We observe the classic SSTD results for both the static and moving conditions, but we find no evidence for the influence of target's motion on the target displacement sensitivity.

ABSTRACT. Attentional Blink (AB) is the reduction in accuracy of detecting the second of two targets which appear in close temporal succession in a Rapid Serial Visual Presentation task. Resource limitation accounts explain AB by suggesting that the perceptual features of Target-1 (T1) interferes with the processing and eventually reporting Target-2 (T2) correctly. Hommel, Müsseler, Aschersleben and Prinz (2001) suggested that action and perceptual events are represented as common distal features and activating one will activate the other process. Similarly, Mussler and Wuhr (2002) suggested that action planning interferes with visual encoding. We propose that the response codes associated with T1 perception is critical for AB. That is, perception of T1 automatically triggers a response selection for it leading to an interference in T2 perception and response selection. Participants completed an AB task in two conditions – experimental condition where the response is contingent on the relationship between both T1 and T2 making it impossible to plan a response when only T1 is seen. The control condition was a standard AB task. Results suggested there is significantly higher T2 accuracy in the contingent response condition as compared to the control condition, providing support for the role of response planning in AB.

ABSTRACT. The aim of this work was to test how perceptual models of color deficiencies predict the performance of dichromats and anomalous trichromats on a hue scaling paradigm. Stimuli presented on a monitor consisted of a square target of variable color displayed at the center of a D65 background. Both target and background had the same luminance of 12 cd/m 2. In order to scale the component sensations of the unique hues observers an arbitrary scoring rule assign a value of 3 when the stimulus is described with a single hue name and if two names were used a value of 2 to the predominant hue and 1 to the other hue. In the normal condition the colors of the target were selected in 24 directions around D65 illuminant with a hue-angle step of 15-deg. The simulation condition corresponds to Brettel’s and DeMarco’s transformation of the normal condition. The experiment was carried out by three normal color observers, four dichromats, and four anomalous trichromats. The results suggest that DeMarco transformation predicts anomalous trichromats perception better than Brettel’s model predicts dichromats perception.

ABSTRACT. The aim of this work was to compare the preferred chromatic composition for unfamiliar paintings with the original. Stimuli were images of paintings generated by rotating the color volume around the L* axis in CIELAB. The images were presented on a CRT. Four paintings were abstract from Amadeo de Souza-Cardoso, two had figurative elements from the same painter, two were abstract and two were figurative from other painters. In each trial a painting selected at random from the set was presented with its color volume rotated by angle selected at random. The task of the observer was to adjust the angle of the chromatic rotation for best subjective impression. Three groups of observers carried out the experiment: 50 without previous knowledge of the paintings, 8 experts in art but unfamiliar with the paintings tested and 6 experts in the paintings of Amadeo de Souza-Cardoso. It was found that the distribution of angles selected for data pooled across paintings for the non-specialist observers could be described by a Gaussian function centered at -2 deg, i.e. close to the original colors of the paintings, and with a FWHM of 64 deg, just about 1.5 above the threshold to detect a chromatic change.

ABSTRACT. Body image disturbance is characteristic of individuals with eating disorders. While the role of media saturation with images of the “thin ideal” has been well documented, relatively little research has examined visual adaptation as a potential mechanism. We sought to examine the nature of body size representations for ‘self’ and ‘others’ using a visual adaptation paradigm. In Experiment 1, participants were adapted to either expanded or contracted images of either ‘self’ or ‘other’. The resultant aftereffect was assessed using both ‘self’ and ‘other’. While aftereffects were largest when the same body type was used for adaptation and testing, substantial aftereffects were also demonstrated for cross adaptation conditions, suggesting overlap in the representation of the two. In Experiment 2, participants were adapted to images of ‘self’ and ‘other’ that had been distorted in opposite directions (e.g. expanded self/contracted other). The direction of each aftereffect was contingent on the test stimulus, demonstrating differences in the representation for these body types. While body representations for self and other appear to be overlapping, yet distinct, the evidence of misperception of one’s own body following exposure to “thin” others demonstrates the viability of visual adaptation as a model of body image disturbance.

