Search Template Research Articles

Persistent resampling of external information despite 25 repetitions of the same visual search templates

We commonly load visual working memory minimally when to-be-remembered information remains available in the external world. In visual search, this is characterised by participants frequently resampling previously encoded templates, which helps minimize cognitive effort and improves task performance. If all search templates have been rehearsed many times, they should become strongly represented in memory, possibly eliminating the benefit of reinspections. To test whether repetition indeed leads to less resampling, participants searched for sets of 1, 2, and 4 continuously available search templates. Critically, each unique set of templates was repeated 25 trials consecutively. Although the number of inspections and inspection durations initially decreased strongly when a template set was repeated, behaviour largely stabilised between the tenth and last repetition: Participants kept resampling templates frequently. In Experiment 2, participants performed the same task, but templates became unavailable after 15 repetitions. Strikingly, accuracy remained high even when templates could not be inspected, suggesting that resampling was not strictly necessary in later repetitions. We further show that seemingly ’excessive’ resampling behaviour had no direct within-trial benefit to speed nor accuracy, and did not improve performance on long-term memory tests. Rather, we argue that resampling was partially used to boost metacognitive confidence regarding memory representations. As such, eliminating the benefit of minimizing working memory load does not eliminate the persistence with which we sample information from the external world – although the underlying reason for resampling behaviour may be different.

The time course of activating, maintaining, and switching between attentional templates in visual search

The time course of activating, maintaining, and switching between attentional templates in visual search

Are Search Templates Target-object Reconstructions?

Are Search Templates Target-object Reconstructions?

MIMTracking: Masked image modeling enhanced vision transformer for visual object tracking

Recently, one-stage trackers achieve state-of-the-art tracking results due to more sufficient integration of search and template representations. These methods usually adopt an encoder for synchronous feature generation and interaction. Despite their high performance, the one-stage approaches tend to feed the encoder full input representations that are highly redundant and dense during training. With a focus on tackling this problem, a novel algorithm termed MIMTracking is developed for visual target tracking. MIMTracking exploits an encoder and a decoder for masked image modeling during training. Randomly sampled discrete search embeddings and template embeddings serve as input of the encoder. The lightweight decoder takes full representations as input and progressively highlights the target region. This design alleviates input redundancy and reduces the computational cost of the training process, thereby allowing for more efficient learning of useful representations. The proposed MIMTracking achieves state-of-the-art tracking results on numerous tracking datasets, e.g., 51.5 % area under curve (AUC) on LaSOText, outperforming the previous top tracker OSTrack by 2 %. Especially, our large model MIMTracking-L further improves the AUC to 53.4 % on LaSOText.

A new method for tracking the preparatory activation of target templates for visual search with high temporal precision.

Efficiently selecting task-relevant objects during visual search depends on foreknowledge of their defining characteristics, which are represented within attentional templates. These templates bias attentional processing toward template-matching sensory signals and are assumed to become anticipatorily activated prior to search display onset. However, a direct neural signal for such preparatory template activation processes has so far remained elusive. Here, we introduce a new high-definition rapid serial probe presentation paradigm (RSPP-HD), which facilitates high temporal resolution tracking of target template activation processes in real time via monitoring of the N2pc component. In the RSPP-HD procedure, task-irrelevant probe displays are presented in rapid succession throughout the period between task-relevant search displays. The probe and search displays are homologously formed by lateralized "clouds" of colored dots, yielding probes that occur at task-relevant locations without confounding template-guided and salience-driven attentional shifts. Target color probes appearing at times when a corresponding target template is active should attract attention, thereby eliciting an N2pc. In a condition where new probe displays appeared every 50 ms, probe N2pcs were reliably elicited during the final 800 ms prior to search display onset, increasing in amplitude toward the end of this preparation period. Analogous temporal profiles were also observed with longer intervals between probes. These findings show that search template activation processes are transient and that their temporal profile can be reliably monitored at high-sampling frequencies with the RSPP-HD paradigm. This procedure offers a new route to approach various questions regarding the content and temporal dynamics of attentional control processes.

