Search Task Research Articles

A Unified Prompt-aware Framework for Personalized Search and Explanation Generation

Product search is crucial for users to find and purchase products they need. Personalized product search, which models users’ search intent and provides tailored results, has become a prominent research problem in industry and academia. Recent studies often leverage knowledge graphs (KGs) to improve search performance and generate explanations for search results. However, existing KG-based methods treat search and explanation tasks separately and explore paths in KGs as explanations, creating a gap between search results and generated explanations. Also, path-formed explanations in KGs are not flexible enough to build correlations with the user’s current query. To address these challenges, we propose P-PEG, a unified prompt-aware framework for personalized product search and explanation generation. P-PEG leverages a pre-trained language model (PLM) and search signal to enhance the generation of user-understandable explanations. We introduce a prompt learning technique and design prompt generators for search and explanation generation tasks based on a fixed PLM. By incorporating search results in explanation-based prompts, we bridge the gap between search results and explanations, facilitating better interaction. Additionally, we utilize the user’s current query, historical search log, and KGs to personalize the explanations and inject task knowledge into PLM. Experimental results show that P-PEG outperforms existing methods in the explanation generation task of the three datasets and the search task of the Electronics dataset, and achieves comparable performance in the search task of the Cellphones & Accessories and CD & Vinyl datasets.

Distractor-response binding influences visual search.

Intertrial priming effects in visual search and action control suggest the involvement of binding and retrieval processes. However, the role of distractor-response binding (DRB) in visual search has been largely overlooked, and the specific processing stage within the functional architecture of attentional guidance where the DRB occurs remains unclear. To address these gaps, we implemented two search tasks, where participants responded based on a separate feature from the one defining the target. We kept the target dimension consistent across trials while varying the color and shape of the distractor. Moreover, we either repeated or randomized the target position in different sessions. Our results revealed a pronounced response priming, a difference between trials where the response changed versus repeated, and importantly this response priming was stronger when distractor features or the target position were repeated than when they changed. These insights affirm the presence of DRB during visual search and support the framework of binding and retrieval in action control as a basis for observed intertrial priming effects related to distractor features. All data are available at: https://github.com/msenselab/distractor_binding .

Visual search for warm and cool colors

Warm versus cool colors are considered a fundamental aspect of color experience, yet the basis for this distinction remains poorly understood. Recently, we found that the warm–cool dimension closely aligns with sensitivity biases implicit in uniform color spaces [J. Vis. 23, 5572 (2023)1534-736210.1167/jov.23.9.5572], which predicts that color vision is less sensitive to the warm–cool axis than other chromatic axes. Here, we tested this prediction by measuring visual search for color targets on backgrounds defined by different axes in color space. The search task involved locating a circular chromatic target sampled from a range of chromaticities that was presented on a background of ellipses that randomly varied in color along the warm–cool (roughly orange–cyan) axis, a blue–yellow axis, or along magenta–greenish axes perpendicular to the warm–cool or blue–yellow axis. Search times were faster on both warm–cool and blue–yellow backgrounds than the orthogonal backgrounds. However, there was no difference between these two axes, or for warm versus cool hues or blue versus yellow hues. These results are consistent with weaker sensitivity for the warm–cool and blue–yellow axes of the color space, but do not reveal a salience difference between colors more strongly associated with objects and surfaces (warm and yellow) versus backgrounds and illumination (cool and blue).

Evaluating geometrically-approximated principal component analysis vs. classical eigenfaces: a quantitative study using image quality metrics

Principal component analysis (PCA) is essential for diminishing the number of dimensions across various fields, preserving data integrity while simplifying complexity. Eigenfaces, a notable application of PCA, illustrates the method's effectiveness in facial recognition. This paper introduces a novel PCA approximation technique based on maximizing distance and compares it with the traditional eigenfaces approach. We employ several image quality metrics including Euclidean distance, mean absolute error (MAE), peak signal-to-noise ratio (PSNR), signal-to-noise ratio (SNR), and structural similarity index measure (SSIM) for a quantitative assessment. Experiments conducted on the Brazilian FEI database reveal significant differences between the approximated and classical eigenfaces. Despite these differences, our approximation method demonstrates superior performance in retrieval and search tasks, offering faster and parallelizable implementation. The results underscore the practical advantages of our approach, particularly in scenarios requiring rapid processing and expansion capabilities.

