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 1,243 records · Page 69Linked to original sources

PubMed1998

Learning in year-old female autoimmune BXSB mice.

BXSB/ MpJ-Yaa and NZB/BINJ mice have been used as animal models for both developmental learning disability and systemic autoimmune disease. Approximately 40-60% of these animals show ectopic clusters of neurons in Layer I of cortex similar to those found in postmortem analyses of human dyslexics, and all exhibit an autoimmune condition similar to systemic lupus erythematosus (SLE) in humans. The expression of immune disease in the BXSB strain, unlike in humans, is more severe in males than females. Most previous studies have examined the behavioral sequelae of neocortical ectopias at a relatively young age, when the BXSB females (unlike the male BXSB and female and male NZBs) are not yet showing high titers of autoantibodies associated with their lupus-like form of autoimmune disease. This study examined the behavior of BXSB females at an age subsequent to autoimmune disease onset. When contrasted with younger animals, year-old BXSB females showed good learning behavior, with no differences in Lashley maze learning and remarkably good performance in a visual discrimination learning task. These results are consistent with other data which indicate that many types of learning behavior are apparently unperturbed by systemic autoimmune disease. Results also showed significant interactions between a measure of lateral paw preference and the presence or absence of ectopias in Lashley maze learning. Animals without ectopias that exhibited a right lateral paw preference showed the greatest number of errors on a number of test measures. These findings support previous results indicating that behavioral effects associated with ectopias may vary based upon the behavioral laterality of affected animals.

Age Factors↗

PubMed2004

[Voluntary postural control learning with a use of visual bio-feedback in patients with spinocerebellar degenerations].

The study aimed at evaluation of possibility and features of voluntary postural control learning using biofeedback from a force platform in patients with spinocerebellar ataxias. Thirty-seven patients with different forms of spinocerebellar degenerations and 13 age-matched healthy subjects were trained to shift the center of pressure (CP) during several stabilographic computer games which tested an ability to learn 2 different types of voluntary postural control: general strategy and precise coordination of CP shifting. Despite the disturbances of static posture and ability for voluntary control of CP position, patients with spinocerebellar degenerations can learn to control a vertical posture using biofeedback on stabilogram. In contrast to healthy subjects, improvement of coordination in the training process does not exert a significant influence on the static posture characteristics, in particular on lateral CP oscillations. The results obtained suggest involvement of the cerebellum in both types of postural control that distinguishes them from pathology caused by motor cortex and nigro-striatal system involved only in one type of postural control.

Adolescent↗

PubMedDate not supplied

Environmental stimulation of 129/SvJ mice causes increased cell proliferation and neurogenesis in the adult dentate gyrus.

New neurons are continuously born in the dentate gyrus of the adult mouse hippocampus, and regulation of adult neurogenesis is influenced by both genetic and environmental determinants. Mice of the 129/SvJ strain have significantly less hippocampal neurogenesis than other inbred mouse strains [1] and do not perform well in learning tasks. Here, the impact of environmental stimuli on brain plasticity during adulthood of 129/SvJ mice was studied using 'enriched environments' where mice receive complex inanimate and social stimulation [2,3]. In contrast to our earlier reports on mice of the C57BL/6 strain - which are competent in learning tasks and in which environmental stimulation did not influence cell proliferation [4,5] - environmentally stimulated 129/SvJ mice were found to have twice as many proliferating cells in the dentate gyrus compared with mice in standard housing. Environmental stimulation fostered the survival of newborn cells in 129/SvJ mice; this effect had also been seen in C57BL/6 mice. Phenotypic analysis of the surviving cells revealed that environmental stimulation resulted in 67% more new neurons. In combination with our earlier results, these data indicate a differential impact of inheritable traits on the environmental regulation of adult hippocampal neurogenesis. In addition, we observed behavioral changes in environmentally stimulated 129/SvJ mice.

Animals↗

PubMed2005

Metabotropic glutamate receptor 1 blockade impairs acquisition and retention in a spatial Water maze task.

