Children with amblyopia read slower than their peers during binocular viewing. Ocular motor dysfunction typical of amblyopia may cause slow reading. It is unclear whether this is due to fixation instability or increased forward saccades. We examined whether removing the requirement of inter-word saccades helps children with amblyopia read at a similar rate as controls using a rapid serial visual presentation (RSVP) task. We also assessed whether reading rate was related to fixation instability. Children with amblyopia (n = 32) and control (n = 30) children ages 8-12 years silently read sentences presented in RSVP (single word presentation at screen center) during binocular viewing. Exposure time per sentence changed with a 2 - down 1 - up staircase to obtain reading speed thresholds (log words/minute [WPM]). Eye movements were tracked to determine fellow eye (FE) and amblyopic eye (AE) fixation stability during RSVP reading. Children with amblyopia read slower than controls (2.75 ± 0.47 log WPM vs 3.06 ± 0.40 log WPM), and had increased AE fixation instability (0.21 ± 0.39 log deg vs - 0.20 ± 0.18 log deg) and increased FE fixation instability (-0.03 ± 0.34 log deg vs - 0.20 ± 0.15 log deg) during RSVP reading. Reading rate in amblyopic children with good FE stability (n = 11) did not differ from controls and was faster than those with poor FE stability (n = 21). Children with poor FE stability read slower than controls. Removing the need for inter-word saccades (i.e., RSVP reading) did not help children with amblyopia read at control speeds. Our data support FE fixation instability as a source of slow reading in amblyopia.

Previous studies have demonstrated that density of texture elements is an important perceptual aspect of textural appearance, and can enable texture segmentation in the absence of other cues. We compared segmentation thresholds for two kinds of second-order boundaries comprised of two species of micropatterns (e.g., horizontal and vertical Gabors): (1) Feature boundaries, with the same number of total micropatterns on opposite sides but different numbers of each micropattern species within each side, and (2) Density boundaries, with different numbers of total micropatterns on opposite sides, but with the same number of both micropattern species within each side. Contrary to the predictions of a standard late-pooling Filter-Rectify-Filter (FRF) model in which different micropattern specific first-order channels are analyzed by different second stage filters before pooling, we observed lower segmentation thresholds for density boundaries than feature boundaries. This suggests that density boundaries may be detected by a different, early-pooling mechanism. In a second experiment, we considered how two species of micropatterns combine for boundary segmentation. When two single-micropattern-species density boundaries are superimposed in-phase to form a new density boundary, the boundaries formed by each micropattern species combine via probability summation. By contrast, when two single-micropattern-species density boundaries are superimposed in opposite-phase to form a feature boundary, segmentation performance is worse than for either single-micropattern-species boundary alone. We conclude that the mechanisms for density-based texture segmentation are not identical to the mechanisms for feature-based segmentation and that density-sensitive mechanisms most likely integrate across multiple first-order filters responsive to different micropattern species.

The stereoscopic system is typically tested by measuring its response to sinusoidal disparity corrugations. Previous research using psychophysical methods like masking and adaptation has provided evidence about the existence of disparity mechanisms tuned to spatial frequency and orientation. Analyses based on individual differences have confirmed only the existence of disparity mechanisms tuned to spatial frequency. The main objective of this study is to investigate the existence of orientation-selective disparity mechanisms using an individual differences approach. We measured the disparity thresholds of 37 subjects using sinusoidal disparity corrugations constructed from dynamic random-dot stereograms with spatial frequencies of 0.1, 0.4, and 0.8 cpd and 7 orientations (ranging from 0° to 90° in steps of 15°). Stereo thresholds for 0.1 cpd showed a strong anisotropy with disparity thresholds increasing from 90° (horizontal) to 0° (vertical). This anisotropy was reduced for the other two spatial frequencies. Component and factor analyses revealed that closer orientations tend to group together, suggesting the presence of underlying orientation-selective mechanisms in stereovision. These results provide new evidence for disparity channels tuned to orientation underlying the processing of disparity corrugations.