EXPRESS: Eye on Context: Individual Differences Reveal the Mechanisms of Statistical Learning.

If a searched-for target object is consistently encountered within repeating spatial distractor arrangements, target detection becomes more efficient relative to non-repeated, i.e., random arrangements (contextual-cueing effect). On the other hand, target location changes within otherwise unchanged distractor arrays substantially weaken the cueing effect. Previous studies reported substantial variations in individual participants' abilities to learn and relearn invariant contexts. Therefore, the current study examined how individual differences in attentional control and focus, as indexed by the well-established Stroop and Navon tasks, respectively, relate to contextual cueing (CC) in a learning phase/relocation phase design. During the visual search, we recorded behavioral reaction times (RTs) and fixation locations, the latter permitting us to decompose search RTs into search- and motor-related substages. We could thus evaluate the processes responsible for CC and the lack thereof after target relocation while also testing whether search and motor components of CC are different for individuals depending on their Stroop/ Navon scores. Repeated contexts yielded faster RTs (and reduced fixation numbers), though there was a substantial decrease in cueing from learning to adaptation, consistent with previous studies. Critically, contextual learning, but not relearning, varied across individuals: participants with high-Stroop interference displayed overall larger CC during early target search, while a more local Navon task bias was associated with increased CC during later processes of target response decisions. Our results demonstrate that analyzing individual differences can help validate the processes responsible for CC in search tasks, particularly distinguishing between early search and later response-related mechanisms.

The effects of aerobic exercise on goal-directed attention and inhibitory control in individuals with high trait anxiety: an EEG study

Anxiety is known to significantly impair cognitive function, particularly attentional control. While exercise has been demonstrated to alleviate these cognitive deficits, the precise neural mechanisms underlying these effects remain poorly understood. This study examines the effects of exercise on attentional control in individuals with high trait anxiety, based on attentional control theory, which suggests that such individuals have reduced top-down attention. Thirty-eight participants were randomly assigned to either an exercise group or a reading group. The exercise group engaged in 30 min of moderate-intensity cycling, while the reading group spent 30 min reading quietly. In Experiment 1, goal-directed attention was assessed using a cue-object paradigm, and in Experiment 2, inhibitory control was evaluated through a visual search task. EEG data indicated that the exercise group exhibited significantly larger Pd components in both experiments, suggesting enhanced attentional focus and improved inhibition of distractors. These findings suggest that aerobic exercise enhances top-down attentional processes, particularly goal-directed attention and distractor inhibition, offering potential as an intervention for individuals with high trait anxiety.

Diverse neuronal activity patterns contribute to the control of distraction in the prefrontal and parietal cortex.

Goal-directed behavior requires the effective suppression of distractions to focus on the task at hand. Although experimental evidence suggests that brain areas in the prefrontal and parietal lobe contribute to the selection of task-relevant and the suppression of task-irrelevant stimuli, how conspicuous distractors are encoded and effectively ignored remains poorly understood. We recorded neuronal responses from 2 regions in the prefrontal and parietal cortex of macaques, the frontal eye fields (FEFs) and the lateral intraparietal (LIP) area, during a visual search task, in the presence and absence of a salient distractor. We found that in both areas, salient distractors are encoded by both response enhancement and suppression by distinct neuronal populations. In FEF, a larger proportion of units displayed suppression of responses to the salient distractor compared to LIP, with suppression effects in FEF being correlated with search time. Moreover, in FEF but not in LIP, the suppression for the salient distractor compared to non-salient distractors that shared the target color could not be accounted for by an enhancement of target features. These results reveal a distinct contribution of FEF in the suppression of salient distractors. Critically, we found that in both areas, the population level representations of the target and singleton locations were not orthogonal, suggesting a mechanism of interference from salient stimuli.