ABSTRACT. A lack of vision at an early age results in some impairment in spatial representation. However, an auditory feedback about arm movements may help to recalibrate the spatial sense. Here we studied the effect of an audio spatial training on the ability to reproduce a hand pointing task toward a moving sound and detect the final location. Forty subjects (20 early blind, 20 healthy blindfolded controls) performed the same task in two sessions. Between the two sessions 2 minutes inter-session was performed, that could either be 1) audio-motor training, i.e. moving the sound around the body mapping the peri-personal space or 2) passive training, i.e a complete rest. Subjects were randomized on the two conditions. The spatial accuracy was calculated for each participant and for each spatial position. In the first session, contrarily to blindfolded controls, early blind individuals showed a worse spatial accuracy, presenting impairment on the lower positions of the plane. After the audio-motor training, the spatial accuracy was highly improved (P>0.05). No statistical difference was present in case of passive training or in blindfolded controls. This result suggests that short audio motor training could improve audio spatial precision in early blind individuals.

ABSTRACT. Predictive coding theories explain core pathologies of schizophrenia as a result of false inference. Typically, higher-level symptoms such as delusions or hallucinations are considered to be the consequence of a misallocation of precision towards sensory evidence at the expense of prior beliefs (Adams et al., 2014). However, the predictive coding framework extends to the explanation of low-level deficits consistently reported in patients with schizophrenia (Fletcher and Frith, 2009), such as the reduced contextual modulation of BOLD–responses in primary visual cortex (Seymour et al., 2013). Here, we sought to probe the influence of contextual surround stimuli on the effective connectivity between retinotopically defined center- and surround-regions in primary visual cortex with Dynamic Causal Modeling in 18 schizophrenic patients and 16 control subjects. We assessed a modelspace of 20 DCMs using Bayesian Model Selection and compared parameter estimates with Bayesian Model Averaging. We found a reduced negative modulation of effective connectivity towards center-regions in primary visual cortex in the presence of a contextual surround in patients with schizophrenia. We conclude that the diminished contextual modulation of effective connectivity might reflect a failure to attenuate sensory precision at early processing stages in the visual hierarchy.

ABSTRACT. Colour names are used to partition colour space into discrete colour categories. Individuals with Williams Syndrome (WS) have been shown to develop perceptual categories more slowly than typically developing (TD) individuals (Capirci et al., 1996). Yet the development of colour naming in WS is unclear, as is its relationship to sensory discrimination ability. Here we compared the performance of mental-age-matched TD children (4-7y) and children with WS (8-18y) on two tasks: 1) free colour naming using the Macbeth ColorChecker Chart; and 2) a computer-based chromatic discrimination (CD) threshold test, which isolates cone-opponent and luminance cardinal axes. The WS and TD groups differed significantly in: (1) the proportion of Basic Colour Terms (BCTs) and non-BCTs used (TDs use more modifiers (e.g. light/dark) but WS more combinations of BCTs with non-BCTs); (2) within-group consensus for naming desaturated and non-focal colours (WS less than TD), in particular for grey-scale, beige, turquoise and magenta colours. There were no significant differences in naming focal colours, nor were there differences in chromatic discrimination thresholds along any axis, between the groups. The results suggest that the atypicality and variability in colour naming for non-focal colours in WS is not explained by deficits in chromatic discrimination ability.

ABSTRACT. The balance of neural inhibition and excitation is key to healthy brain function, but may be abnormal in several clinical conditions, including epilepsy (Porciatti, Bonanni, Fiorentini & Guerrini, 2000). The brain’s primary inhibitory neurotransmitter is gamma-aminobutyric acid (GABA). We investigated whether increasing availability of GABA precursors could affect neural responsivity, indexed by steady state visual evoked potentials (SSVEPs). Fourteen participants were shown sine wave gratings flickering at 7Hz at a range of contrast levels, with and without an overlaid orthogonal mask flickering at 5Hz. They then consumed a 5ml daily dose of yeast extract, high in B vitamins and glutamate (both GABA precursors), over a four week period, before being retested using the same stimuli. The dependent variables were the SSVEP responses at the target and mask frequencies. A significant interaction between target contrast and yeast extract (F(2.09,27.17)=4.68, p<0.02) demonstrated that the intervention reduced neural responses at higher contrast levels by up to 20%, but did not affect baseline activity to a blank screen. This was confirmed by a main effect of the intervention on responses to the mask (F(1,13)=5.19, p<0.05). We conclude that dietary intake can influence neural activity, suggesting a potentially valuable supplement to seizure medications.