Metabotropic glutamate receptors, including the mGlu1 receptor, have received considerable attention as potential targets for anxiolytic, antidepressant, antipsychotic and antinociceptive drugs. mGlu1 receptors have also been suggested to play a role in the modulation of cognitive processes, but knowledge is still very limited. In the present study the effects of the selective mGlu1 receptor antagonist 3,4-dihydro-2H-pyrano[2,3]beta-quinolin-7-yl)(cis-4-methoxycyclohexyl)methanone (JNJ16259685, 0.63-10 mg/kg s.c.) on more or less spatially demanding learning and spatial memory (retention and re-acquisition) were investigated in mice performing in a water maze. Selective mGlu1 receptor blockade with JNJ16259685 impaired spatial acquisition processes, irrespective of spatial load, as well as spatial re-acquisition, already at the lowest dose tested (0.63 mg/kg). In contrast, effects on spatial retention performance were relatively mild in mice that had learned to locate the position of the escape platform prior to treatment. Thigmotaxic behaviour and locomotor activity appeared to be unaffected by JNJ16259685. These data suggest that blockade of the mGlu1 receptor primarily affects learning of new information, but leaves retention of spatial information relatively unaffected. Blockade of the mGlu5 receptor with MPEP also impaired spatial learning, although only at the highest dose tested (10 mg/kg). An ex vivo receptor occupancy study in rats revealed that MPEP occupied central mGlu5 receptors with an ED(50) of 2.0 mg/kg one hour after subcutaneous administration. This is 50-150 times higher than the ED(50) reported for JNJ16259685 at central mGlu1 receptors and suggests that one reason why the two compounds cause cognitive effects at different doses might be due to differences in central mGlu receptor occupancy, rather than fundamentally different roles of mGlu1 and mGlu5 receptors in the modulation of cognitive function.

Animals↗

PubMed2006

The role of recollection and familiarity in the context variability mirror effect.

Context variability refers to the number of preexperimental contexts that are associated with concepts. In four experiments, we investigated the basis for increased recognition memory for low context variability words. Low context variability was associated with greater recollection in the hit rates, and high context variability was associated with greater familiarity in the false alarms. Shortening the study time reduced recollection, but low context variability still influenced recollection in the hit rates. A modality change from study to test also reduced recollection but preserved recollective differences for low versus high context variability items. One interpretation of the results suggests that low context variability evokes more specific and, perhaps, idiosyncratic recollective associations during learning and that these associations support better recognition in the hit rates. By contrast, activating the larger number of associations for high context variability items may be mistaken for familiarity in the false alarm rates.

Association Learning↗

PubMed2001

Anticipatory responses of dopamine neurons and cortical neurons reproduced by internal model.

Animals seem to learn and use internal models; they learn to anticipate predictable events, and their behavior in the sensory preconditioning paradigm reflects formation of novel associative chains. To investigate possible neural correlates, the temporal difference model (TD model) was extended to an internal model approach. The proposed model learns reward prediction error signals that resemble dopamine neuron activity. In contrast to the original TD model, the reward prediction error signals of the proposed model are influenced by the formation of novel associative chains in the sensory preconditioning experiment. This is consistent with experimental findings, as striatal dopamine concentration is influenced by the formation of novel associative chains in this paradigm. Comparison of the model architecture with biological neural networks suggests that chains of neurons with tonic anticipatory activity may underlie the formation of novel associative chains. These findings suggest that dopamine neuron activity may reflect the processing of an internal model.

Algorithms↗

PubMed2005

Implicit and explicit memory in children with congenital and acquired brain disorder.

Implicit and explicit memory were examined in 8- to 15-year-old children with myelomeningocele and shunted hydrocephalus, severe traumatic brain injuries, or orthopedic injuries. Each group included between 22 and 29 children. Children completed a fragmented picture identification task to assess perceptual priming and a semantic decision-making task to assess conceptual priming. Each task also assessed procedural learning as well as explicit recall and recognition. All 3 groups showed significant perceptual and semantic priming of similar magnitude. In contrast, both brain-disordered groups displayed poorer explicit memory than did the comparison group. No group showed significant procedural learning on either task. Age and IQ were stronger predictors of explicit recall than of implicit memory. The findings indicate that implicit memory is relatively intact in many children with congenital and acquired brain disorders, despite deficits in explicit memory, and support the existence of separate memory systems in children.

Adolescent↗

PubMed1981

Neonatal Frontal Lesions in the rat: sparing of learned but not species-typical behavior in the presence of reduced brain weight and cortical thickness.

