SEARCH · PubMed Health

Results for “Contrastive learning”

Explore indexed PubMed citations for clinical trials, systematic reviews and public health research. Read source abstracts and follow each citation to its original PubMed record.

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 37 records · Page 2Linked to original sources

PubMed2004

Perceptual learning in contrast discrimination: the effect of contrast uncertainty.

Performance in perceptual tasks improves with repetition (perceptual learning), eventually reaching a saturation level. Typically, when perceptual learning effects are studied, stimulus parameters are kept constant throughout the training and during the pre- and post-training tests. Here we investigate whether learning by repetition transfers to testing conditions in which the practiced stimuli are randomly interleaved during the post-training session. We studied practice effects with a contrast discrimination task, employing a number of training methods: (i) practice with a single, fixed pedestal (base-contrast), (ii) practice with several pedestals, and (iii) practice with several pedestals that included a spatial context. Pre- and post-training tests were carried out with the base contrast randomized across trials, under conditions of contrast uncertainty. The results showed that learning had taken place with the fixed pedestal method (i) and with the context method (iii), but only the latter survived the uncertainty test. In addition, we were able to identify a very fast learning phase in contrast discrimination that improved performance under uncertainty. We contend that learned tasks that do not pass the uncertainty test involve modification of decision strategies that require exact knowledge of the stimulus.

Contrast Sensitivity↗

PubMed2025

Deep learning-based annotation of plant abiotic stress resistance genes for crops.

The declining costs of DNA sequencing have expanded genomic data, crucial for understanding plant abiotic stress responses and crop improvement. However, accurate gene annotation remains challenging. To address this limitation, we propose the PASRGA, a deep learning approach that leverages transfer learning and contrastive learning to annotate genes related to drought, salt, cold, and UV resistance. PASRGA achieves high F1-scores, area under the receiver operating characteristic (AUROC), area under the precision-recall curve (AUPRC), and Matthews correlation coefficient (MCC) in annotating stress resistance genes, significantly outperforming the general protein annotation model CLEAN, the plant phosphatase gene annotation model PF-NET, the top-ranked model in the CAFA5 challenge NetGO 4.0, and four traditional machine learning methods. Its effectiveness was further validated with a salt stress treatment experiment in Eutrema salsugineum. To facilitate crop breeding practices, we utilized PASRGA to annotate the genomes of 17 major crops. To improve accessibility and utility, we incorporated both manually curated and PASRGA-predicted gene data, together with the PASRGA tool, into the PlantASRG database (https://bioinfor.nefu.edu.cn/PlantASRG/). This comprehensive resource aims to support crop breeding initiatives and ensure food security.

Crops, Agricultural↗

PubMed2026

CLASPP: A unified model for predicting post-translational modifications.

Post-Translational Modifications (PTMs) are a fundamental mechanism for regulating cellular pathways and increasing the functional diversity of the proteome. Accurately predicting the PTM types that are likely to occur at a given site in the primary sequence is a key challenge in functional proteomics. Existing PTM prediction models predominantly focus on either single PTM types or employ ensemble methods that combine multiple models to predict different PTM types. This fragmentation is largely driven by the vast imbalance in data availability across PTM types, making it difficult to predict multiple PTM types with a single model. To address this limitation, we present the Contrastively Learned Attention-based Stratified PTM Predictor (CLASPP), a unified PTM prediction model. CLASPP addresses imbalance challenges by leveraging unsupervised clustering-based undersampling and a novel contrastive learning framework tailored to PTM data. Additionally, our hierarchical data organization and curation are shown to improve CLASPP's performance by balancing the representation of individual PTM types and provides a standardized dataset to train and validate future model designs. Drawing inspiration from advancements in image and natural language processing, the CLASPP model employs a multi-stage training strategy and a high-quality, curated training dataset to improve PTM prediction performance. To uncover what is learned during the contrastive learning stage, the CLASPP model is shown to distinguish known protein kinase substrate specificity profiles as a form of explainability. Finally, we evaluate the application of CLASPP in predicting PTMs in different model organisms and experimentally validated ubiquitination sites in the understudied DCLK3 kinase. Overall, CLASPP represents a unified model for PTM prediction that addresses key bottlenecks in data imbalance and offers new strategies for biological data curation, thereby improving PTM-type prediction performance across diverse organisms.