ABSTRACT. The ability to integrate auditory and visual information is crucial to everyday life and results are mixed regarding how Autism Spectrum Disorder (ASD) influences audiovisual integration. To investigate this question we examined the Temporal Integration Window (TIW), which indicates how precisely sight and sound need to be temporally aligned so that a unitary audiovisual event can be perceived. A total of 26 adult males with ASD and age and IQ-matched typically developed males were presented with flash-beep, body-sound, and face-voice displays with varying degrees of asynchrony and asked to make Synchrony Judgements (SJ) and Temporal Order Judgements (TOJ). Analysis included fitting Gaussian functions as well as using an Independent Channels Model (ICM) to fit the data (Garcia-Perez & Alcala-Quintana, 2012). Curve fitting for SJs showed that the ASD group had a wider TIW, but for TOJ no group effect was found. The ICM supported these results and model parameters indicated that the wider TIW for SJs in the ASD group was not due to sensory processing at the unisensory level but rather due to decreased temporal resolution at a decisional level of combining sensory information.

ABSTRACT. Amblyopia is a developmental visual disorder that affects first-order metrics like acuity and contrast sensitivity. Many amblyopes also suffer losses on higher-order visual tasks, the neural bases of which remain unexplained. We recently reported that sensitivity to the higher-order statistics of naturalistic texture images provides a signature for processing in area V2 (Freeman et al., 2013). We therefore investigated whether amblyopes show deficiencies in detecting these statistics and whether there is a corresponding neural deficit in V2. We tested 5 amblyopes (4 macaques,1 human) using spatial 2AFC. They discriminated texture patterns retaining variable amounts of the higher-order statistical structure of original natural images from noise images that retain only the orientation and spatial frequency content. All amblyopes were impaired on the discrimination when viewing with their amblyopic eyes. To investigate any neural deficit, we measured sensitivity to naturalistic structure in recordings from 5 amblyopic macaques under anesthesia. We used 96-electrode “Utah” arrays to record multiunit activity and found that V2 sites driven by the amblyopic eye showed a reduced ability to distinguish naturalistic images from their noise counterparts relative to the fellow eye. We conclude that amblyopia modifies the processing of naturalistic visual structure.

ABSTRACT. Embodiment is the experience of one’s own body. The Rubber Hand Illusion (RHI) alters embodiment such that participants feel that a rubber hand becomes their own hand. Following the illusion participants also recognise the appearance of the rubber hand as being more like their own hand. This short term effect of the illusion is moderated by an individual’s awareness of their internal body sensations or interoceptive sensitivity. To assess these effects over time we conducted the RHI and measured visual changes over two sessions, one week apart (N=40). Each day participants listened to a 15 minute audio guide. Half listened to a Body Scan Meditation guide designed to increase awareness body (BSM-group) and half listened to a factual anatomy guide (control-group). Immediately following the illusion, both groups recognised the appearance of the rubber hand as more like their own hand. One week later this effect was still present in the control-group (8% vs. 6%, p=.304) but was significantly reduced in the BSM-group (9% vs. 1%, p=.008). The magnitude of the long term effect was negativity correlated with increases in interoceptive sensitivity (BSM-group: r2=40.6%, p=.011, overall: r2= 23%, p=.007), indicating that focussing on internal body sensations reduces illusory body experiences.