Conventional visual acuity (VA) tests, using black optotypes on white background, may not fully detect subtle ON-pathway amblyopia deficits or increases in visual dark dominance. This pilot study tests the hypothesis that VA testing with reverse contrast polarity, white optotypes on black background, is more sensitive in detecting amblyopia than conventional VA testing. Two groups of children aged 8-12 were enrolled. The amblyopic group (N = 13) had a best-corrected VA of 20/32 or worse in the amblyopic eye and an interocular VA difference of ≥ 0.2 logMAR. The normal control group (N = 16) had best-corrected VA equal to or better than 20/25 and an interocular VA difference of ≤ 0.1 logMAR. Monocular VA was tested for each polarity using an ETDRS program using the Amblyopia Treatment Study protocol, with results reported as a Score. Two optotype polarities were tested in randomized order, and tests were repeated after a 30-minute interval. VA was compared among amblyopic, fellow, and the right eyes of the control group for both polarities. For amblyopic eyes, the mean VA Score was significantly lower for white optotypes than for black optotypes by 3.6 letters or approximately 0.07 logMAR, indicating that amblyopic eyes had more difficulty seeing white than black optotypes. For fellow eyes, the mean Score was also significantly lower for white than for black optotypes. No significant differences in mean Scores between two polarities were found within the control eyes. Our findings suggest that visual acuity testing with reverse polarity could be more sensitive to detect amblyopia in children.

As a classic theory in cognitive psychology, Gestalt theory elucidates basic principles in visual perception. However, the Gestalt principles are validated mainly by psychological experiments, lacking quantitative research supports and theoretical coherence. In this paper, we utilize persistent homology, a mathematical tool in computational topology, to develop a unified computational model for Gestalt principles, addressing the challenges of quantification and computation. This Gestalt computational model provides a quantitative approach to several key Gestalt principles, and it shows that these Gestalt principles can be uniformly calculated using persistent homology, thus developing a coherent theory for Gestalt principles in computation. Moreover, it is anticipated that the Gestalt computational model can serve as a significant computational model in the field of computational psychology, and help the understanding of human visual perception.

Attention-Deficit/Hyperactivity Disorder (ADHD) is a prevalent neurodevelopmental condition in young populations and is often associated with visual disturbances. This study investigated clinical measurements in individuals with ADHD and compared them with individuals without ADHD. This study included 80 participants: 40 with ADHD and 40 age- and sex-matched controls. The mean age ± standard deviation was 19.93 ± 6.42 (range 6-36). The optometric tests performed assessed accommodative and binocular visual functions, including accommodative amplitude, monocular and binocular accommodative facility, Cover Test, near point of convergence, monocular estimated method, and calculated accommodation convergence/accommodation ratio. Subgroup analysis within the ADHD group, based on sex and medication use, explored the associations with optometric findings. The ADHD group showed a significantly higher lag of accommodation (+0.30 ± 0.17 D) than the control group (+0.18 ± 0.23 D) and a higher accommodative amplitude (ADHD group = 13.15 ± 1.73, control group = 12.07 ± 1.60). ADHD was also associated with a higher prevalence of near-heterophoria. No differences were observed between medicated and medication-free at testing individuals with ADHD. Furthermore, males demonstrated higher near heterophoria than females, with males having significantly more phorias at near. The results of this paper demonstrate that individuals with ADHD commonly experience compromised accommodative responses and a heightened occurrence of heterophoria compared with individuals without ADHD. These results emphasize the importance of eye exams for ADHD patients, as vision issues may reduce attention and concentration, especially for near-work tasks.

Populations of sensory neurons are not homogeneous. Even neighboring neurons located in the same brain area can process identical stimuli in significantly different ways. Retinal ganglion cells (RGCs) are a prominent example of such heterogeneity - they exhibit diverse properties whose computational role and purpose remain mysterious. In this review, we explore normative theories of neural computation that attempt to explain the origins and role of functional variability in the retina. We first express a general mathematical formulation of normative theories of neural computation and identify components of these theories that can explain the heterogeneity of sensory populations. We then organize existing theoretical studies of retinal coding according to the factors they highlight as explanations of the computational diversity in the retina - the beginning of the visual hierarchy.