Rats with complete removal of the cortex anterior to bregma in adulthood (frontal cortex) were compared behaviorally and neuroanatomically with rats with similar removals at 7 or 25 days of age. Excision of the frontal cortex in adult rats produced transient aphagia, chronic motor abnormalities in feeding, a chronic drop in body weight, increased activity in running wheels, impaired performance at a spatial reversal learning task, and chronic abnormalities in a variety of species-typical behaviors, including swimming, food hoarding, and defensive burying. In contrast, similar lesions in infant rats failed to produce aphagia, a chronic drop in body weight, increased activity, or impaired learning of a spatial reversal task. Infant lesions did not allow sparing of complex species-typical behaviors, however, such as those involved in feeding, swimming, hoarding, or defensive burying. Furthermore, when the brains of neonatally operated rats were compared with those of control rats or rats operated on in adulthood, there were striking differences. The cerebral hemispheres of the neonatal operates were smaller both in surface dimensions and weight, the thalamus was smaller, and the cerebral cortex was thinner. These data imply that there may be substantially less sparing of function following frontal cortex lesions in infancy than previously believed and that neonatal frontal lesions in rats have significant effects on brain development in regions far removed from the actual site of surgical excision.

Aging↗

PubMed2006

The role of the basolateral amygdala in stimulus-reward memory and extinction memory consolidation and in subsequent conditioned cued reinstatement of cocaine seeking.

The consolidation of cue-cocaine associations and extinction learning (i.e. cue-no cocaine associations) into long-term memory probably regulates the long-lasting control of conditioned stimuli (CS) over cocaine-seeking behaviour, and the basolateral amygdala (BLA) may play a role in this phenomenon. To test this hypothesis, rats previously trained to self-administer cocaine underwent a single classical conditioning (CC) session, during which they received passive pairings of cocaine infusions and a novel light + tone stimulus complex. After additional self-administration sessions in the absence of CS presentation and subsequent extinction training sessions, the ability of the CS to reinstate cocaine-seeking on five test days was assessed. Rats received intra-BLA microinfusions of vehicle or the Na+-channel blocker tetrodotoxin (TTX) immediately after CC (consolidation of CS-cocaine associations) or immediately after reinstatement testing (consolidation of extinction learning). TTX administered immediately after CC attenuated subsequent CS-induced reinstatement. In contrast, TTX administered after the first reinstatement test impaired the extinction of cocaine-seeking behaviour during a second reinstatement test by disrupting extinction memory. Overall, these findings suggest that Na+ channel-mediated mechanisms within the BLA mediate the consolidation of both cocaine-stimulus association and extinction learning, two processes that have opposite effects on subsequent cue-induced cocaine-seeking behaviour.

Amygdala↗

PubMed1999

Facilitation of NMDAR-independent LTP and spatial learning in mutant mice lacking ryanodine receptor type 3.

To evaluate the role in synaptic plasticity of ryanodine receptor type 3 (RyR3), which is normally enriched in hippocampal area CA1, we generated RyR3-deficient mice. Mutant mice exhibited facilitated CA1 long-term potentiation (LTP) induced by short tetanus (100 Hz, 100 ms) stimulation. Unlike LTP in wild-type mice, this LTP was not blocked bythe NMDA receptor antagonist D-AP5 but was partially dependent on L-type voltage-dependent Ca2+ channels (VDCCs) and metabotropic glutamate receptors (mGluRs). Long-term depression (LTD) was not induced in RyR3-deficient mice. RyR3-deficient mice also exhibited improved spatial learning on a Morris water maze task. These results suggest that in wild-type mice, in contrast to the excitatory role of Ca2+ influx, RyR3-mediated intracellular Ca2+ ([Ca2+]i) release from endoplasmic reticulum (ER) may inhibit hippocampal LTP and spatial learning.

Animals↗

PubMed2006

Age effects on brain activity during repetition priming of targets and distracters.

The effects of age on repetition priming and how such differences were related to intentional learning and working memory status were examined. Fourteen older (age 65-75) and 14 younger (age 18-28) healthy adults performed a modified delayed match-to-sample task consisting of a target object held in mind followed by nine test objects. Sixty four-channel EEGs were recorded as participants indicated whether each test object was the same or different from the target object. Half of all target and distractor objects were intentionally studied prior to the task, and both target and distractor objects were repeatedly presented up to four times in each trial. Although both age groups showed repetition priming effects, speed increases due to repetition were more enhanced for elderly. ERP repetition effects for both younger and older adults were indexed via early (200-550) and late (550-850 ms) components. The early repetition effect was affected by whether a distractor was previously studied or not for younger but not for older adults. In contrast, the late repetition effect was not affected by prior intentional learning, and a marginal age effect suggested that repetitions of distractors likely affected older and younger adults differently. These findings suggest that at least two distinguishable repetition mechanisms differentially affect adult aging.

Adolescent↗

PubMed2003

Towards valid measures of self-directed clinical learning.