ABSTRACT. Affordances reflect the relationship between action capacity (e.g., grasping ability) of the observer and action potentials (e.g., whether an object can be grasped). The action-specific account of perception suggests that our perception of spatial features of objects scales according to our perceived action capacity (Proffitt & Linkenauger, 2013). For example, Linkenauger, Witt and Proffitt (2011) reported that estimates of object size are scaled according to the perceived maximum grasp of the acting hand. We examined whether two independent judgements (affordances and object size) were conflated in Linkenauger et al's (2011) study, such that the reported scaling effects represented changes in perceived affordances rather than spatial perception. In the present studies, these two judgements were clearly distinguished. Participants estimated both their maximum grasping capacity for each hand in a perceived affordance task, and they also visually matched the physical size of objects in a spatial perception task. We found no effect of grasping capacity on object size perception. In addition, when action capacity was restricted, estimates of affordances reflected this change but object size estimates remained unchanged. Thus changes in action capacity influenced perception of affordances, but not spatial perception. Our results provide evidence against the action-specific account of perception.

ABSTRACT. Previous research found that a patient with cortical blindness (homonymous hemianopia) was able to successfully avoid an obstacle placed in his blind field, despite reporting no conscious awareness of it (Striemer, Chapman & Goodale, 2010). This finding led to the suggestion that dorsal stream areas, that are assumed to mediate obstacle avoidance behaviour, obtain their visual input primarily from subcortical pathways. Thus, it was concluded that normal obstacle avoidance behaviour can proceed without input from the primary visual cortex (V1). Here we tried to replicate this finding in a larger patient population (N=6) and also tested if obstacle avoidance behaviour still occurs when obstacles are placed simultaneously in the blind and intact visual field. We have found that even though patients successfully avoided obstacles placed in their intact visual field, they were not able to avoid obstacles in their blind field. These effects were not dependent on whether one or two obstacles were presented. Overall, our findings indicate that normal obstacle avoidance behaviour in the absence of V1 input is a rare occurrence.

ABSTRACT. Much is known about spatial corrections to consecutive reaching movements, but little about the role of temporal error signals. We could use the temporal error (i.e. time-gap between the target and the hand at the aimed position) from previous trials to make adjustements. This requires judging the target speed which can be noisy. Alternatively, subjects could approximate this temporal error by combining the spatial error and the velocity of the hand at the moment of impact. We test this hypothesis in a interceptive task in which the target moved at different speeds and with a designated interceptive area (no spatial error). We analysed the dependency of the next movement on the previous physical temporal error and the proposed combination. First, hand speed combined with the spatial error explained most of the the temporal error, so it could be used to adjust future movements. By applying linear mixed models we analysed the sequential structure and revealed a significant dependency of hand speed at interception on the previous values of hand speed and spatial error, while the physical temporal error had no significant effect on the next movement velocity. The results suggest that subjects do not use previous temporal errors.

ABSTRACT. Rapid motor responses to visual stimuli can involve both activation and inhibition of motor responses. Here, we trace the early processing dynamics of response generation, examining whether activation and inhibition events form a strict sequence when elicited by sequential stimuli, as expected when motor events are carried by fast stimulus-triggered feedforward sweeps. We investigate response priming and negative compatibility effects (NCE) with primed pointing movements going in ten possible directions with response-relevant, response-irrelevant, or no masks interleaved between primes and targets. We show that in response priming (short prime-target intervals, positive priming effects), initial responses are controlled exclusively by the prime. In contrast, in the negative compatibility effect (long prime-target intervals, negative priming effects), even the earliest movement phase is controlled jointly by prime, mask, and target information, and there is a massive force in counterdirection to the primed response that reverses priming effects specifically in slow responses. Thus, response priming reflects a strict sequence of feedforward response activations, while activation/inhibition events in the NCE are not strictly serial but integrate information from different stimuli over time. Even though mask features and visual attention modulate the NCE, its major source is a mask-induced, direction-specific thrust reversal of the initial response.

ABSTRACT. Humans achieve better performance when reaching and grasping in the lower visual field (LVF) than in the upper visual field (UVF). Moreover, the brain regions involved in visuomotor control also show a LVF preference for hand actions (Rossit et al., 2013). It has been suggested that this LVF specialization for action is linked to experience, as most of our actions occur in our LVF, however this claim has never been directly investigated. Juggling involves intense practise of LVF movements and interestingly learning how to juggle produces structural changes in visuomotor brain regions (Scholtz et al., 2009). For the first time we investigated whether the LVF advantage for action is directly related to experience by comparing the performance of jugglers to non-jugglers. Participants were asked to grasp objects positioned in the LVF and UVF. In line with previous research, we found an advantage for grasping in the LVF: maximum grip aperture (MGA) was wider and more variable in the UVF than in the LVF. Remarkably, jugglers achieved MGA quicker and presented higher peak velocities than non-jugglers, particularly in the LVF. These results suggest that the LVF specialization for action is shaped by experience.

