Motion-induced blindness as a noisy excitable system.

Perceptual disappearance of a salient target induced by a moving texture mask (MIB: Motion-Induced Blindness) is a striking effect, currently poorly understood. Here, we investigated whether the dynamics of MIB qualify as an excitable system. Excitable systems exhibit fast switches from one state to another (e.g., visible/invisible) induced by an above-threshold perturbation and stimulus-independent dynamics, followed by a refractory period. In the experiments, disappearance was induced by masks consisting of slowly rotating radial bars with a gap at the target location, leading to periodic perturbation of the visual field around the target (a bright parafoveal spot). When passed around the target location, masks frequently induced an abrupt target disappearance, pointing to locality. As expected from excitable systems, the disappearance time was not affected by additional bars crossing the target during invisibility, and there was little dependence on the mask configuration. After the target reappeared, it stayed for at least 0.5-2 s (the refractory period). Therefore, the dynamics governing MIB represent an example of an excitable system, where the transition to the invisible state is induced by the mask. The dynamics that follow were determined mostly by the internal network properties.

Isolating objective and subjective filling-in using the drift diffusion model.

Spatial context is known to influence the behavioral sensitivity (d') and the decision criterion (c) when detecting low-contrast targets. Of interest here is the effect on the decision criterion. Polat and Sagi (2007) demonstrated that, for a Gabor target positioned between two similar co-aligned high-contrast flankers, the observers' reports of seeing the target (Hit and False Alarm) decreased with increasing target-flanker distance. This effect was more pronounced when the distance was randomized within testing blocks compared to when it was fixed. According to signal detection theory (SDT), the latter result suggests that the decision criterion is adjusted to a specific distance-dependent combination of signal (S) and noise (N) when the S and N statistics are fixed, but not when they vary across trials. However, SDT cannot differentiate between changes in the decision bias (the criterion shift) and changes introduced by variations in S and N (the signal and noise shift). To circumvent this limitation of SDT, we analyzed the reaction time (RT) data within the framework of the drift diffusion model (DDM). We performed an RT analysis of the target-flanker interactions using data from Polat and Sagi (2007) and Zomet et al. (2008; 2016). The analysis revealed a stronger dependence on flankers for faster RTs and a weaker dependence for slower RTs. The results can be explained by DDM, where an evidence accumulation process depends on the flankers via a change in the rate of the evidence (signal and noise shift) and on observers' prior knowledge via a change in the starting point (criterion shift), leading to RT-independent and RT-dependent effects, respectively. The RT-independent distance-dependent response bias is attributed to the observers' inability to learn multiple internal distributions required to accommodate the distance-dependent effects of the flankers on both the signal and noise.

Traces of early developmental bias in the adult brain.

During the first 2 years of life, there is a high prevalence of optical distortions in the human eye, causing vertical blur on the retina (astigmatism), which is naturally resolved by the age of 5; thus, it is not treated. Here we determined the possible long-term effects on visual grouping resulting from optical distortions during the development of visual perception. Our results show a clear directional bias in shape perception for optically corrected astigmatic adults, compared with non-astigmatic ones, with remarkably slow decision times. These effects can be explained by a mismatch between the developmental timescales of different components in the visual system.

Contrast adaptation improves spatial integration.

The effects of contrast adaptation and contrast area summation (spatial integration) were investigated using a contrast discrimination task. The task consisted of a target of variable size, and a pedestal with a fixed base contrast. Discrimination performance was examined for a condition in which the pedestal size was fixed, equal to the largest target size, and for a condition in which the pedestal size matched the target size and thus varied with it. Repeated performance of the task produced rapid within-session improvements for both conditions. For stimuli with a matching size of target and pedestal, the performance improved only for the larger targets, indicating the development of spatial integration, which was initially absent for these stimuli. However, the improvements were mostly temporary, and were not fully retained between subsequent daily sessions. The temporary nature of the sensitivity gains implies that they resulted, at least in part, from rapid adaptation to the stimulus contrast. We suggest that adaptation decorrelates and thus reduces the spatial noise generated by a high-contrast pedestal, leading to improved spatial integration (area summation) and better contrast sensitivity. A decorrelation model successfully predicted our experimental results.

A decision-time account of individual variability in context-dependent orientation estimation.