Fixation-related potentials during a virtual navigation task: The influence of image statistics on early cortical processing.

Historically, electrophysiological correlates of scene processing have been studied with experiments using static stimuli presented for discrete timescales where participants maintain a fixed eye position. Gaps remain in generalizing these findings to real-world conditions where eye movements are made to select new visual information and where the environment remains stable but changes with our position and orientation in space, driving dynamic visual stimulation. Co-recording of eye movements and electroencephalography (EEG) is an approach to leverage fixations as time-locking events in the EEG recording under free-viewing conditions to create fixation-related potentials (FRPs), providing a neural snapshot in which to study visual processing under naturalistic conditions. The current experiment aimed to explore the influence of low-level image statistics-specifically, luminance and a metric of spatial frequency (slope of theamplitude spectrum)-on the early visual components evoked from fixation onsets in a free-viewing visual search and navigation task using a virtual environment. This research combines FRPs with an optimized approach to remove ocular artifacts and deconvolution modeling to correct for overlapping neural activity inherent in any free-viewing paradigm. The results suggest that early visual components-namely, the lambda response and N1-of the FRPs are sensitive to luminance and spatial frequency around fixation, separate from modulation due to underlying differences in eye-movement characteristics. Together, our results demonstrate the utility of studying the influence of image statistics on FRPs using a deconvolution modeling approach to control for overlapping neural activity and oculomotor covariates.

Multimodal mapping of spatial attention for unilateral spatial neglect in VR: a proof of concept study using eye-tracking and mobile EEG

Unilateral spatial neglect (USN) is a complex spatial attentional disorder consisting of a failure to attend to the contralesional side of space, frequently seen after a stroke. However, the majority of cases go undiagnosed due to the lack of a valid and reliable tool that is able to assess USN and its many variants. Recent technological advances in virtual reality (VR) and physiological sensors, allow for the study of this disorder under controlled, and ecologically-valid environments, which hold the promise of reliable and early detection. This proof of concept study aims to evaluate the feasibility of a system for discriminating different attentional states using a multimodal dataset derived from a spatial attention task conducted in VR. Nine healthy young adults underwent two experimental conditions: a Control condition and a Left Occlusion condition. Participants performed a visual search task while their behavioral data, including performance metrics, eye-gaze, head, and controller movement data, were recorded. Additionally, electroencephalography data was synchroniously collected to capture neural correlates of attentional processing. Analysis of results of this within-subjects study found worse performance (higher RT), changes in behavior (right-ward gaze bias, left-ward bias in head and controller movement) in the Left Occlusion condition. Neural differences were found (parieto-occipital mean alpha band power and event related potentials) between the two conditions. If validated, this system could be utilized as a diagnostic VR tool, while it holds the potential to facilitate the participation of stroke patients with USN in VR-driven rehabilitation.

Self-Bias and Self-Related Mentalizing are Unaltered in Adolescents with Autism.

The self is a multidimensional concept that can be represented at a pre-reflective (first-order) level, at a deeper, reflective level (second-order), or even at a meta-level (representing one's own thoughts, i.e. self-related mentalizing). Since self-related processing and mentalizing are crucial for social cognition, both constructs have been researched in individuals with autism, who experience persistent socio-communicative difficulties. Some studies suggested autism-related reductions of the self-bias, i.e. tendency to preferentially process self-related content; while others observed a decreased ability to mentalize on one's own thoughts in autism. However, prior research examined distinct levels of self-related processing in isolation, in the context of separate studies. In this investigation, we directly compared self-bias, self- and other-related mentalizing within the same sample of adolescents with and without autism, to identify which of these are altered in this condition. Thirty adolescents with autism and 26 age- and IQ-matched controls performed a visual search task (first-order self-bias), a trait adjectives task (second-order self-bias), a feeling-of-knowing task (self-related mentalizing) and the Frith-Happé animations task (other-related mentalizing). Parents also completed two questionnaires (i.e. SRS, SCQ) assessing the adolescent's degree of autism traits. Our findings replicated previous research showing reduced other-related mentalizing in autism. However, we did not find any difference between adolescents with and without autism in terms of first- or second-order self-bias, nor in the ability to mentalize on one's own thoughts. In line with recent investigations, our results do not support earlier claims of altered self-related information processing in autism.