ABSTRACT. The Material-Weight Illusion (MWI) is an example of how visually evoked expectations influence perception and action. Objects that appear to be made of a lighter-looking material, e.g. polystyrene, are perceived to be heavier than equally weighted and sized objects of a heavier-looking material such as stone. Grip and load forces are initially driven by these visual expectations and soon adjusted to the actual mass of the objects (Buckingham, Cant, & Goodale, 2009). Real-world objects are often made of more than one material. In the present study we therefore investigated the perceived heaviness of symmetrical blocks whose two halves appeared to be made of different materials (polystyrene, wood and stone), whereby their true mass was identical (400 g) and evenly distributed around the geometric center. 48 participants lifted the bipartite objects at a central handle, while forces were recorded, and were then asked to judge the weight or center of mass of the objects. Prior expectations seemed to affect heaviness perception. In contrast to the classic MWI, the heavier-looking side of the object was perceived to weigh more. This perceptual illusion lasted until the end of the experiment. No analogous effect was observed on the forces applied to lift the objects.

ABSTRACT. An object seen through a window appears to shrink in apparent size as the observer approaches the window. Paradoxically, the distant object appears smaller as its visual angle increases. The effect was first reported by Cornish (1935) and baptized 'vista paradox' by Walker, Rupich, & Powell (1989), who investigated it experimentally in relation to approaching a window and looking at a scene outside the window. However, in this study there was no control over where the observers were fixating. We investigated the vista-paradox by varying object size, distance, point of fixation, texture of the frame and of the object. It turned out that fixation is essential for the illusion. Fixating the window frame led to an apparent shrinking of the object, whereas fixation on the object did not. Texture of the frame intensified the apparent shrinking of the object. Quite paradoxically, when separating the point of fixation from the frame, the illusion persisted. It also persists when removing the window altogether. That is, the window or frame is dispensable for the vista paradox.

ABSTRACT. Perception is usually non-retinotopic. For example, a reflector on the wheel of a bicycle is perceived to rotate on a circular orbit, while its retinotopic motion is cycloidal. To investigate non-retinotopic motion perception, we used the Ternus-Pikler display. Two disks are repeatedly flashed on a computer screen. A dot moves linearly up-down in the left disk and left-right in the right disk (retinotopic percept). If a third disk is added alternatingly to the left and right, the three disks form a group moving predictably back and forth horizontally. The dot in the central disk now appears to move on a circular orbit (non-retinotopic percept), because the brain subtracts the horizontal group motion from the up-down and left-right motion. Here, we show that predictability is not necessary to compute non-retinotopic motion. In experiment 1, the three disks moved randomly in any direction. In experiment 2, we additionally varied the shape and contrast polarity of the stimuli from frame to frame. In both cases, strong non-retinotopic rotation was perceived. Hence, the visual system can flexibly solve the non-retinotopic motion correspondence problem, even when the retinotopic reference motion is unpredictable and no efference copy-like signals can be used.

ABSTRACT. Alignment shifts in the Poggendorff figure (PF) have been explained by cross-orientational inhibition in visual cortex (Blakemore et al, 1970). Hotopf et al (1972a,b,1974) matched the orientation of a test line with a neutral line; Blakemore et al, rotated the line to point at a distant dot. In both cases the shift (< 1 deg) was much too small to account for the PF in its entirety. We used a 2AFC task designed to reduce response biases . Observers (n=8) saw 2 figures and had to decide in which of them a pointer was more aligned with another pointer or a distant dot. We confirmed Blakemore et al (p=0.0006) but the mean effect was small (-0.92 deg over subjects), whereas in the PF it was much larger (5.2 deg; p=.0002). A small gap of the apex of the Blakemore et al stimulus reduced their effect still further (0.42 deg; p=0.18). Hotopf et al note other factors that could boost the PE and we have discovered another: the perceived angle of virtual lines also shows repulsion from the vertical parallels (2.21 deg; p=0.007). Like other ‘illusions’ the PF has evolved to be conspicuous rather than informative.