We examined the relationship between stereopsis and fusional vergence in groups of amblyopic and stereo-normal control observers. As absolute disparity is thought to be the basis for relative disparity and for disparity-driven vergence, we hypothesized that vergence anomalies would be accompanied by impaired stereopsis. Specifically, we examined whether patterns of impaired stereopsis across the central 20° of the visual field were accompanied by impaired fusional vergence for stimuli confined to these regions. Stereopsis was measured locally across the visual field with disparity steps of 5 to 20 arcmin. Fusional vergence to large disparity steps (2 to 3°) was measured with binocular eye tracking. The vergence stimuli were random dot stereograms, in one of 3 spatial configurations: a large disc 16° in diameter, a small disc 4° in diameter, and an annulus with outer and inner diameters corresponding to the large and small discs. Of the controls (n = 25) with no history of abnormal visual development, 12 individuals exhibited normal stereopsis across the visual field and normal vergence gains for all configurations. Thirteen individuals with weak stereopsis in the central field tended to have anomalous vergence for small stimuli, but normal vergence for larger stimuli. Amblyopic/strabismic individuals (n = 12) had poor stereopsis and poor vergence for small stimuli. We report a strong correlation between vergence, coarse and fine stereopsis, with no double dissociation (no cases of impaired vergence with normal stereopsis). Taken together, the results suggest that compromised binocular interaction is the cause of both stereopsis and vergence deficits.

Monocular deprivation (MD) amblyopia caused by a dense unilateral congenital or infantile cataract leads to both sensory and ocular motor deficits, which can in turn affect motor performance. Previous research shows reduced fine motor skills in children with MD amblyopia on standardized tasks. Here, we evaluate eye-hand coordination during visually-guided reaching in MD amblyopia. A group of 17 children aged 7-15 years with MD amblyopia resulting from a unilateral cataract and a group of 41 age-similar control children were enrolled. During binocular viewing, children's reaching movements (LEAP Motion Controller) and eye movements (EyeLink 1000 binocular eye tracker) were recorded as they reached to touch a dot displayed at one of four locations (±5 deg or ±10 deg) on a computer monitor. Saccade and reach kinematic measures were assessed between groups, and factors associated with impairments in the MD amblyopia group were evaluated. The MD amblyopia group as a whole had impaired saccade (lower saccade gain, reduced saccade precision, more reach-related saccades) and reach (longer total reach duration, slower peak velocity, reduced touch accuracy) kinematics compared to controls. However, performance was worse in those with a poorer visual acuity outcome (≥0.7 logMAR) compared to good visual acuity outcome (≤0.6 logMAR). MD amblyopia impacts the development of eye-hand coordination during reaching, particularly in those with a poorer visual acuity outcome. Longer deceleration in the final approach and more reach-related saccades may suggest an inability to adapt or form an efficient compensatory strategy and may also be indicative of impaired on-line control.

Scene context is known to significantly influence visual perception, enhancing object recognition particularly under challenging viewing conditions. Behavioral and neuroimaging studies suggest that high-level scene information modulates activity in object-selective brain areas through top-down mechanisms, yet the underlying mechanism of this process remains unclear. Here, we introduce a biologically inspired context-based computational model (CBM) that integrates scene context into object recognition via an explicit feedback mechanism. CBM consists of two distinct pathways: Object_CNN, which processes localized object features, and Place_CNN, which extracts global scene information to modulate object processing. We compare CBM to a standard feedforward model, AlexNet, in a multiclass object recognition task under varying levels of visual degradation and occlusion. CBM significantly outperformed a standard feedforward model (AlexNet), demonstrating the effectiveness of structured contextual feedback in resolving ambiguous or degraded visual input. However, behavioral experiments revealed that while humans also benefited from congruent context - particularly at high occlusion levels - the effect was modest. Human recognition remained relatively robust even without contextual support, suggesting that mechanisms such as global shape processing and pattern completion, likely mediated by local recurrent processes, play a dominant role in resolving occluded input. These findings highlight the potential of contextual feedback for enhancing model performance, while also underscoring key differences between human and models. Our results point toward the need for models that combine context-sensitive feedback with object-intrinsic local recurrent processes to more closely approximate the flexible and resilient strategies of human perception.