Action of the Euclidean versus projective group on an agent’s internal space in curiosity driven exploration

According to the Projective Consciousness Model (PCM), in human spatial awareness, 3-dimensional projective geometry structures information integration and action planning through perspective taking within an internal representation space. The way different perspectives are related to and transform a world model defines a specific perception and imagination scheme. In mathematics, such a collection of transformations corresponds to a ‘group’, whose ‘actions’ characterize the geometry of a space. Imbuing world models with a group structure may capture different agents’ spatial awareness and affordance schemes. We used group action as a special class of policies for perspective-dependent control. We explored how such a geometric structure impacts agents’ behaviors, comparing how the Euclidean versus projective groups act on epistemic value in active inference, drive curiosity, and exploration. We formally demonstrate and simulate how the groups induce distinct behaviors in a simple search task. The projective group’s nonlinear magnification of information transformed epistemic value according to the choice of frame, generating behaviors of approach toward objects with uncertain locations due to limited sampling. The Euclidean group had no effect on epistemic value: no action was better than the initial idle state. In structuring a priori an agent’s internal representation, we show how geometry can play a key role in information integration and action planning. Our results add further support to the PCM.

The Impact of Color Combination on Visual Search Efficiency and User Experience in Human-Machine Interface: A Case Study of Metro Electronic Guide Screen

Color is an important attribute of human-perceived human-machine interface (HMI), and it is unclear which color combinations produce the best performance and user experience in visual search tasks. Taking the metro electronic guide screen (MEGS) as an example, the eye-tracking (E-T) device is used to record the pupil diameter and search time, and the Likert scale is used to evaluate the satisfaction and usability, and the influence of the four factors of background hue, saturation, brightness and text color in HMI on user search efficiency and user experience is discussed in depth. The results show that the recognition efficiency is best when the background hue is medium brightness, medium saturation green with white text color. Meanwhile, the results of user satisfaction and ease of use are basically consistent with the objective experimental results, and there is a certain correlation between the subjective preference of color combination and visual performance.

Enhancing Snippet Visualizations to Improve Web Search

Various snippet visualizations have been suggested to assist users in assessing the relevance of their online search results, including highlighting query terms and incorporating target page thumbnails. This study introduced two enhancements: shades and multiple thumbnails. In the shades visualization, query terms in the snippet are background highlighted in different shades based on the number of adjacent query terms. The multiple thumbnail visualization presents reduced-size thumbnails of the target landing page. A 3 × 3 experimental design examined the effects of thumbnails (no thumbnail, single thumbnail, multiple thumbnails) and highlighting (no highlighting, yellow background highlighting, shades), and their combined use on 37 participants’ effectiveness, efficiency, and satisfaction when performing complex search tasks. The results showed that although the shades and multiple thumbnail visualizations had no significant effect on users’ effectiveness compared to the baseline textual snippet visualization, they both significantly (p < 0.0001) improved users’ efficiency and satisfaction, and their combined effect was synergetic. The findings have operative implications for web search interface design.

Contributions of distractor dwelling, skipping, and revisiting to age differences in visual search.

Visual search becomes slower with aging, particularly when targets are difficult to discriminate from distractors. Multiple distractor rejection processes may contribute independently to slower search times: dwelling on, skipping of, and revisiting of distractors, measurable by eye-tracking. The present study investigated how age affects each of the distractor rejection processes, and how these contribute to the final search times in difficult (inefficient) visual search. In a sample of Dutch healthy adults (19-85years), we measured reaction times and eye-movements during a target present/absent visual search task, with varying target-distractor similarity and visual set size. We found that older age was associated with longer dwelling and more revisiting of distractors, while skipping was unaffected by age. This suggests that increased processing time and reduced visuo-spatial memory for visited distractor locations contribute to age-related decline in visual search. Furthermore, independently of age, dwelling and revisiting contributed stronger to search times than skipping of distractors. In conclusion, under conditions of poor guidance, dwelling and revisiting have a major contribution to search times and age-related slowing in difficult visual search, while skipping is largely negligible.