Following exposure to an oriented stimulus, the perceived orientation is slightly shifted, a phenomenon termed the tilt aftereffect (TAE). This estimation bias, as well as other context-dependent biases, is speculated to reflect statistical mechanisms of inference that optimize visual processing. Importantly, although measured biases are extremely robust in the population, the magnitude of individual bias can be extremely variable. For example, measuring different individuals may result in TAE magnitudes that differ by a factor of 5. Such findings appear to challenge the accounts of bias in terms of learned statistics: is inference so different across individuals? Here, we found that a strong correlation exists between reaction time and TAE, with slower individuals having much less TAE. In the tilt illusion, the spatial analogue of the TAE, we found a similar, though weaker, correlation. These findings can be explained by a theory predicting that bias, caused by a change in the initial conditions of evidence accumulation (e.g., priors), decreases with decision time (*Communications Biology 3 (2020) 1-12). We contend that the context-dependence of visual processing is more homogeneous in the population than was previously thought, with the measured variability of perceptual bias explained, at least in part, by the flexibility of decision-making. Homogeneity in processing might reflect the similarity of the learned statistics.

Interaction of contexts in context-dependent orientation estimation.

The processing of a visual stimulus is known to be influenced by the statistics in recent visual history and by the stimulus' visual surround. Such contextual influences lead to perceptually salient phenomena, such as the tilt aftereffect and the tilt illusion. Despite much research on the influence of an isolated context, it is not clear how multiple, possibly competing sources of contextual influence interact. Here, using psychophysical methods, we compared the combined influence of multiple contexts to the sum of the isolated context influences. The results showed large deviations from linear additivity for adjacent or overlapping contexts, and remarkably, clear additivity when the contexts were sufficiently separated. Specifically, for adjacent or overlapping contexts, the combined effect was often lower than the sum of the isolated component effects (sub-additivity), or was more influenced by one component than another (selection). For contexts that were separated in time (600 ms), the combined effect measured the exact sum of the isolated component effects (in degrees of bias). Overall, the results imply an initial compressive transformation during visual processing, followed by selection between the processed parts.

Perceptual bias is reduced with longer reaction times during visual discrimination.

Fast and slow decisions exhibit distinct behavioral properties, such as the presence of decision bias in faster but not slower responses. This dichotomy is currently explained by assuming that distinct cognitive processes map to separate brain mechanisms. Here, we suggest an alternative single-process account based on the stochastic properties of decision processes. Our experimental results show perceptual biases in a variety of tasks (specifically: learned priors, tilt aftereffect, and tilt illusion) that are much reduced with increasing reaction time. To account for this, we consider a simple yet general explanation: prior and noisy decision-related evidence are integrated serially, with evidence and noise accumulating over time (as in the standard drift diffusion model). With time, owing to noise accumulation, the prior effect is predicted to diminish. This illustrates that a clear behavioral separation-presence vs. absence of bias-may reflect a simple stochastic mechanism.

Orientation-selective adaptation improves perceptual grouping.

The role of visual pattern adaptation, and learning, in spatial integration was investigated. Observers reported whether a grid of identical tilted bars was perceived as rows or columns (perceptual grouping task). Performance was measured multiple times during a session to determine effects of repeated exposure to the stimuli. To test for possible effects of learning on the within-session dynamics, observers repeated the experiment on five days. We found that repeated performance produced rapid within-day improvements, which were largely transient and were not retained on subsequent days. In addition, exposure to stimuli with equal orientation contributed to the within-session improvement, whereas stimuli having an orientation differing by 45° from the original orientation diminished the improvement previously obtained in the same session. Practice with the task over days resulted in faster improvements. The transient nature of these exposure-driven improvements and their susceptibility to interference by stimuli designed to reduce adaptation suggest that adaptation was their main cause. Finally, to investigate the effects of adaptation on internal noise and on spatial integration, we employed an external-noise paradigm, showing that internal-noise reduction resulted from adaptation. Internal noise was reduced only when spatial integration was effective, suggesting that adaptation improved perception of global stimulus properties. Overall, our results suggest that the grouping task benefits from an adaptation process that rapidly adjusts the visual system to the statistics of the visual stimuli. We suggest that this effect is achieved through spatial decorrelation of neural responses. With practice, those adjustments are made faster.

Real-time visual interactions across the boundary of awareness.