Foveal crowding refers to the impaired recognition of a foveal stimulus due to the presence of adjacent flankers. Previous research has produced inconsistent results regarding the maturation of foveal crowding, either at ages 5-7 or remaining elevated from ages 5 to at least 11. We investigated this developmental trajectory using a specialized set of digit stimuli (Pelli fonts) tailored for measuring foveal crowding. We measured foveal crowding in preschoolers, school-age typically developing children, and school-age children with developmental dyslexia, as well as in a group of adults. The results show that foveal crowding decreases with age, reaching adult-like levels around 8 years among preschoolers and typically developing children. Furthermore, dyslexic children exhibited heightened foveal crowding compared to their typical peers by approximately the same amount, regardless of age and reading level. Notably, preschoolers exhibited the most pronounced foveal crowding effects with considerable individual variability: some displayed crowding similar to that of older typical children and adults, while others exhibited similar or even higher levels of crowding compared to dyslexic children. This large variability suggests that foveal crowding may have the potential to serve as an early indicator for identifying developmental dyslexia, a possibility that warrants further longitudinal investigation.

Earlier work has shown that the eyes preferably select stimuli that are presented close to central fixation over stimuli presented further away, suggesting the existence of a central selection bias. However, so far studies have confounded retinal eccentricity with distance from the center of a display, and the observed effects may thus have been driven by what is known as the central fixation bias, which is the preference for items near the center of a display rather than the center of the retina. This study aimed to dissociate the central selection bias from the central fixation bias, and to uncover its time course. In two experiments, participants were instructed to make a single eye movement to one of two simultaneously presented singletons. The singletons were always presented at the same distance from the center of the display (thus controlling for the central fixation bias) but their eccentricity relative to the initial fixation point was varied (thus allowing for a central selection bias to operate). When the two singletons were displayed at different eccentricities, participants preferred selecting the nearest item. This central selection bias occurred rapidly and transiently, peaking around 230 ms and lasting until approximately 320 ms after display onset. Together, these results suggest that retinal eccentricity is a major factor when multiple objects compete for selection.

Eye-hand coordination is a key aspect of visuomotor control essential for performing most daily activities. Disruption in visuomotor control, characterized by slower arm movements and grasping errors, has been documented in children with amblyopia. This study aimed to characterize the effects of amblyopia on the temporal pattern of eye and hand coordination during the performance of a task that involves reaching, precision grasping, and placement. The study recruited 28 children with a history of amblyopia and 56 typically developing peers (age range 6-14 years). Children performed a bead-threading task while their eyes and hand movements were recorded concurrently. As hypothesized, children with amblyopia demonstrated poorer task performance, with greater deficits for the object manipulation compared to the reaching (transport) components. In comparison to their peers with normal vision, children with amblyopia had shorter reaction time for initiating eye and hand movement, longer object fixation duration to guide grasp execution and object placement, and lower eye-hand latency difference for the second movement indicating that the hand movement preceded eye initiation. These results suggest that children with amblyopia have poorer motor planning ability, which impacts movement execution. Longer fixations during object manipulations indicate that more time is required to transform the noisy visual input into a motor response. Overall, the study adds to the growing body of evidence highlighting deficits in visuomotor control in amblyopia.

Humans not only perceive material properties of natural surfaces but also evaluate their affective qualities, such as pleasantness or unpleasantness. Recent psychophysical studies suggest that such emotional impressions can arise directly from low-level image statistics, independent of object recognition. To elucidate the neural mechanisms underlying these immediate affective responses, we recorded visual evoked potentials (VEPs) while participants viewed 150 images of natural surfaces varying in affective valence. We identified occipital VEP components emerging around 100-150 ms after stimulus onset that were significantly correlated with subjective unpleasantness ratings. Moreover, these unpleasantness-related VEPs were accurately predicted by a linear combination of VEP components associated with a small set of diagnostic image statistics. Our findings indicate that early visual cortical activity encodes image features that give rise to unpleasant affective responses, supporting the notion that rapid, low-level visual processing can directly contribute to the emotional evaluation of visual textures and materials.