Protein Processing, Post-Translational↗

PubMed2006

Perceptual learning in contrast detection: presence and cost of shifts in response criteria.

Contemporary theoretical accounts of perceptual learning typically assume that observers are either unbiased or stably biased across the course of learning. However, standard methods for estimating thresholds, as they are typically used, do not allow this assumption to be tested. We present an approach that allows for this test specific to perceptual learning for contrast detection. We show that reliable decreases in detection thresholds and increases in hit rates are not uniformly accompanied by reliable increases in sensitivity (d'), but are regularly accompanied by reliable liberal shifts in response criteria (c). In addition, we estimate the extent to which sensitivity could have increased in the absence of these liberal shifts. The results pose a challenge to the assumption that perceptual learning has limited or no impact on response criteria.

Adult↗

PubMed1998

Encapsulation of implicit and explicit memory in sequence learning.

Contrasts between implicit and explicit knowledge in the serial reaction time (SJRT) paradigm have been challenged because they have depended on a single dissociation; intact implicit knowledge in the absence of corresponding explicit knowledge. In the SRT task, subjects respond with a corresponding keypress to a cue that appears in one of four locations. The cue follows a repeating sequence of locations, and subjects can exhibit knowledge of the repeating sequence through increasingly rapid performance (an implicit test) or by being able to recognize the sequence (an explicit test). In our study, amnesic patients were given extensive SRT training. Their implicit and explicit test performance was compared to the performance of control subjects who memorized the training sequence. Compared with control subjects, amnesic patients exhibited superior performance on the implicit task and impaired performance on the explicit task. This crossover interaction suggests that implicit and explicit knowledge of the embedded sequence are separate and encapsulated and that they presumably depend on different brain systems.

Aged↗

PubMed2005

Perceptual learning improves contrast sensitivity and visual acuity in adults with anisometropic amblyopia.

To evaluate the effects of perceptual learning on contrast-sensitivity function and visual acuity in adult observers with amblyopia, 23 anisometropic amblyopes with a mean age of 19.3 years were recruited and divided into three groups. Subjects in Group I were trained in grating detection in the amblyopic eye near pre-training cut-off spatial frequency. Group II received a training regimen of repeated contrast-sensitivity function measurements in the amblyopic eye. Group III received no training. We found that training substantially improved visual acuity and contrast-sensitivity functions in the amblyopic eyes of all the observers in Groups I and II, although no significant performance improvement was observed in Group III. For observers in Group I, performance improvements in the amblyopic eyes were broadly tuned in spatial frequency and generalized to the fellow eyes. The latter result was not found in Group II. In a few cases tested, improvements in visual acuity following training showed about 90% retention for at least 1 year. We concluded that the visual system of adult amblyopes might still retain substantial plasticity. Perceptual learning shows potential as a clinical tool for treating child and adult amblyopia.

Adolescent↗

PubMed2001

The effect of orientation learning on contrast sensitivity.

Regan and Beverley [Regan, D., & Beverley, K. I. (1985). Postadaptation orientation discrimination. Journal of the Optical Society of America A, 2(2), 147-155] previously demonstrated that adapting to an oriented visual stimulus improves sensitivity to subtle orientation differences while impairing contrast sensitivity. Here, we investigated whether practice-based improvements in orientation sensitivity would, like adaptation, impair contrast sensitivity. To the contrary, we found that contrast sensitivity actually improved significantly after observers demonstrated practice-based increases in orientation sensitivity. Therefore, while orientation sensitivity can be enhanced either by orientation-discrimination training or by adapting to visual stimuli, these two procedures have opposite effects on contrast sensitivity. This difference suggests that adaptation and perceptual learning on orientation discrimination cannot be explained sufficiently by a shared underlying cause, such as a reduction in neural activity.