Recently, using Motion Induced-Blindness (MIB), we have shown that two visual stimuli, one consciously experienced and one not, interact as a function of feature and object similarity, pointing to preserved visual representations of objects, and their constitutive features, in the absence of perceptual awareness. Here we investigated whether these representations preserve the memory of the previously perceived stimulus by testing interactions with the unperceived stimulus modified while it is invisible. Observers performed the MIB task, wherein an object 'Target' (a plaid object) was morphed into one of its features (an oriented Gabor patch) once its disappearance was reported. Reappearances of the morphed target were induced by a visible 'Cue' (object or feature), with reappearance frequency used to quantify the interaction between the visible cue and the invisible target. Reappearance rates were highest when the morphed target and the cue shared the same orientations, with the plaid-cue showing reappearance rates equal to that of the orthogonal-cue. Our findings indicate that target-cue interactions do not depend on memory-stored representations, but rather, on the current state of the consciously unavailable target. We suggest that visual objects can be constructed and deconstructed in the absence of conscious perception, but only objects are consciously available.

Visual learning with reduced adaptation is eccentricity-specific.

Visual learning is known to be specific to the trained target location, showing little transfer to untrained locations. Recently, learning was shown to transfer across equal-eccentricity retinal-locations when sensory adaptation due to repetitive stimulation was minimized. It was suggested that learning transfers to previously untrained locations when the learned representation is location invariant, with sensory adaptation introducing location-dependent representations, thus preventing transfer. Spatial invariance may also fail when the trained and tested locations are at different distance from the center of gaze (different retinal eccentricities), due to differences in the corresponding low-level cortical representations (e.g. allocated cortical area decreases with eccentricity). Thus, if learning improves performance by better classifying target-dependent early visual representations, generalization is predicted to fail when locations of different retinal eccentricities are trained and tested in the absence sensory adaptation. Here, using the texture discrimination task, we show specificity of learning across different retinal eccentricities (4-8°) using reduced adaptation training. The existence of generalization across equal-eccentricity locations but not across different eccentricities demonstrates that learning accesses visual representations preceding location independent representations, with specificity of learning explained by inhomogeneous sensory representation.

Expectations and visual aftereffects.

The tilt aftereffect (TAE) is traditionally regarded as a consequence of orientation-selective sensory adaptation, a low-level stimulus-driven process. Adaptation has been recently suggested to be the outcome of predictive coding. Here, we tested whether the TAE is modulated by predictability, and specifically, whether TAE depends on the congruency of adapted and expected orientations. Observers were presented with successive pairs of oriented Gabor patches. Pairs were arranged in blocks, forming two conditions with the orientation of the second pair member either predictable or not. For all pairs, the orientation of the first Gabor was tilted clockwise (CW) or counterclockwise (CCW) (±20° relative to vertical, randomized). In the "Expected" conditions, the orientation of the second Gabor was fixed relative to the first Gabor (the same or a mirror orientation, blocked). In the "no-expectation" condition, the orientation of the second Gabor was independent of the first Gabor (randomized ±20°). Intermixed test pairs were used to measure observers' perceived vertical, with the second pair member serving as a target, oriented around the vertical, permitting an estimate of the TAE produced by the presentation of the first Gabor. Results show an increase in TAE with the expected orientation matching the inducing orientation, but a decrease with the expected mirror orientation, consistent with additivity of the adaptation and the expectation effects. A second experiment, with the first oriented Gabor replaced by a colored circular blob, showed that expectation alone does not modulate the perceived orientation. These findings indicate a role for expectation in generating the perceptual TAE and are in line with predictive coding models of perception. We suggest that orientation dependent adaptation is affected by both the mean orientation (first order statistics) and by temporal contingencies (second order statistics).

A dissociation between consolidated perceptual learning and sensory adaptation in vision.

Perceptual learning refers to improvement in perception thresholds with practice, however, extended training sessions show reduced performance during training, interfering with learning. These effects were taken to indicate a tight link between sensory adaptation and learning. Here we show a dissociation between adaptation and consolidated learning. Participants trained with a texture discrimination task, in which visual processing time is limited by a temporal target-to-mask window defined as the Stimulus-Onset-Asynchrony (SOA). An initial training phase, previously shown to produce efficient learning, was followed by training structures with varying numbers of SOAs. Largest interference with learning was found in structures containing the largest SOA density, when SOA was gradually decreased. When SOAs were largely kept unchanged, learning was effective. All training structures yielded the same within-session performance reduction, as expected from sensory adaptation. The results point to a dissociation between within-day effects, which depend on the number of trials per se regardless of their temporal structure, and consolidation effects observed on the following day, which were mediated by the temporal structure of practice. These results add a new dimension to consolidation in perceptual learning, suggesting that the degree of its effectiveness depends on variations in temporal properties of the visual stimuli.

Asymmetric visual interactions across the boundary of awareness.