AIM: To compare the validity of different measures of self-directed clinical learning. METHODS: We used a quasi-experimental study design. The measures were: (1) a 23-item quantitative instrument measuring satisfaction with the learning process and environment; (2) free text responses to 2 open questions about the quality of students' learning experiences; (3) a quantitative, self-report measure of real patient learning, and (4) objective structured clinical examination (OSCE) and progress test results. Thirty-three students attached to a single firm during 1 curriculum year in Phase 2 of a problem-based medical curriculum formed an experimental group. Thirty-one students attached to the same firm in the previous year served as historical controls and 33 students attached to other firms within the same module served as contemporary controls. After the historical control period, experimental group students were exposed to a complex curriculum intervention that set out to maximise appropriate real patient learning through increased use of the outpatient setting, briefing and supported, reflective debriefing. RESULTS: The quantitative satisfaction instrument was insensitive to the intervention. In contrast, the qualitative measure recorded a significantly increased number of positive statements about the appropriateness of real patient learning. Moreover, the quantitative self-report measure of real patient learning found high levels of appropriate learning activity. Regarding outpatient learning, the qualitative and quantitative real patient learning instruments were again concordant and changed in the expected direction, whereas the satisfaction measure did not. An incidental finding was that, despite all attempts to achieve horizontal integration through simultaneously providing community attachments and opening up the hospital for self-directed clinical learning, real patient learning was strongly bounded by the specialty interest of the hospital firm to which students were attached. Assessment results did not correlate with real patient learning. CONCLUSIONS: Both free text responses and students' quantitative self-reports of real patient learning were more valid than a satisfaction instrument. One explanation is that students had no benchmark against which to rate their satisfaction and curriculum change altered their tacit benchmarks. Perhaps the stronger emphasis on self-directed learning demanded more of students and dissatisfied those who were less self-directed. Results of objective, standardised assessments were not sensitive to the level of self-directed, real patient learning. Despite an integrated curriculum design that set out to override disciplinary boundaries, students' learning remained strongly influenced by the specialty of their hospital firm.

Clinical Competence↗

PubMed1998

Constructive incremental learning from only local information

We introduce a constructive, incremental learning system for regression problems that models data by means of spatially localized linear models. In contrast to other approaches, the size and shape of the receptive field of each locally linear model, as well as the parameters of the locally linear model itself, are learned independently, that is, without the need for competition or any other kind of communication. Independent learning is accomplished by incrementally minimizing a weighted local cross-validation error. As a result, we obtain a learning system that can allocate resources as needed while dealing with the bias-variance dilemma in a principled way. The spatial localization of the linear models increases robustness toward negative interference. Our learning system can be interpreted as a nonparametric adaptive bandwidth smoother, as a mixture of experts where the experts are trained in isolation, and as a learning system that profits from combining independent expert knowledge on the same problem. This article illustrates the potential learning capabilities of purely local learning and offers an interesting and powerful approach to learning with receptive fields.

Journal Article↗

PubMed2006

Early learners' discrimination of second-language vowels.

It is uncertain from previous research to what extent the perceptual system retains plasticity after attunement to the native language (L1) sound system. This study evaluated second-language (L2) vowel discrimination by individuals who began learning the L2 as children ("early learners"). Experiment 1 identified procedures that lowered discrimination scores for foreign vowel contrasts in an AXB test (with three physically different stimuli per trial, where "X" was drawn from the same vowel category as "A" or "B"). Experiment 2 examined the AXB discrimination of English vowels by native Spanish early learners and monolingual speakers of Spanish and English (20 per group) at interstimulus intervals (ISIs) of 1000 and 0 ms. The Spanish monolinguals obtained near-chance scores for three difficult vowel contrasts, presumably because they did not perceive the vowels as distinct phonemes and because the experimental design hindered low-level encoding strategies. Like the English monolinguals, the early learners obtained high scores, indicating they had shown considerable perceptual learning. However, statistically significant differences between early learners and English monolinguals for two of three difficult contrasts at the 0-ms ISI suggested that their underlying perceptual systems were not identical. Implications for claims regarding perceptual plasticity following L1 attunement are discussed.

Adult↗

PubMed1997

Assessment of learning by the Morris water task and fear conditioning in inbred mouse strains and F1 hybrids: implications of genetic background for single gene mutations and quantitative trait loci analyses.