Emerging treatments, including virtual reality (VR)-based therapies, video games, and movies, have been proposed to enhance stereoacuity in individuals with binocular vision disorders such as amblyopia and strabismus. However, their comparative effectiveness remains uncertain. This systematic review and meta-analysis aimed to evaluate the effectiveness of these emerging treatments in improving stereoacuity through within-group analyses, and to compare their outcomes with occlusion, in studies with direct group comparisons. We conducted comprehensive literature searches in PubMed, MEDLINE, Cochrane Library, Scopus, and Web of Science. Eligible studies included randomized controlled trials, cohort studies, and case-control studies reporting stereoacuity outcomes. The primary outcome was the change in stereoacuity (log arcsec). A random-effects meta-analysis, subgroup comparisons, and meta-regressions were performed. Twenty-six studies were included. The pooled mean improvement in stereoacuity was 0.26 log arcsec, i.e. a factor of 1.82 (95% CI: 0.19-0.33). Emerging treatments yielded significant within-group improvements, with no significant difference compared to occlusion therapy. VR-based interventions did not show statistically significant advantages over non-VR binocular treatments. Movies showed slightly greater gains than video games, but differences were not significant after correction. In regression analyses, no predictors remained significant after Bonferroni correction. Heterogeneity was moderate, reflecting variability across studies. In conclusion, emerging therapies demonstrate measurable benefits in enhancing stereoacuity. However, they have not consistently outperformed occlusion.

Visual patterns with large shifts in chromaticity can be perceived as uncomfortable and may even trigger photosensitive epilepsy. In the temporal domain, not only does large chromatic contrast tend to increase discomfort, but red flicker also appears to evoke strong aversion. The mechanisms underlying this red-specific discomfort, however, remain unclear. To address this issue, the present study measured discomfort ratings for two-color alternating stimuli defined in the MB-DKL cone-opponent space, in which the magnitudes of modulation along the L/(L + M) and S/(L + M) axes were systematically manipulated. We found that modulation along the L/(L + M) axis showed a positive correlation with discomfort ratings, whereas modulation along the S/(L + M) axis also predicted discomfort but with a comparatively weaker effect. We also replicated previous findings that greater temporal chromatic contrast, quantified in a uniform color space, was associated with increased discomfort. Importantly, the relationship between L/(L + M) modulation and discomfort remained robust even when chromatic contrast was controlled. These results suggest that visual discomfort from temporal chromatic variation involves multiple stages of color processing. Specifically, the aversion to red flicker may originate at early stages of cone-opponent processing, prior to the formation of categorical red perception.

Visual discomfort, the unpleasant, aversive experience associated with some visual stimuli, is most pronounced for flickering and spatially repetitive stimuli. It has been proposed that the degree of visual discomfort for such stimuli can be predicted by the contrast sensitivity function, peaking at midrange spatial and temporal frequencies. We evaluated the spatio-temporal tuning of visual discomfort for flickering, sinusoidal stimuli. Discomfort increased with spatial frequency for static and slowly flickering stimuli, but decreased with spatial frequency for stimuli flickering at 16 Hz. Discomfort increased with temporal frequency for spatially uniform stimuli, and for all spatial frequencies. Flickering stimuli were more uncomfortable than static stimuli of any spatial frequency. Spatially uniform stimuli flickering at 16 Hz, the highest frequency tested, were rated as the most uncomfortable. These results deviate from the contrast sensitivity function, which predicts that discomfort should be highest for static stimuli, with bandpass spatial frequency tuning. This discrepancy indicates that threshold-level visual sensitivity is not a good predictor of visual discomfort for high contrast stimuli. Our results are however consistent with efficient coding models, which predict higher levels of excitation for high spatial and temporal frequencies when stimuli are presented at a high contrast. They are also consistent with physiological measures of cortical responses to high contrast stimuli.

Cortical prostheses offer the potential for partial vision restoration in individuals with blindness by stimulating neurons to produce phosphenes. However, the low number of discrete phosphenes that can be simultaneously elicited in practice makes encoding of whole objects difficult, hindering recognition. Hence, we aim at determining the minimal number of discrete visual elements needed for object recognition. To answer this question, we replaced object contours by a varying number of round patches, serving as idealized phosphene simulation. Forty-six sighted participants identified the fragmented objects in a free-naming task. We increased the number as well as the patch density gradually until the objects were correctly identified. Depending on the object, we found that a minimum of 29 and a maximum of 65 round patches were necessary for recognition. While this still exceeds the current number of phosphenes which can simultaneously be elicited in cortical visual protheses, we found much smaller estimates than previous studies of simulated prosthetic vision. We next investigated whether local visual information of non-circular patches could be leveraged to reduce the number of discrete elements necessary for object recognition. For this, objects were fragmented into straight and curved segments matching the size, density, and location of the previously used round patches. Participants required 27% fewer segments than round patches for recognition, indicating that contour information leads to perceptual benefits. Our study provides a lower bound estimate on the minimal number of phosphenes needed to recognize objects, should simultaneous stimulation protocols be pursued. These findings provide realistic bounds for the design of visual prostheses.