Adherent Peanut Image Segmentation Based on Multi-Modal Fusion.

Aiming at the problem of the difficult segmentation of adherent images due to the not fully convex shape of peanut pods, their complex surface texture, and their diverse structures, a multimodal fusion algorithm is proposed to achieve a 2D segmentation of adherent peanut images with the assistance of 3D point clouds. Firstly, the point cloud of a running peanut is captured line by line using a line structured light imaging system, and its three-dimensional shape is obtained through splicing and combining it with a local surface-fitting algorithm to calculate a normal vector and curvature. Seed points are selected based on the principle of minimum curvature, and neighboring points are searched using the KD-Tree algorithm. The point cloud is filtered and segmented according to the normal angle and the curvature threshold until achieving the completion of the point cloud segmentation of the individual peanut, and then the two-dimensional contour of the individual peanut model is extracted by using the rolling method. The search template is established, multiscale feature matching is implemented on the adherent image to achieve the region localization, and finally, the segmentation region is optimized by an opening operation. The experimental results show that the algorithm improves the segmentation accuracy, and the segmentation accuracy reaches 96.8%.

Attentional Control Setting: Type, Quantity and Processing Stage

The attentional control setting is a search template containing target features. Individuals can establish attentional control settings based on the color, size, semantics, joint location-feature of stimulus and even relationships between targets and non-targets. When multiple targets exist, participants may establish multiple attentional control settings or a control setting based on the feature singleton item. However, the processing stage of the attentional control setting is still controversial. Future studies should further analyze the time course of attention control setting in different experimental paradigms to provide solutions to existing controversies, while examining the quantitative limits of attention control setting and the influence of distractor inhibition as a control setting on visual search.

Isolating the impact of a visual search template's color and form information on search guidance and verification times.

Visual search can be guided by biasing one's attention towards features associated with a target. Prior work has shown that high-fidelity, picture-based cues are more beneficial to search than text-based cues. However, typically picture cues provide both detailed form information and color information that is absent from text-based cues. Given that visual resolution deteriorates with eccentricity, it is not clear that high-fidelity form information would benefit guidance to peripheral objects - much of the picture benefit could be due to color information alone. To address this, we conducted a search task with eye-tracking that had four types of cues that comprised a 2 (text/pictorial cue) × 2 (no color/color) design. We hypothesized that color information would be important for efficient search guidance while high-fidelity form information would be important for efficient verification times. In Experiment 1 cues were a colored picture of the target, a gray-scaled picture of the target, a text-based cue that included color (e.g., "blue shoe"), or a text-based cue without color (e.g., "shoe"). Experiment 2 was a replication of Experiment 1, except that the color word in the text-based cue was presented in the precise color that was the dominant color in the target. Our results show that high-fidelity form information is important for efficient verifications times (with color playing less of a role) and color is important for efficient guidance, but form information also benefits guidance. These results suggest that different features of the cue independently contribute to different aspects of the search process.

A pilot search for MeV gamma-ray emission from five galaxy clusters using archival COMPTEL data

We search for MeV gamma-ray emission between 0.75–30 MeV from five galaxy clusters, viz. Coma, VIRGO, SPT-CL J2012-5649, Bullet, and El Gordo, using archival data from the COMPTEL telescope. For this purpose we use three search templates: point source, radial disk and radial Gaussian. We do not detect any signals from Coma, SPT-CL J2012-5649, Bullet and El Gordo clusters with the 95% c.l. photon energy flux limit ∼ 10-10erg/cm2/s.For VIRGO, we detect a non-zero signal between 0.75 to 1.50 MeV having marginal significance of about 2.5σ, with the observed energy flux equal to ∼ 10-9 ergs/cm2/s. However, we do not confirm the previously reported evidence in literature for a gamma-ray line from Coma and VIRGO clusters between 5–7 MeV.