A salient visual object can disappear from conscious perception when surrounded by a moving texture, a phenomenon known as MIB, Motion-Induced Blindness (Bonneh, Cooperman, & Sagi, 2001). Here we tested the information available in the brain from such stimuli that do not access awareness by examining interactions across the boundary of awareness between stimuli that reach awareness and those that do not. Observers performed the MIB task in which a "Cue" was presented next to the "Target" after observers reported the perceptual disappearance of the target (Kawabe, Yamada, & Miura, 2007). Oriented Gabor patches were used as targets and cues; observers reported the target's reappearance. The results indicated an interaction between the target and the cue, depending on the orientation difference (∼30° bandwidth) and distance (∼1° range), indicating preserved properties of features in the absence of awareness. Object-based representation (binding) of unseen stimuli was tested by examining the interaction between a compound stimulus and its composing features. Here we used vertical and horizontal Gabor patches and their combinations (plaids) as targets and cues. Results indicated asymmetric relations between aware and unaware object representations; a plaid cue was not effective with a component target, but a plaid target efficiently reappeared by its component cues. This result suggests that the unseen, but not the seen plaid, is decomposed into its features. Plaid targets also reappeared with plaid cues, supporting binding without awareness. Our findings suggest preconscious representations of objects and their features, with conscious perception confined to object representations.

Target-selective tilt aftereffect during texture learning.

Sensory adaptation and perceptual learning are two forms of plasticity in the visual system, with some potential overlapping neural mechanisms and functional benefits. However, they have been largely considered in isolation. Here we examined whether extensive perceptual training with oriented textures (texture discrimination task, TDT) induces adaptation tilt aftereffects (TAE). Texture elements were oriented lines at -22.5° (target) and 22.5° (background). Observers were trained in 5 daily sessions on the TDT, with 800-1000trials/session. Thresholds increased within the daily sessions, showing within-session performance deterioration, but decreased between days, showing learning. To evaluate TAE, perceived vertical (0°) was measured prior to and after each daily session using a single line element. The results showed a TAE of ∼1.5° at retinal locations consistently stimulated by the target, but none at locations consistently stimulated by the background texture. Retinal locations equally stimulated by target and background elements showed a significant TAE (∼0.7°), in a direction expected by target-driven sensory adaptation. Moreover, these locations showed increasing TAE persistence with training. Additional experiments with a modified target, in order to have balanced stimulation around the vertical direction in all target locations, confirmed the locality of the task-dependent TAE. The present results support a strong link between perceptual learning and local orientation-selective adaptation leading to TAE; the latter was shown here to be task and experience dependent.

Tilt aftereffect due to adaptation to natural stimuli.

The human visual system continuously adjusts to the current environment. To investigate these adjustments, biases in observers' perceptions owing to changes in the visual environment are measured (visual aftereffects). Typically, the stimuli used are synthetic and are composed of oriented patterns such as lines or gratings. These patterns are known to activate individual neurons in the visual cortex, but cover only a small subset of actual visual stimulations. To overcome this drawback, recent research has focused on synthetic patterns that mimic several aspects of natural stimulation. However, the aftereffects of natural stimulation per-se remain largely unexplored. Here, we interleaved presentations of unmodified natural image adaptors, selected according to criteria favoring content at a particular orientation, with presentations of targets that test a perceived orientation. This allowed us to measure the change in the perceived orientation, namely the tilt aftereffect (TAE), which resulted from repeated image presentations. Results show a close to standard TAE with adaptor durations around 500ms, which is reduced with longer presentations. Importantly, our method can be generalized to investigate other aftereffects by selecting images differently.

Binocular summation of chance decisions.

Seeing with two eyes usually helps one respond faster. Here we show that with ambiguous stimuli, binocular viewing can paradoxically slow down reaction time. This is explained by the observers basing their decision on a noisy neuronal representation within the visual system, with the added noise breaking the symmetry between the two possible interpretations. Binocular integration improves the representation by reducing the noise, increasing ambiguity, and decision time. The neuronal Accumulator (Race) model is applied to quantify the underlying binocular integration. The model accounts for the distributions of reaction times, and predicts suboptimal integration between eyes. We conclude that under ambiguous stimulation neuronal noise within the visual system determines responses.

Perceptual learning in autism: over-specificity and possible remedies.

Inflexible behavior is a core characteristic of autism spectrum disorder (ASD), but its underlying cause is unknown. Using a perceptual learning protocol, we observed initially efficient learning in ASD that was followed by anomalously poor learning when the location of the target was changed (over-specificity). Reducing stimulus repetition eliminated over-specificity. Our results indicate that inflexible behavior may be evident ubiquitously in ASD, even in sensory learning, but can be circumvented by specifically designed stimulation protocols.