Statistical learning of spatiotemporal target regularities in the absence of saliency.

In previous studies, it was established that individuals can implicitly learn spatiotemporal regularities related to how the distribution of target locations unfolds across the time course of a single trial. However, these regularities were tied to the appearance of salient targets that are known to capture attention in a bottom-up way. The current study investigated whether the saliency of target is necessary for this type of learning to occur. In a visual search task, participants were instructed to search for a unique circle with a gap (Landolt C) among two other circles and indicate the location of the gap. Unbeknownst to them, the onset timing of search displays predicted the target location. Specifically, the target appeared more frequently at one peripheral location with an early onset of the search display and at the opposite peripheral location with a late onset of the search display. Additionally, we manipulated the gap size of the Landolt C to create either nonsalient (small gap) or salient (big gap) target events. Results showed that, regardless of prior exposure to salient targets (Experiment 1) or nonsalient targets (Experiment 2), visual search efficiency increased when the target appeared at the temporally valid location compared with the temporally invalid location. In conclusion, the saliency of targets and the associated bottom-up capture is not a prerequisite for learning dynamic distributional regularities of target locations during visual search.

Eye posture and screen alignment with simulated see-through head-mounted displays.

When rendering the visual scene for near-eye head-mounted displays, accurate knowledge of the geometry of the displays, scene objects, and eyes is required for the correct generation of the binocular images. Despite possible design and calibration efforts, these quantities are subject to positional and measurement errors, resulting in some misalignment of the images projected to each eye. Previous research investigated the effects in virtual reality (VR) setups that triggered such symptoms as eye strain and nausea. This work aimed at investigating the effects of binocular vertical misalignment (BVM) in see-through augmented reality (AR). In such devices, two conflicting environments coexist. One environment corresponds to the real world, which lies in the background and forms geometrically aligned images on the retinas. The other environment corresponds to the augmented content, which stands out as foreground and might be subject to misalignment. We simulated a see-through AR environment using a standard three-dimensional (3D) stereoscopic display to have full control and high accuracy of the real and augmented contents. Participants were involved in a visual search task that forced them to alternatively interact with the real and the augmented contents while being exposed to different amounts of BVM. The measured eye posture indicated that the compensation for vertical misalignment is equally shared by the sensory (binocular fusion) and the motor (vertical vergence) components of binocular vision. The sensitivity of each participant varied, both in terms of perceived discomfort and misalignment tolerance, suggesting that a per-user calibration might be useful for a comfortable visual experience.

Semantically congruent auditory primes enhance visual search efficiency: Direct evidence by varying set size.

Task-irrelevant sounds that are semantically congruent with the target can facilitate performance in visual search tasks, resulting in faster search times. In three experiments, we tested the underlying processes of this effect. Participants were presented with auditory primes that were semantically congruent, neutral, or incongruent to the visual search target, and importantly, we varied the set size of the search displays. According to seminal accounts of semantic priming, priming effects can be explained by processes not related to search (i.e., facilitation of target encoding; McNamara, 2013), which would predict a priming effect that is independent of set size. Alternatively, we tested if auditory priming can serve as a source of guidance for visual attention toward the primed target (i.e., in terms of altering attention-directing priorities; Wolfe, 2021), as indexed by higher search efficiency with congruent priming. Experiment 1 found that auditory color word primes resulted in faster responses and, importantly, flatter search slopes for congruent compared to incongruent color targets, indicating a more efficient search. As with many naturalistic search behaviors, we used multiple-target search. Experiment 2 replicated the findings of Experiment 1 with a reduced target set. Experiment 3 extended these findings to complex audiovisual objects. Our results provide direct evidence that cross-modal priming can guide visual selective attention, as reflected by enhanced visual search efficiency. (PsycInfo Database Record (c) 2025 APA, all rights reserved).