UnifiedTT: Visual tracking with unified transformer

Target tracking is an important research task in computer vision. Existing tracking algorithms based on Siamese networks often suffer from the problem of information redundancy between adjacent frames and a lack of ability to capture global dependencies. When similar backgrounds appear around the target, the tracking performance usually significantly decreases. Although target tracking algorithms based on deep convolution and the Transformer have partially addressed these issues, achieving a good balance between the two remains a challenge. In this work, we propose a unified convolution and self-attention Siamese network for target tracking. By utilizing a feature extraction backbone network based on integrated convolution and self-attention styles, we are able to capture globally important regions and key frames while greatly reducing local redundant computations, thereby improving tracking performance. We apply this approach to the task of target tracking and enhance the feature extraction capabilities of both the target template and search template. Experimental results show that our proposed tracking algorithm outperforms some recent classical tracking algorithms, especially achieving improvements of 10.7 % on the high-diversity dataset GOT-10K and 24.7 % on the large-scale and high-quality LaSOT dataset.

Do We Prepare for What We Predict? How Target Expectations Affect Preparatory Attentional Templates and Target Selection in Visual Search.

Visual search is guided by representations of target-defining features (attentional templates) that are activated in a preparatory fashion. Here, we investigated whether these template activation processes are modulated by probabilistic expectations about upcoming search targets. We tracked template activation while observers prepared to search for one or two possible color-defined targets by measuring N2pc components (markers of attentional capture) to task-irrelevant color probes flashed every 200 msec during the interval between search displays. These probes elicit N2pcs only if the corresponding color template is active at the time when the probe appears. Probe N2pcs emerged from about 600 msec before search display onset. They did not differ between one-color and two-color search, indicating that two color templates can be activated concurrently. Critically, probe N2pcs measured during two-color search were identical for probes matching an expected or unexpected color (target color probability: 80% vs. 20%), or one of two equally likely colors. This strongly suggests that probabilistic target color expectations had no impact on search preparation. In marked contrast, subsequent target selection processes were strongly affected by these expectations. We discuss possible explanations for this clear dissociation in the effects of expectations on preparatory search template activation and search target selection, respectively.

Contingent capture by color is sensitive to categorical color perception.

Contingent capture (CC) theory postulates that attention can only be captured by top-down matching stimuli. Although the contingent capture of attention is a well-known and thoroughly studied phenomenon, there is still no consensus on the characteristics of the top-down template which guides the search for colors. We tried to replicate the classical contingent capture effect on color (Experiment 1) and then added linguistic processing to this perceptual effect (Experiment 2). In Experiment 1, attention was indeed captured by the cues of the same color as the target, while the cues of different colors were successfully ignored. In Experiment 2, the cue color was never identical to the target color but would either belong to the same linguistic category or not (i.e., linguistic matching and linguistic nonmatching cues). In both cases, cues were made to be equally perceptually distant from the target. Although, attention was captured by both cue types, the degree of capture was significantly higher for linguistic matching cues. Our research replicated the classic contingent capture effect but on color, and also demonstrated the effect of color categories in the search task. In short, we demonstrated the effect of color categories in the search task. Results show that the template for color search contains physical characteristics of color, as well as information about color category names.

Consistent Weighted Correlation-Based Attention for Transformer Tracking

Attention mechanism takes a crucial role among the key technologies in transformer-based visual tracking. However, the current methods for attention computing neglect the correlation between the query and the key, which results in erroneous correlations. To address this issue, a CWCTrack framework is proposed in this study for transformer visual tracking. To balance the weights of the attention module and enhance the feature extraction of the search region and template region, a consistent weighted correlation (CWC) module is introduced into the cross-attention block. The CWC module computes the correlation score between each query and all keys. Then, the correlation multiplies the consistent weights of the other query–key pairs to acquire the final attention weights. The weights of consistency are computed by the relevance of the query–key pairs. The correlation is enhanced for the relevant query–key pair and suppressed for the irrelevant query–key pair. Experimental results conducted on four prevalent benchmarks demonstrate that the proposed CWCTrack yields preferable performances.