Genetic methods including the creation of transgenic or null mutant models and mapping studies using quantitative trait loci strategies can be used to identify candidate genes in mice that regulate learning processes. Interpretations as to the impact of single gene mutations for polygenic behaviours like learning will depend in part on the genetic background of the animals used for these manipulations. To address the issue of genetic variability, 12 inbred strains and seven different F1 hybrids were tested on multiple behavioural tasks, including two complex learning paradigms: the Morris water task and fear conditioning. Strain differences were found for all variables measured. In the hidden platform version of the Morris task, the albino animals performed poorly while overall the F1 hybrids showed the best selectivity for the trained quadrant as measured in a probe trial. In contrast, almost all genotypes performed well on the contextual fear conditioning task and learned to associate the test context with the pairing of a foot shock and auditory stimulus as demonstrated 24 h after training by increased freezing in the test environment compared to an altered context. Significant genetic correlations were obtained for behavioural measures suggesting that the same genes regulate various aspects of performance on behavioural tasks. Scores from these multiple inbred strains and F1 hybrids provide a baseline level of learning ability for fear conditioning and the Morris water task. The results of the present study confirm the importance of genetic background in the performance of various learning tasks. This variability should be considered when developing new transgenic or null mutant animal models.

Acoustic Stimulation↗

PubMed2000

Disconjugate vertical memory-guided saccades to disparate targets.

We studied the binocular coordination of normal memory-guided saccades and the possibility of inducing memory-based disconjugate learning. First, we report that normal vertical memory-guided saccades to non-disparate targets are yoked vertically in the two eyes as well as visually guided vertical saccades. To induce disconjugate vertical learning, at each trial we flashed a target that was disparate (i.e. 8% more elevated or more depressed for one eye); the memory delay was 1 s. Memory-guided vertical saccades developed a vertical disconjugacy that was appropriate for the disparity of the remembered target. After 15 min of repetition, this vertical disconjugacy persisted even when the target to be remembered was no longer disparate; this indicates disconjugate vertical learning based on short-term memory. However, this was observed only for a few individual cases and its amplitude was small. This contrasts with prior findings on horizontal saccades associated with horizontal disparities. We conclude that vertical memory-based disconjugate learning is possible but very limited. Together with other studies, this study suggests that the natural vertical conjugacy of vertical saccades relies little on rapid learning mechanisms. Rather it seems to be built-in, and this is consistent with earlier electrophysiological findings.

Convergence, Ocular↗

PubMed1997

[Nondeclarative memory--neuropsychological findings and neuroanatomic principles].

The contents of long-term memory will influence behaviour, even if the acquired knowledge or the original learning episode are not remembered. These phenomena have been termed "non-declarative" or "implicit" memory, and they are contrasted with "declarative" or "explicit" memory which is characterised by conscious search and retrieval procedures. Non-declarative memory encompasses non-associative learning, simple conditioning, priming effects as well as motor, perceptual and cognitive skill acquisition. The dissociation of both forms of memory is documented by studies in health subjects which indicated that experimental manipulations or drugs may differentially affect declarative and non-declarative memory processes. Damage to the medial temporal or the medial thalamic regions is known to result in declarative memory deficits whereas non-declarative memory is largely unaffected by such lesions. Animal research and clinical findings indicate that several components of non-declarative memory such as motor and cognitive skill acquisition or certain types of classical conditioning are dependent upon the integrity of the basal ganglia or the cerebellum. These issues are therefore of increasing importance for the understanding of extrapyramidal and cerebellar diseases. This paper presents recent neuropsychological findings and neuroanatomical data relating to the issue of non-declarative memory.

Animals↗

PubMed2005

Performance of male C57BL/6J mice and Wistar rats in the water maze following various schedules of phencyclidine treatment.

In order to establish an animal model of cognitive impairments relevant to schizophrenia, we set out to obtain an optimal treatment protocol with phencyclidine (PCP) that would lead to robust cognitive impairment with minimal PCP-related adverse effects. Effects of various doses (0.63-5 mg/kg), pre-treatment period (0, 3, 7 and 10 days before the beginning of acquisition) and treatment schedules (before the first or immediately after the last trial on each day) of PCP on the performance of male C57BL/6J mice and Wistar rats in the spatial version of the water maze were studied. In mice, a 10-day pre-treatment period was required to prevent PCP-induced motor impairments, whereas a 3-day pre-treatment was sufficient in rats. PCP impaired spatial learning in both rats and mice, if animals were administered PCP prior to the first trial. The optimal dose was 2.5 mg/kg. In contrast, animals given PCP immediately after the daily training sessions performed as well as controls. Thus, PCP impairs spatial learning in the water maze only when present in the organism. It can be concluded that PCP interferes with learning, and perhaps retrieval, but not consolidation of newly acquired information.

Animals↗