Adaptation, Physiological↗

PubMed1999

Detection and modeling of disease susceptibility locus effects: how much can be learned from contrast of populations?

We report the results of our analyses of the GAW11 Problem 2 data set, using information from three different populations. In the first part of the paper, we used classical population genetic tests to compare affected individuals from the different populations, stratifying on the environmental factors. Thanks to existing linkage disequilibrium in one population, we found one of the disease susceptibility loci. In the second part of the paper, we used the marker association segregation chi 2 method to model the role of this disease susceptibility locus in the different populations and draw some inferences regarding the model used at that locus to generate the data.

Chromosome Mapping↗

PubMed1998

Learning of sequential movements by neural network model with dopamine-like reinforcement signal.

Dopamine neurons appear to code an error in the prediction of reward. They are activated by unpredicted rewards, are not influenced by predicted rewards, and are depressed when a predicted reward is omitted. After conditioning, they respond to reward-predicting stimuli in a similar manner. With these characteristics, the dopamine response strongly resembles the predictive reinforcement teaching signal of neural network models implementing the temporal difference learning algorithm. This study explored a neural network model that used a reward-prediction error signal strongly resembling dopamine responses for learning movement sequences. A different stimulus was presented in each step of the sequence and required a different movement reaction, and reward occurred at the end of the correctly performed sequence. The dopamine-like predictive reinforcement signal efficiently allowed the model to learn long sequences. By contrast, learning with an unconditional reinforcement signal required synaptic eligibility traces of longer and biologically less-plausible durations for obtaining satisfactory performance. Thus, dopamine-like neuronal signals constitute excellent teaching signals for learning sequential behavior.

Algorithms↗

PubMed2005

Activation of the precuneus is related to reduced reaction time in serial reaction time tasks.

Multiple brain areas are activated during serial reaction time (RT) tasks (SRTTs), but the part of the brain that facilitates reductions in RT remains unclear. The present study attempted to determine the brain region contributing most to improved RTs during explicit SRTTs. Subjects comprised 18 healthy volunteers who were instructed to press one of four buttons corresponding to visual stimuli as quickly as possible and with minimal errors during functional MRI. Stimuli were presented either in random order (control condition) or in a repeated six-item sequence (learning condition). Conventional analysis contrasting learning and control conditions revealed activation in the prefrontal-parietal area, which shifted to motor area. Subjects with high RT reduction showed more prominent activation in the precuneus than subjects with low RT reduction. Intra-subject correlation analysis revealed that time course of precuneus activation was unrelated to time-course of RT reduction. However, inter-subject correlation analysis revealed that RT changes correlate only with precuneus activation, meaning that subjects showing more prominent RT reduction revealed more prominent activation of the precuneus, which is known to play critical roles in controlling finger movements with reference to buffered memory.

Adult↗

PubMed1984

Transfer of learning strategies: contrast of self-instructional and traditional training formats with EMR children.

Twelve mentally retarded children were taught anticipation and paraphrase strategies imbedded in either a self-instructional format that provided strategy initiation and modification routines or a traditional didactic training format: six children in a control condition received no explicit strategy instructions. Maintenance tests were administered 1 and 3 weeks following three training sessions; generalization tests were given in the afternoon of each training and maintenance session. Although no differences were found for self-instructional vs. traditional training formats, both conditions were superior to the control condition on the maintenance tests. Strategy generalization did not occur during training or short-term maintenance. On the final generalization test, however, significant improvements in recall accuracy were noted for both format conditions. In the self-instructional condition, several children successfully generalized the strategies during the early phases of training to facilitate gist recall.

Child↗

PubMed1992

Behavioural and electrophysiological studies of entorhinal cortex lesions in the rat.