The Twinkle-Goes illusion is an apparent perceptual extrapolation of moving objects that suddenly disappear. Object disappearance must occur against a white noise background that is dynamically updating, or against an initially static white noise background that becomes dynamic within a brief time window after disappearance. Here, we have refined the critical time window for dynamic noise onset to just ∼ 20 ms after moving object disappearance. We have also successfully measured the illusion using a sensory reproduction task, that minimizes decisional biases that could have influenced the binary forced choice tasks that have previously been used to measure the illusion. This suggests that the Twinkle-Goes illusion is likely perceptual in origin. We also measured the illusion using a task where people made a saccade to the perceived moving object disappearance position, showing the illusion impacts motor planning. The magnitude of the effect in our data was ∼ 22 ms, at most. This is too slight to reflect on full compensation for delayed sensory processing in early visual brain regions (∼67 ms). However, it is possible that our measures have only partially captured the activity of an extrapolation process that is involved in a full compensation for information processing delays, in order to facilitate motor planning - but this possibility remains speculative, and we review counter arguments and evidence.

In a uniformity illusion, participants experience that peripheral stimuli appear identical to the central stimulus even though they are different. The uniformity illusion generally occurs after prolonged central fixation. The uniformity illusion thus seems to evoke illusory peripheral perceptions of colour, shape, size, and even movement. This could be the result of a passive process where the periphery is unattended and ignored, resulting in erroneous reporting of peripheral stimulus properties. The current experiment, however, points to an active filling-in process that resulted in an actual percept of the periphery by showing that the illusory periphery is "sticky": Even after the central stimuli inducing the illusion were removed, the peripheral illusion persisted, and continued to influence perceptual reports of the participants. Participants viewed displays featuring colour or size variations between the center and periphery for a set time to induce the uniformity illusion, reporting if they experienced a uniform screen. Subsequently, the central patch was altered to match the original periphery, creating a truly uniform display. Participants then evaluated whether the new display appeared uniform and reproduced the size or colour of the peripheral texture. Reaction time and reproduction accuracy revealed that experiencing an illusory periphery interfered with processing subsequent physical stimuli, especially in trials where participants explicitly reported the illusion. These findings suggest that the uniformity illusion can produce a persistent illusory periphery, which disrupts the perception of the actual peripheral stimulus even seconds after the original display has been modified. The results underscore the active, hierarchical nature of perceptual reconstruction in visual processing.

Some exemplars are more representative of their category than other exemplars. Here we ask whether good exemplars (high representativeness) of real-world scene categories are more informative with respect to their category than bad exemplars (low representativeness) by leveraging machine learning methods and a set of deep neural networks, including convolutional neural network (AlexNet, PlaceNet, ResNet), and transformer models (Vision transformer, transformer-based CLIP). We use a one-shot Support Vector Machine, in which the training set only contains one exemplar per category (two images in total) to ask whether good exemplars generalize to other members of their category better than bad. The resulting classification accuracy is interpreted as reflecting the degree to which information regarding category is present in the training exemplars. We used four natural scene categories (beaches, cities, highways, mountains). Experiment 1 showed good exemplars produced higher classification accuracy than bad exemplars in all the features tested. Experiment 2 demonstrated that multiple bad training exemplars were needed to reach the category information of a single good exemplar. Both experiments indicate that there is more category-related information in good exemplars than bad. Experiment 3 showed that the most informative images and the good exemplars did not fall in the center of the category space but rather towards the edge of the space, and this was true of the category spaces derived from the feature spaces and human similarity judgements. Overall, this work demonstrated that DNNs can capture human representativeness and provide a useful measure for capturing human scene category spaces.