Knowledge-based multiple point statistics for soil stratigraphy simulation

An accurate representation of subsurface stratigraphy in the form of geotechnical profiles is a prerequisite for optimal underground design and safe construction. This paper proposes a novel knowledge-based multiple point statistics (KMPS) method to simulate the heterogeneities and spatial trends of subsurface strata and to generate geotechnical profiles. This approach incorporates domain knowledge in the form of cone penetration test and standard penetration test data as well as results from laboratory testing to increase simulation accuracy. A novel varying search template size and similarity threshold is adopted to alleviate the computational cost of the simulations. Ablation analyses are used to quantify the improvement of the proposed KMPS algorithm over existing baseline multiple point statistics algorithms. By way of example, the proposed approach is applied to real-world exploration data from the Suzhou No.6 metroline project. The results show that the computational cost of KMPS is effectively minimized whilst the simulation accuracy can be improved by up to 13.3%. The algorithm is further validated by extending the simulation to a test image. With sparse exploration data, the KMPS method is shown to effectively and efficiently predict soil profiles based on prior experience embodied by a training image, thus enabling a reduction in the number of field exploration boreholes.

Detectability of microlensed gravitational waves

ABSTRACT Gravitational lensing describes the bending of the trajectories of light and gravitational waves due to the gravitational potential of a massive object. Strong lensing by galaxies can create multiple images with different overall amplifications, arrival times, and image types. If, furthermore, the gravitational wave encounters a star along its trajectory, microlensing will take place. Previously, it has been shown that the effects of microlenses on strongly-lensed type-I images could be negligible in practice, at least in the low magnification regime. In this work, we study the same effect on type-II strongly-lensed images by computing the microlensing amplification factor. As opposed to being magnified, type-II images are typically demagnified. Moreover, microlensing on top of type-II images induces larger mismatches with un-microlensed waveforms than type-I images. These results are broadly consistent with recent literature and serve to confirm the findings. In addition, we investigate the possibility of detecting and analysing microlensed signals through Bayesian parameter estimation with an isolated point mass lens template, which has been adopted in recent parameter estimation literature. In particular, we simulate gravitational waves microlensed by a microlens embedded in a galaxy potential near moderately magnified type-I and II macroimages, with variable lens masses, source parameters and macromagnifcations. Generally, an isolated point mass model could be used as an effective template to detect a type-II microlensed image but not for type-I images, demonstrating the necessity for more realistic microlensing search templates.

Testing a relational account of search templates in visual foraging

Search templates guide human visual attention toward relevant targets. Templates are often seen as encoding exact target features, but recent studies suggest that templates rather contain “relational properties” (e.g., they facilitate “redder” stimuli instead of specific hues of red). Such relational guidance seems helpful in naturalistic searches where illumination or perspective renders exact feature values unreliable. So far relational guidance has only been demonstrated in rather artificial single-target search tasks with briefly flashed displays. Here, we investigate whether relational guidance also occurs when humans interact with the search environment for longer durations to collect multiple target elements. In a visual foraging task, participants searched for and collected multiple targets among distractors of different relationships to the target colour. Distractors whose colour differed from the environment in the same direction as the targets reduced foraging efficiency to the same amount as distractors whose colour matched the target colour. Distractors that differed by the same colour distance but in the opposite direction of the target colour did not reduce efficiency. These findings provide evidence that search templates encode relational target features in naturalistic search tasks and suggest that attention guidance based on relational features is a common mode in dynamic, real-world search environments.

