In virtual reality, observers readily calibrate to the bodies of digital human characters and consider changes to these in estimates of their action capabilities. This study investigates whether such calibration persists when somatosensory and visual information specifying the spatial location of limbs are incongruent. Across four experiments, virtual hand size and location was manipulated systematically while participants were placed in an immersive environment. Participants performed a calibration exercise to gain experience with the virtual hand before estimating their perceived grasping and reaching capabilities. In all experiments, observers consistently calibrated to the virtual hands and considered this in their perceived grasping and reaching estimates. This demonstrates that the properties of an artificial limb are integrated into observers' perception of their action capabilities, even when conditions for embodiment are not optimal.
It is critical to provide real-time environmental information to ensure takeover safety during automated driving. However, prior research has largely neglected the role of peripheral situational awareness (SA), which is essential for safe takeover. This study investigates how drivers develop SA in different takeover scenarios through a driving simulator experiment with three peripheral visual displays: baseline, interface light and ambient light. Participants were required to takeover and avoid potential hazards after receiving a takeover request. The results indicated that ambient light and interface light displays improve takeover performance, increase drivers' visual attention to rearview mirrors and enhance overall SA. The ambient light displays improved drivers' SA in higher-urgency scenarios. These findings suggest that ambient light effectively guides drivers' attention to hazards visible in the rearview mirror and improve their SA. This study provides valuable insights for the design of peripheral displays in automated vehicle and helps improve human-automation safety.
Visual attention is guided by attentional templates that store task-relevant information, but whether these templates contain sensory details, and how task factors influence their content and guiding strength remains unclear. We investigated these questions by combining sustained feature-based attention with a search task containing an irrelevant distractor that either matched or mismatched the target color in the sustained attention task. The difference in attentional capture by the distractor in the same-color versus different-color condition indexed the strength of the sensory template. We manipulated selection demand by varying the similarity between the two colors in the sustained attention task and presented cues in either sensory or non-sensory formats. With sensory cues, capture magnitude remained equally strong under both low and high selection demand. However, with non-sensory cues, low selection demand attenuated but did not fully eliminate capture. These results demonstrate that attentional templates store sensory content of cued features. The guiding strength of templates is not susceptible to selection demand with sensory cues, but non-sensory cues delayed or reduced the activation and maintenance of sensory templates under low selection demand. This suggests that cue format and selection demand contribute to how sensory information is represented in attentional templates to guide selection.
It is well established that individuals recognize faces of their own age better compared to others, known as the own-age bias. However, its underlying mechanisms, especially the roles of holistic and part-based information, remain underexplored. In a variable viewing position paradigm. participants fixated on either the left or right eye of a target face and then chose the previously seen face from four options: the target, a face with congruent peripheral but incongruent central features (holistic match), a face with congruent central but incongruent peripheral features (part-based match), and a fully incongruent face. Undergraduate participants were tested with upright (Experiment 1) and inverted (Experiment 2) faces of young adults (20-year-olds) and children (4-year-olds). Results showed that for upright faces, participants achieved higher accuracy for own-age faces versus other-age faces, demonstrating the own-age bias, and this bias was not observed for inverted faces. Additionally, for upright faces, participants made more holistic errors on adult faces than child faces, but more part-based errors on child faces than adult faces. These error patterns disappeared for inverted faces. The findings suggest that the own-age bias results from a nuanced adjustment in face processing that involves both holistic and part-based information.
It is widely agreed that attentional selection can be guided by goal-related control settings. While many studies have shown that such guidance operates for targets defined by simple perceptual features, it remains unclear whether control settings defined by abstract concepts can also guide spatial selection, or whether apparent abstract guidance is mediated by concrete perceptual templates. We addressed this question in two experiments using a contingent-capture spatial-cueing paradigm. Participants searched for a target defined by relative magnitude (the digit with the smallest or largest value in a display), such that the target could not be specified in advance by a fixed perceptual feature. In Experiment 1, spatial-cueing effects were larger when a physical-size-singleton cue (small or large) matched the abstract target definition than when it did not. Experiment 2 ruled out an alternative account, according to which both cue types captured attention due to their saliency, such that the cue-goal match did not affect guidance, but instead delayed disengagement from matching relative to non-matching cues. Together, the results show that attentional control settings defined by an abstract concept can bias spatial selection when reliance on a fixed perceptual target template is not possible.