A well-documented phenomenon in research on eye movement control during reading is the systematic relationship between the landing positions of forward saccades and target word characteristics. However, the behaviour of regressive saccades, which move the eyes in the opposite direction, remains less explored. This study delves into the landing positions of regressive saccades, emphasizing the distinction between intra-word and inter-word regressions, across diverse languages. Using data from the MECO L1 project, which includes eye-tracking data from 589 participants across 13 languages, we scrutinize the precise landing positions of regressions vis-à-vis forward saccades. Our analysis shows a robust effect of launch distance on landing positions for progressive saccades, with undershoots increasing as launch distance grows and overshoots with shorter launch distances. In contrast, regressive inter-word saccades show only minimal variation in landing positions, typically landing near the centre of the target word regardless of launch distance or word length. Intra-word regressions, however, display a pattern similar to progressive saccades, where the landing position is influenced by launch distance, tending to overshoot the optimal viewing position as the launch site moves away from the word's end. This pattern is consistent across all languages. These findings support the notion of cross-linguistic universality in oculomotor control mechanisms during reading, particularly the precision of regressive saccades. They align with the spatial coding hypothesis, suggesting that precise spatial memory of word positions guides regressive saccades.

Amblyopia is a common disorder of spatial vision and is frequently associated with the presence of anisometropia, strabismus, or both, during visual development. For highly visible stimuli, subjects with strabismic amblyopia often report marked spatial distortions, but the neural basis of this supra-threshold deficit is not well understood. Here, we used a combination of behavioural measurements and visual field mapping with high spatial-resolution functional magnetic resonance imaging (fMRI) at 7 T to assess perceptual distortions in 12 participants with strabismic amblyopia and 9 control subjects. We measured both behavioural and cortical visual field maps monocularly through each eye. Although amblyopic subjects showed increased perceptual distortions, the layout of V1 maps, as measured through the eccentricity and size of population receptive fields, was largely unaltered compared to controls, with no discernible difference in cortical magnification between groups. This suggests that disruptions to V1 retinotopy do not explain the perceptual distortions experienced by amblyopes.

Humans frequently encounter crowds in daily life, and the collective emotional state of these groups provides vital social information that influences behavior and decision-making. Variance in facial expressions within a crowd serves as an indicator of the diversity of emotions, modulating the strength and stability of the group's emotional signal. Previous studies of low-level visual features have shown that the visual system adapts to the statistical properties of variance, producing systematic aftereffects in subsequent variance perception, such that the perceived level of variance shifts in the direction opposite to the adaptor. However, it remains unclear whether similar adaptation to variance occurs for complex, socially meaningful information such as facial expressions. In this study, we examined whether adaptation to the variance of facial expressions in crowds leads to aftereffects in perceived variance. Using morphed facial stimuli that varied incrementally between happy and angry expressions, we created crowd images composed of individuals with different degrees of emotional variability. In Experiment 1, participants judged the variance of facial expressions before and after adapting to stimuli with small or large variance. Experiment 2 examined adaptation using more intense expressions to enhance perceived variability. Across both experiments, perceived emotional variance shifted in the opposite direction to the adaptor, indicating robust aftereffects of facial-expression variance. These findings provide behavioral evidence consistent with the idea that the human visual system encodes the variance of facial expressions, and that adaptation to ensemble variance dynamically recalibrates perception in social contexts.

Topographic maps early in visual processing preserve the spatial relations of visual stimuli but the metric relationships between these visual directions is not directly accessible. To investigate the magnocellular pathway's role in metric spatial vision, we employed an adaptation paradigm. Exposure to a 60 Hz flickering disc array (subjectively invisible) induced a systematic compression in the perceived distance between subsequently presented dot pairs. This compression was strongest when adaptation preferentially modulated low spatial frequency channels, consistent with the properties of transient channels tuned to low spatial and high temporal frequencies. Crucially, this compression was attenuated when the adaptor consisted of two cyan lattices rotating on a magenta background near isoluminance, as confirmed by a global motion direction discrimination task. The same pattern emerged when test dots were isoluminant with the background, ruling out test-adaptor similarity as a critical factor. Finally, an isoluminant red-green adaptor flickering on a yellow background induced compression at 3 Hz, but not at 60 Hz. This dissociation aligns with the known properties of magnocellular neurons, which are insesitive to high temporal frequency isoluminant red-green modulation, but can respond to slow isoluminant red-green modulations. These findings reveal a novel role of the magnocellular pathway in metric spatial vision.