Transformative approaches for clinical trials in ophthalmology: Virtual reality as a metric for functional vision in inherited retinal diseases

Determining the efficacy of therapeutic interventions in ophthalmology requires not only the assessment of retinal structure and visual function but also a comprehensive evaluation of functional vision. Functional vision pertains to the patient's capacity to execute essential, vision‐dependent tasks, thereby maintaining autonomy. Traditional measures like visual acuity, contrast sensitivity, and visual field evaluations can fall short in encapsulating the full spectrum of visual deficits experienced by patients. To address this gap, translational researchers are increasingly focusing on developing performance outcomes (PerfOs) that provide a holistic measure of the therapy's impact on daily life activities.PerfOs are designed to strike a balance between measurement precision, relevance to the patient's daily life, and ecological validity, all while being adaptable for use in multicenter studies. A promising approach to achieving these objectives is the use of virtual reality (VR) for performance assessment. VR offers unparalleled control over experimental conditions, such as lighting, and enables rapid, objective behavioral measurements. Additionally, its reproducibility across various assessment centers and the safety it provides to participants make VR an ideal tool for this context.In this presentation, we introduce two innovative PerfOs developed for inherited retinal diseases, employing VR to enhance clinical endpoint accuracy in ophthalmology. The first is a mobility test, validated extensively in both real‐world and virtual settings. It demonstrates high reproducibility and reliability, effectively differentiates between retinitis pigmentosa patients and healthy controls, and accurately categorizes disease stages. The second is a VR‐based visual search task, specifically designed for the same patient population. These studies collectively affirm the potential of VR in establishing novel, meaningful clinical endpoints, significantly advancing the field of ophthalmology and enhancing patient‐centric care.

Autonomic contributions to attentional modulation of the cardiac defense response.

The cardiac defense response to aversive auditory stimulation is characterized by two acceleration/deceleration heart rate components. The first component is ascribed to attentional processes that facilitate detection and processing of potential threat, and the second one to protective actions. This study investigated attentional modulation of the cardiac defense response and the role of autonomic cardiac control therein. In 60 healthy subjects, the cardiac defense response was elicited, while electrocardiography, impedance cardiography and continuous blood pressure recordings were accomplished. Pre-ejection period represented sympathetic, heart rate variability parasympathetic, and systolic blood pressure sympathetic and parasympathetic, control. Half of the subjects performed a visual search task following the noise stimulus; the other half was exposed to the stimulus without any subsequent task. Task execution was associated with potentiation of the second heart rate acceleration/deceleration component of the defense response. Moreover, there was a greater systolic blood pressure decline during the second component. While pre-ejection period was unaffected by the task, the heart rate variability response was smaller overall in subjects performing the task. The findings suggest mediation of the first acceleration/deceleration component of the cardiac defense response by parasympathetic cardiac control; sympathetic and parasympathetic mechanisms contribute to the second component. While sympathetic control was unaffected by attentional demands, the potentiation of the cardiac defense response due to demands on external attention may relate to increased parasympathetic withdrawal.

Semantic information boosts the acquisition of a novel grammatical system in different presentation formats

Abstract Designing effective language learning settings requires an understanding of the processes taking place in language learning and the way they interact. One important issue concerns the interaction between meaning and grammar. A number of studies have shown a beneficial effect of semantics in grammar learning. What is unclear, however, is how far this effect may be influenced by the presentation formats of the semantic content. In two experiments, participants performed rule search tasks on Latin sentences. In Experiment 1, we presented semantic information in the form of naturalistic photographs, whereas in Experiment 2, the semantic information was implemented by quasi-translations. The control groups did not receive any semantic information. Learning performance was assessed by a grammaticality-judgment task combined with a source-attributions task. In both experiments, participants in the with-semantics group outperformed the respective control groups. Yet, only in Experiment 1, participants report having more explicit than implicit knowledge. We argue that semantic information boosts the acquisition of grammatical structures regardless of the presentation format. Furthermore, we suggest that, consistent with multimedia learning theories, the pictorial presentation format of Experiment 1 helped to use working memory capacity efficiently, which may have led to the generation of more explicit knowledge.