Previous studies showed that eye contact alters time perception. However, inconsistent findings have been reported regarding whether eye contact leads to an underestimation or overestimation of time. This study investigated how the race of facial stimuli influences time perception during eye contact among Japanese university students. In Experiment 1, participants completed the time bisection task using static images of Japanese and Caucasian faces with either direct or averted gaze. The results showed no significant effect of gaze direction but significant effect of race: participants perceived the duration of Japanese faces as shorter. In Experiment 2, we created a dynamic eye contact situation by dynamically presenting sequential face images with different gaze directions. The results revealed that participants were more likely to perceive time as longer when the gaze was directed toward them, indicating a clear eye-contact effect. These findings suggest that the method of facial image presentation influences time perception.
Most studies of visuospatial short-term memory (VSSTM) rely on simplified tasks performed on vertical screens. In this study, we examine VSSTM in a more naturalistic three-dimensional context, using goal-directed throwing toward stimuli projected on the floor. Across four experiments, participants memorized spatial locations under different demands involving serial order, item-location binding, and memory load. We developed a method to dissociate spatial localization errors from transposition errors, allowing their independent contribution to performance to be quantified. When both order and location of sequentially presented targets had to be recalled, throwing accuracy followed an inverted U-shaped serial position curve, largely driven by order-related errors. Reducing spatial uncertainty during recall selectively decreased transposition errors, revealing an interaction between spatial and sequential processes. Removing serial requirements markedly reduced spatial error and improved performance for recent items. When stimuli were presented simultaneously and memory load was varied, both spatial and transposition errors increased with load, diverging from each other from three items onward. Together, these results provide behavioral evidence for multiple, separable sources of error in VSSTM and introduce a methodological framework to decompose visuomotor performance into distinct cognitive components, consistent with probabilistic models of working memory.
Accurate perception of time and space is crucial for real-time interactions with our environment. This process demands flexibility as it combines information from our actions and the outside context. Probing the visual system during the updating process reveals spatiotemporal distortions, where sequential stimuli appear closer in time and space than they actually are. These effects occur perisaccadically or when a visual mask follows the stimuli. The study examined whether non-overlapping visual masks could influence temporal order judgments (TOJs), suggesting that a temporal signal might act as an anchor during updating. In Experiment 1, participants judged the temporal order of two stimuli under three conditions: no mask, a full-field mask, or a partial mask that avoided the stimuli's locations. Compared to no mask, both masks triggered reversals in TOJs when presented within 30 milliseconds of the second stimulus. In two control experiments, delaying mask onset by 30 milliseconds eliminated the inversion effect. In Experiment 2, reversals in TOJs were observed for both ipsilateral and contralateral masks, suggesting that long-range inhibitory signals might also contribute to the effect. Together, these findings indicate that temporal inversions can occur with a non-overlapping stimuli masks configuration, pointing to a non-spatial signal related to mask timing as the underlying mechanism.
The transposed-word effect, where readers misperceive transposed-word sentences as grammatical, has been central to the debate between parallel and serial processing accounts of reading. While parallel processing accounts attribute this effect to simultaneous word activation, evidence from serial presentations challenges this assumption. In this study, sixty-six native Chinese readers completed a grammaticality judgment task on Chinese sentences under three presentation modes: parallel, horizontal serial, and central serial, with eye tracking used to control for reading time and spatial information. The results showed a robust transposed-word effect in all presentation modes, with larger effect sizes for error rates in parallel compared to both serial presentations (central and horizontal), and for response times in parallel and horizontal serial compared to central serial. These findings provide further support for serial accounts of lexical processing in which word-order encoding remains flexible and highlight the influence of spatial position and word availability on this process.