Beyond general relativity: Designing a template-based search for exotic gravitational wave signals

Accurate waveform models describing the complete evolution of compact binaries are crucial for the maximum likelihood detection framework, testing the predictions of general relativity (GR) and investigating the possibility of an alternative theory of gravity. Deviations from general relativity could manifest in subtle variations of the numerical value of the gravitational wave signal's phasing coefficients. Once the search pipelines confirm an unambiguous signal detection, deviations of the signal phasing coefficients at various post-Newtonian orders are routinely measured and reported. As the search templates themselves do not incorporate any deviations from general relativity, they may miss astrophysical signals carrying a significant departure from GR. We present a parametrized template-based search for exotic gravitational-wave signals beyond general relativity by incorporating deviations to the signal's phasing coefficients at different post-Newtonian orders in the search templates. We present critical aspects of the new search, such as improvements in search volume and its effect on various parts of the parameter space. In particular, we demonstrate a twofold increase in search sensitivity (at a fixed false-alarm rate) to beyond-GR exotic signals by using search templates that admit a range of departures from general relativity. We also present the results from a reanalysis of ten days long duration of LIGO data from the first science run that includes the epoch of the GW150914 event.

Search templates for real-world objects in natural scenes

Search templates for real-world objects in natural scenes

The allocation of working memory resources determines the efficiency of attentional templates in single- and dual-target search.

Attentional templates are representations of target features in working memory (WM). Although two attentional templates can guide visual search in dual-target search, search efficiency is reduced compared with one attentional template in single-target search. Here, we investigated whether the allocation of WM resources contributes to these differences. Participants always memorized two colors, but the use of the corresponding WM representations varied. In the blocked conditions, the two colors were either maintained as attentional templates for dual-target search or as simple WM representations for recall only. In the mixed condition, one color was maintained as an attentional template for single-target search and the other as a simple WM representation for recall only. Reaction times (RTs) were delayed and recall precision reduced with two attentional templates in the blocked condition compared with one attentional template in the mixed condition, indicating that search efficiency and WM resources decreased in dual- compared with single-target search. Moreover, the attentional template was always recalled more precisely than the simple WM representation in the mixed condition, despite lowered visual search frequency (Experiment 2) and retro-cueing (Experiment 3). Consistent with the existence of an "active" WM state, resources were strongly biased toward the attentional template in single-target search. In dual-target search, however, resources were balanced between two attentional templates and flexibly adjusted with retro-cues, as with two simple WM representations. Therefore, the allocation of WM resources goes beyond the traditional dichotomy between "active" and "accessory" WM states and explains how attentional templates guide visual search with variable efficiency. (PsycInfo Database Record (c) 2022 APA, all rights reserved).

A Flippase-Like Domain-Containing Protein From Deltaproteobacteria Bacterium (Marine Sediment Metagenome) Has Anti-prostate Cancer Effects

Background: Prostate cancer is the second prevalent cancer and the fifth cause of deaths related to the men malignancies. The cancerous cell can travel from prostate to other parts of the body and cause metastasis in other organs such as bones and lymph nodes. For the treatment of the prostate cancer, chemotherapy and hormone therapy are being used. Hormone therapy in this disease is primarily based on the androgen suppression to prevent the growth and division of the cancerous cells. Different classes of drugs are used for hormone therapy. One group of these drugs is reducers of androgen production in adrenal glands. These drugs prevent the androgen production in adrenal glands and cancerous cells. Leuprorelin which is also known as leuprolide is a synthetic peptide that is used for treatment of prostate cancer. Objectives: The aim of this study is to explore novel anti-prostate cancer compounds using proteomics of microorganisms. Methods: Leuprorelin synthetic peptide was chosen as the template for amino acid sequence homology search using blast, and the resulting protein of the best candidate sequences were employed as a query to perform docking test against the gonadotropin-releasing hormone receptor (GRHR) using HDOCK web server. Results: By employing an in silico approach, natural products with structure and function similar to leuprorelin were explored, and their features were characterized. A flippase-like domain-containing protein from Deltaproteobacteria bacterium showed strong binding to gonadotropin-releasing hormone receptor GRHR with docking score of -425. The absorption, distribution, metabolism, and excretion (ADME) tests showed no potential toxicity of this natural product to the body. Conclusion: The present study demonstrated that the proteomics of living organisms contains natural compounds that can be considered a valuable source of medically important resources.