Recent behavioral studies suggest that rapid short-term plasticity in adult binocular vision may occur over several minutes. In the example of adult monocular patching, a difference in luminance between the eyes may cause plasticity. Analogous work in animal models exists, but relevant physiological data from humans is sparse. We measured human binocular function with fMRI by determining how the BOLD signal evoked by test stimuli in one eye is modified by the effects of fellow eye stimulation. The tested eye viewed moving checkerboard patterns that varied in contrast. Three fellow eye viewing conditions were compared over various time periods: black screen, gray screen, or the same checkerboard as the tested eye. Overall, the same checkerboard in both eyes elicited the highest brain activity in early visual cortices and area MT+. In addition, we found higher brain activity for a black screen compared to gray screen in early visual areas, consistent with luminance masking. However, this effect is absent if the luminance presented to the fellow eye changes several times per minute, rather than remaining constant for several minutes. We suggest that longer viewing periods allow for greater fellow eye adaptation effects to accumulate over time. Results are discussed along with previous studies of monocular patching and contrast normalization models of vision.

The dragonfly visual system contains neurons that respond to the movement of small targets, even when embedded in visual clutter. Some of these target-detecting neurons facilitate their spiking activity to targets moving along continuous trajectories and selectively attend to one target when presented in a pair. When a target is rapidly repeated, the spiking activity of these 'small target motion detector' (STMD) neurons decreases. This 'adaptation' is spatially localised with a time course dependent on the frequency of stimulation. Here, we presented rapidly repeating targets of varying contrast and direction, as well as perturbing facilitatory and selective pathways by evoking localised adaptation. Using extracellular electrophysiology, we show that higher target contrast elicits stronger adaptation; however, the time course of diminishing responses was similar across contrasts. This relationship between target saliency and adaptation is also observed with targets moving in the preferred direction causing stronger adaptation. Targets that jitter, as is likely the case during rapid head foveation of prey, induce less adaptation as the jitter radius increases. We observed that the strengthening effect of facilitation competes against suppressive adaptation. Moreover, we observed that STMDs are unlikely to selectively attend to targets that traverse an adapted location. These results provide insight into how dragonfly target-detecting neurons adapt their responsiveness to visual targets in their environment that would be encountered during flight when pursuing prey or conspecifics.

Human vision must adapt to varying lighting conditions, particularly in low-light environments where depth perception and size constancy are challenged. This study examined size-distance scaling for real objects and afterimages under sudden and adapted darkness conditions. Real objects exhibited more stable scaling than afterimages, closely aligning with Emmert's law, while afterimages showed substantial deviations. Although dark adaptation modestly improved size-distance scaling for both stimulus types, it did not fully restore veridical perception. Contrary to expectations, afterimages did not benefit more from adaptation than real objects, suggesting that their reliance on external depth cues exceeds what mesopic adaptation alone can compensate for. Additionally, perceived brightness influenced distance misperceptions, with brighter stimuli associated with greater errors-particularly for real objects. These findings highlight the limitations of internally generated stimuli in maintaining perceptual stability in darkness and emphasise the critical role of environmental and stimulus-based depth cues in supporting size constancy.

Adaptive psychophysical procedures aim to increase the efficiency and reliability of measurements. With increasing stimulus and experiment complexity in the last decade, estimating multi-dimensional psychometric functions has become a challenging task for adaptive procedures. If the experimenter has limited information about the underlying psychometric function, it is not possible to use parametric techniques developed for the multi-dimensional stimulus space. Although there are non-parametric approaches that use Gaussian process methods and specific hand-crafted acquisition functions, their performance is sensitive to proper selection of the kernel function, which is not always straightforward. In this work, we use a neural network as the psychometric function estimator and introduce a novel acquisition function for stimulus selection. We thoroughly benchmark our technique both using simulations and by conducting psychovisual experiments under realistic conditions. We show that our method outperforms the state of the art without the need to select a kernel function and significantly reduces the experiment duration.