Rectangularity can be an a priori constraint for the visual system to recover a 3D shape from a 2D image. Two psychophysical experiments (n = 16 each) were conducted to test whether observers are biased to perceive trihedral corners as rectangular and whether this bias is perceptual rather than a response bias. The results of Experiment 1 showed that observers were biased to report that the 3D shapes of the corners were rectangular when their 2D images were consistent with rectangular interpretations. Experiment 2 tested whether this bias affects 3D shape perception by using a task about a 3D shape that was not directly concerned with rectangularity. The results of Experiment 2 were consistent with a prediction based on a hypothesis that the rectangularity bias is perceptual. These results suggest that the visual system uses rectangularity as an a priori constraint for 3D shape perception.
Wearing glasses can alter various facial appearances, potentially interfering with face processing during person identification. Although prior research has examined their effects on face recognition and matching, much less is known about how wearing glasses influences the perceptual detection of changes in face identity. This study examined how wearing glasses influences face identity perception. In Experiment 1, 45 participants judged whether two sequentially presented faces were the same individual. The results indicated that putting on and taking off glasses had opposite effects on face identity change detection relative to not wearing glasses: faces were more likely to be perceived as the same identity when glasses were taken off between successive faces, whereas a greater degree of facial similarity was required for two faces to be judged as the same identity when glasses were put on. Experiment 2 employed the same task but shifted the glasses upward. Under this manipulation, the effects observed in Experiment 1 were very weak or eliminated. Collectively, these findings demonstrate opposite effects of putting on versus taking off glasses on face identity perception, which are most pronounced when glasses are properly worn.
Attentional capture is strongest when cues exactly match a target feature, but recent evidence suggests linguistic categories can also guide attention. Cues that share a category label with the target can capture attention even when their hue differs, yet it remains unclear whether this categorical guidance extends to fine-grained, confusable colours near category boundaries. We used a spatial cueing task to test whether attention is biased by colour categories. In Experiment 1, singleton cues and targets spanned the blue - green boundary. In Experiment 2, cues and targets were restricted to the centroid of a single category (blue or green) to control for ensemble effects. In Experiment 1, validity effects were stronger when cues and targets belonged to the same colour category, even near the category boundary. Model comparisons revealed that a stepwise, categorical function at the boundary fit the data better than a linear gradient. In Experiment 2, with targets within a single category, only a baseline validity effect emerged, with no categorical modulation. These results show that categorical structure for blue-green colours shapes attention: attention is flexibly tuned to colour categories rather than limited to exact feature values, enabling efficient selection of relevant stimuli.
Internalizing disorders, like depression and anxiety, reportedly disrupt facial emotion expression recognition. Here, we evaluated whether internalizing symptoms alter expression perception, allowing us to determine whether visual, temporally early, non-linguistic components of expression processing are altered as a function of symptoms. In a within-subjects design, we tested perceptual sensitivity to expressions that varied as a function of motivational signaling (approach, avoid), complexity (basic, complex), and valence (positive, negative). Participants included 123 emerging adults tested online during the COVID-19 pandemic. Participants completed a just noticeable difference perceptual staircase task to determine the minimal amount of perceptual information needed to differentiate an expression from a neutral face. They also reported depression and anxiety symptoms. Participants were more sensitive to perceiving approach versus avoid expressions, but their sensitivity to complexity and valence was not systematic. Depression symptoms failed to predict expression perception suggesting that early, non-linguistic components of expression perception are intact. Anxiety symptoms predicted reduced sensitivity to perceive anger and fear expressions, suggesting altered early, non-linguistic processing of these expressions. These findings have implications for predicting and interpreting different neural systems that affect emotion expression processing and therapeutic intervention approaches in depression and anxiety.