Bilateral ibotenic acid injections aimed at the entorhinal cortex (EC) lesioned the EC and subiculum in 30% of animals (group EC/S) and caused additional hippocampal damage in 50% (group RH). Both lesions increased acetylcholinesterase (AChE) staining in the intermediate molecular layer of the dentate gyrus. EC/S lesions increased diurnal deep sleep and the incidence of spindles but decreased REM sleep. RH lesions increased nocturnal deep sleep and decreased nocturnal quiet sleep. Both lesions reduced power over the theta frequency range from 6-10 Hz for epochs of REM sleep and quiet waking but not deep sleep. Peak frequency was unaffected. The RH group and a subset of the EC/S group were nocturnally, but not diurnally, hyperactive. Six weeks after the lesion there was no evidence for hyperactivity in a novel open field. The EC/S lesion impaired exploration as indicated by reduced motility and rearing in an open field and by the failure of EC/S-lesioned rats to increase contact time in response to a novel olfactory cue. Place navigation learning in a Morris maze was not affected by EC/S or RH lesions. However, when the spatial location of the hidden platform was shifted EC/S-lesioned rats were impaired. The sprouting response, reduced theta power and exploration deficits resemble those reported following electrolytic lesions, but the lack of effect on place navigation learning contrasts with reports of impaired spatial learning following electrolytic lesions. The data prompt a reexamination of the role which the EC projection to the hippocampus plays in spatial learning.

Animals↗

PubMed2026

Survival prediction for clear cell renal cell carcinoma based on deep multimodal synergistic survival network.

Objective.To propose a deep multimodal synergistic survival analysis framework (Deep Multimodal Synergistic Survival Network, DMSSN) to achieve accurate prognostic analysis for clear cell renal cell carcinoma (ccRCC).Methods.This study (DMSSN) utilized matched multimodal data from the Cancer Genome Atlas-KIRC database, including CT imaging data, whole slide images, copy number variation (CNV) features, and clinical data. Deep Canonical Correlation Analysis was employed to map heterogeneous modalities into a shared latent space. Contrastive learning was introduced to enhance semantic consistency across multimodal features, and a gating network was utilized for the adaptive fusion of multimodal information to achieve precise survival risk prediction for patients.Results.Experimental results demonstrated that DMSSN achieved a Concordance Index (C-index) of 0.8153 ± 0.0994, with a Log-rank testp-value of 1.6553×10-11. DMSSN exhibited significant performance advantages over traditional statistical methods like Log-rank-Cox (0.7055 ± 0.0670) and machine learning methods such as Random Survival Forest (RSF) (0.6836 ± 0.1048). Furthermore, in comparison with similar deep learning approaches, DMSSN outperformed late fusion strategies (0.7493 ± 0.1211) and discrete-time survival models such as DeepHit (0.7655 ± 0.1041) and Nnet-surv (0.7694 ± 0.0635). Notably, DMSSN still achieved the best predictive performance when compared to the classic deep survival model DeepSurv (0.7919 ± 0.0978) and advanced state-of-the-art multimodal fusion frameworks like Context-Aware Transformer (0.7735 ± 0.0818) and Multimodal Co-Attention Transformer (0.8102 ± 0.0972). Ablation studies showed that removing any single modality led to a decline in performance, with the largest numerical decrease occurring after removing CT imaging features (C-index decreased to 0.7327), validating the complementarity of multimodal data and the pivotal role of radiomic features in prognostic assessment. Module ablation experiments further confirmed the effectiveness of the core components.Conclusion:By effectively integrating imaging, pathology, genomic, and clinical features, the DMSSN framework demonstrates superior performance and robustness in the survival prediction of ccRCC.

Carcinoma, Renal Cell↗

PubMed2001

Consistent contrast aids concept learning.

We suggest that coherence among concepts and correspondence between concepts and the world are important in concept learning. We identify one aspect of coherence, consistent contrast, and investigate its role in supervised concept learning. Concepts that contrast consistently carry information about the same attributes across the concepts within a contrast set. Concepts that contrast inconsistently predict and are predicted by values of different attributes. Experiment 1 revealed a large advantage for consistent contrast in learning and generalization. Experiment 2 pitted similarity against consistency and still revealed an advantage of consistency. Experiment 2 also broadened the range of tasks considered to include inductions about novel categories and subjects' category descriptions. We discuss relations to theories of concept learning, to attentional mechanism, and to alignability, and we suggest practical implications.