ABSTRACT. The Ebbinghaus illusion has been extensively studied, but there is no consensus about its explanation. The phenomenon was often described essentially as follows: identical targets look smaller/larger when surrounded by several larger/smaller figures. This description in terms of size contrast has inspired a corresponding class of explanations, which claims that size contrast is the cause of the phenomenon. However, the illusion can also be described in terms of the distances of the surrounding figures rather than their sizes: identical targets look smaller/larger when surrounded by several further/nearer figures. Distance has indeed long been recognized as an independent factor affecting the strength of the illusion. We report four experiments using variants of the Ebbinghaus display, whose purpose was to test whether size of surrounding figures as such is an independent causal factor. For better control of distances between the elements of the stimulus constellations, figures with straight contours were used instead of conventional circular shapes. We found that in one variant the illusion was reversed despite the presence of size contrast, whereas in another variant it was present despite the absence of size contrast. The results strongly suggest that size is not a separate causal factor of the Ebbinghaus illusion.

ABSTRACT. Images disjointed in space and time evoke the perception of moving objects by virtue of spatiotemporal grouping. We study how spatiotemporal proximity between image parts determines the perception of motion and spatial configuration. We use two types of multistable motion displays: motion lattices and rotating dipoles. In motion lattices, alternative groupings underlie perception of motion in different directions (Gepshtein & Kubovy, 2000). In rotating dipoles, alternative groupings underlie perception of different spatial configurations: multiple spinning dot pairs vs. two pulsating rings of dots (Anstis & Kim, 2011). For perception of both motion and spatial configuration, we identify points of equilibrium: the conditions at which alternative percepts are equally likely. We find that individual points of equilibrium of perceptual organization are predicted from the individual limits of contrast detection summarized by individual spatiotemporal contrast sensitivity functions. Our results demonstrate that perception of motion and spatial configuration in high-contrast discrete visual displays fall under the same constraints as detection of continuous visual stimuli near the threshold of visibility.

ABSTRACT. We have previously demonstrated that a perceptual object can attract attention automatically (Kimchi et al., 2007). Here we explored: (a) Does the attentional capture by an object involve a spatial component? (b) Which Gestalt factors suffice for an object to capture attention? (c) Does the strength of organization affect the object’s ability to capture attention? Participants viewed multi-elements displays and identified the color of one element or responded to a Vernier target. On some trials, a subset of the elements grouped by Gestalt factors into an object that was irrelevant to the task and unpredictive of the target. Object effect – faster performance for targets within than outside the object – was found also when the target appeared after the object offset, suggesting deployment of attention to the object location. Object effects of similar magnitude were found for objects grouped by collinearity, closure, and symmetry combined, by closure and symmetry, or by collinerity, but not symmetry, suggesting that collinearity alone, and closure and symmetry, suffice for attentional capture by an object, but symmetry does not. Finally, grouping strength in modal completion, manipulated by varying contrast polarity between and within elements, affected the effectiveness of the attentional capture by the object.

ABSTRACT. Visual object recognition typically happens very fast and it has therefore been difficult to disentangle its constituent processes. Extended recognition times have been observed for images with emergent properties, suggesting these may help examine perceptual and cognitive processes in object recognition. Until now, their use has been constrained by limited availability. Here, we used a new set of 15 stimuli with emergent properties – akin to the famous Gestalt image of a Dalmatian – in combination with eye tracking to examine the processes underlying human object recognition. Recognition times were relatively long (median ~ 5 s), confirming the objects' emergent properties. Surprisingly, already within the first 500 ms, the majority of fixations were aimed at the hidden objects. Contemporary saliency models – that emulate the early feature processing stages of human vision – fail to predict these eye movements. The fast detection suggests that observers selected potentially relevant image sections based on image statistics presently not captured by these saliency models. The quick detection yet slow recognition of emergent images points towards discrete contributions of perceptual and cognitive processes to human object recognition. This may eventually aid the development of better biologically plausible computer vision algorithms.