When we interact with each other, we frequently see faces from different viewpoints. Changes in viewpoint can result in significant variations in the way emotional facial expressions appear. Although the visibility of features is not qualitatively affected between front and three-quarter views, in profiles some features are hidden, and actions such as mouth widening and eyebrow narrowing, which contain important horizontally-oriented information, are partly hidden. To evaluate how profiles uniquely affect emotional facial expression processing, we recorded response times and error rates in a two-choice discrimination task involving two emotions (happiness and anger) from five different viewpoints (front 0 degrees, left and right 45 degrees three-quarter, left and right 90 degrees profile) in normal and inverted presentations, with 99 young adult participants. We found higher response times and error rates for profiles than for other views, and the inversion effect was significant only for profiles. There was no difference between the three-quarter and front views. In normal presentations, differences in viewpoint were more obvious with response times than with error rates. We also found an asymmetry with slower response times for the right than for the left profiles. Our results support the idea that emotional facial expressions are processed in a view-dependent manner.
This study investigated the effects of display size and aging on heading-speed perception from the optic flow observed during forward locomotion. In the experiment, younger and older participants viewed two types of 5-sec video clips: standard and test stimuli. Standard stimuli were extracted from video clips recorded during forward locomotion at the original playback speed, whereas the test stimuli were extracted from video clips with different playback speeds and display sizes (the peripheral area of optic flow was systematically cropped). Participants rated the perceived heading speed of the test stimulus relative to that of the standard stimulus. Across both age groups, ratings of the perceived heading speed increased with an increase in the visible area. However, the dependence on the peripheral area of the stimulus was weaker among older participants. These results highlight the importance of the peripheral area of optic flow in heading speed perception and indicate that older adults rely more on the central area of optic flow, whereas younger adults use both central and peripheral areas of optic flow.
Interactive search refers to tasks wherein a searcher actively manipulates objects to find a target. Here, we examined how physical effort influences behaviour and performance during an interactive search task. Participants searched in person for a "T" shape embedded on the side of a set of cubes that varied in weight. Interactive search behaviours were recorded throughout the experiment using an optical motion tracking system. In all cases, we found no evidence that any discernible effects of effort emerged. Despite this, we offer three important contributions to the literature. First, our methodology enabled us to determine that, within our study, interactive search utilized a "nearest-next" strategy, with searchers choosing to interact with the nearest cube to them on each trial. Second, contrary to expectations from purely visual search tasks, rather than waiting and examining the already-visible cube faces at the start of each trial, searchers opted instead to begin interacting with objects immediately. Third, response accuracy rates were no different between target-present and target-absent trials, suggesting that there is at least one point at which interactive search differs from visual search. Our findings lay down the foundation for future and more detailed examinations of live, in-person interactive search tasks.
Humans experience a sense of uncanniness when encountering human-like robots, a phenomenon known as the uncanny valley. Although the underlying mechanism remains unclear, the categorization failure hypothesis suggests that the visual system automatically attempts - but fails - to categorize the perceived object, leading to a negative affective response. However, it remains uncertain whether this categorization process does indeed occur automatically. This study examined whether the uncanny valley effect is triggered automatically by a task-irrelevant dimension. Participants categorized the colour of a morphed figure that previously elicited the uncanny valley effect based on shape, and then rated its likeability. Despite successful colour-based categorization of the figure, participants exhibited the uncanny valley effect based on the task-irrelevant shape dimension. These findings imply that the visual system processes an object's shape automatically, and incorporates this information into affective evaluation, supporting the categorization failure hypothesis.
The present study examined the role of vowelization diacritics in the visual recognition of written Arabic verbs among typical (N = 20) and poor (N = 20) native Arab readers in the 10th grade. A lexical decision task was used, in which stimuli were presented under two vowelization conditions: vowelized and non-vowelized. The findings revealed that typical readers demonstrated higher accuracy and shorter response times in the non-vowelized condition. In contrast, the effect of vowelization on the performance of poor readers was inconsistent across measures. The results are discussed in light of previous findings in the literature, with particular emphasis on the contribution of morpho-orthographic components to the visual recognition of Arabic verbs.