Adult↗

PubMed2002

Learning nonlinearly separable categories by inference and classification.

Previous research suggests that learning categories by classifying new instances highlights information that is useful for discriminating between categories. In contrast, learning categories by making predictive inferences focuses learners on an abstract summary of each category (e.g., the prototype). To test this characterization of classification and inference learning further, the authors evaluated the two learning procedures with nonlinearly separable categories. In contrast to previous research involving cohesive, linearly separable categories, the authors found that it is more difficult to learn nonlinearly separable categories by making inferences about features than it is to learn them by classifying instances. This finding reflects that the prototype of a nonlinearly separable category does not provide a good summary of the category members. The results from this study suggest that having a cohesive category structure is more important for inference than it is for classification.

Analysis of Variance↗

PubMed2006

Language and action control: the acquisition of action goals in early childhood.

This study examined the role of verbal labeling in 4-year-old children's acquisition of action-effect learning. The acquisition of action-effect associations was tested by having children first perform a two-choice key-pressing task in which each key press was followed by an effect (i.e., a particular sound) and then respond to the previously perceived effects under either consistent or inconsistent key-sound mappings. During acquisition, the children overtly described the actions, the effects, both the actions and the effects, or, in a control condition, something irrelevant to the actions and effects. Action-effect learning was reliable only if the description related actions to effects, even though some evidence of learning was also obtained in the control condition. In contrast, learning was prevented if only the actions or only the effects were described. The results suggest that verbal labeling plays an important role in integrating and isolating event representations.

Child, Preschool↗

PubMed2025

scMGCL: accurate and efficient integration representation of single-cell multi-omics data.

MOTIVATION: Single-cell multi-omics data integration is essential for understanding cellular states and disease mechanisms, yet integrating heterogeneous data modalities remains a challenge. We present scMGCL, a graph contrastive learning framework for robust integration of single-cell ATAC-seq and RNA-seq data. Our approach leverages self-supervised learning on cell-cell similarity graphs, in which each modality's graph structure serves as an augmentation for the other. This cross-modality contrastive paradigm enables the learning of biologically meaningful, shared representations while preserving modality-specific features. RESULTS: Benchmarking against state-of-the-art methods demonstrates that scMGCL outperforms others in cell-type clustering, label transfer accuracy, and preservation of marker-gene correlations. Additionally, scMGCL significantly improves computational efficiency, reducing runtime and memory usage. The method's effectiveness is further validated through extensive analyses of cell-type similarity and functional consistency, providing a powerful tool for multi-omics data exploration. AVAILABILITY AND IMPLEMENTATION: Code and datasets are released at https://github.com/zlCreator/scMGCL.

Single-Cell Analysis↗

PubMed1999

Deletion of the ryanodine receptor type 3 (RyR3) impairs forms of synaptic plasticity and spatial learning.

Deletion of the ryanodine receptor type 3 (RyR3) results in specific changes in hippocampal synaptic plasticity, without affecting hippocampal morphology, basal synaptic transmission or presynaptic function. Robust long-term potentiation (LTP) induced by repeated, strong tetanization in the CA1 region and in the dentate gyrus was unaltered in hippocampal slices in vitro, whereas weak forms of plasticity generated by either a single weak tetanization or depotentiation of a robust LTP were impaired. These distinct physiological deficits were paralleled by a reduced flexibility in re-learning a new target in the water-maze. In contrast, learning performance in the acquisition phase and during probe trial did not differ between the mutants and their wild-type littermates. In the open-field, RyR3(-/-) mice displayed a normal exploration and habituation, but had an increased speed of locomotion and a mild tendency to circular running. The observed physiological and behavioral effects implicate RyR3-mediated Ca(2+) release in the intracellular processes underlying spatial learning and hippocampal synaptic plasticity.

Animals↗