Search PubMed⌕ Search

SEARCH · Search PubMed

Results for “representation learning”

Search indexed PubMed citations on genomics, clinical trials, systematic reviews and public health. Explore titles, authors and supplied subject terms, then open the PubMed record.

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

At least 1,099 records · Page 61Linked to original sources
PubMed1997

Influence of experience on the representation of the "mothering call" in frontoparietal and auditory cortex of pups of the rodent Octodon degus: FDG mapping.

Several types and subtypes of vocalizations which have a behavioral impact on degu pups were identified. Among these the complex "mothering call" which is exclusively uttered by females and first during extensive nursing periods in the nest is a candidate for filial learning. In 14C-2-fluoro-2-deoxyglucose (FDG) experiments two-weeks-old pups raised by normal mothers showed higher metabolic activity in somatosensory frontoparietal and frontal cortex upon play back of a mothering call than pups raised by muted mothers. It is suggested that pups learn to associate the mothering call with close body contact with their mother early in life. In addition, FDG representation of the call, of its components and of tone and noise stimuli were studied in degu auditory cortex. Five fields and some aspects of tonotopic organization were identified. The mothering call activated all fields, but with more spatial extent of labeling in normally raised pups. A rostral field was activated by play-back of the mothering call, noise, and two-tone sequences, but hardly by single-frequency tones and the narrow-band component of the mothering call.

Acoustic Stimulation↗
PubMed2004

Climbing neuronal activity as an event-based cortical representation of time.

The brain has the ability to represent the passage of time between two behaviorally relevant events. Recordings from different areas in the cortex of monkeys suggest the existence of neurons representing time by increasing (climbing) activity, which is triggered by a first event and peaks at the expected time of a second event, e.g., a visual stimulus or a reward. When the typical interval between the two events is changed, the slope of the climbing activity adapts to the new timing. We present a model in which the climbing activity results from slow firing rate adaptation in inhibitory neurons. Hebbian synaptic modifications allow for learning the new time interval by changing the degree of firing rate adaptation. This event-based representation of time is consistent with Weber's law in interval timing, according to which the error in estimating a time interval is proportional to the interval length.

Action Potentials↗
PubMed2006

Induction of neocortical long-term depression results in smaller movement representations, fewer excitatory perforated synapses, and more inhibitory synapses.

Long-term depression (LTD) is one of the most widely investigated models of the synaptic mechanisms underlying learning and memory. Previous research has shown that induction of LTD in the neocortex decreases measures of pyramidal cell dendritic morphology in both layers III and V. Here, we investigated the effects of LTD induction on 1) the time course of recovery of synaptic efficacy, 2) movement representations, 3) cortical thickness and layer V neuron density, and 4) the density of excitatory and inhibitory synapses in layer V of sensorimotor neocortex. Rats carried a stimulating electrode in the midline corpus callosum and a recording electrode in the right sensorimotor neocortex. Each rat received either low-frequency stimulation composed of 900 pulses at 1 Hz or handling daily for a total of 20-25 days. Callosal-neocortical evoked potentials were recorded in the right hemisphere before and after stimulation or handling. Our results show that LTD induction lasts for 3 weeks and results in smaller motor maps of the caudal forelimb area. We did not observe any reduction in neocortical thickness or neuron density. There was a reduction in the density of excitatory perforated synapses and an increase in the density of inhibitory synapses in layer V of the sensorimotor neocortex, thereby providing a general mechanism for the reduction in motor map size. This study sheds light on the interaction between an artificial model of learning, receptive field characteristics, and synaptic number in the sensorimotor cortex.

Animals↗
PubMed2005

Distinguishable brain activation networks for short- and long-term motor skill learning.

The acquisition of a new motor skill is characterized first by a short-term, fast learning stage in which performance improves rapidly, and subsequently by a long-term, slower learning stage in which additional performance gains are incremental. Previous functional imaging studies have suggested that distinct brain networks mediate these two stages of learning, but direct comparisons using the same task have not been performed. Here we used a task in which subjects learn to track a continuous 8-s sequence demanding variable isometric force development between the fingers and thumb of the dominant, right hand. Learning-associated changes in brain activation were characterized using functional MRI (fMRI) during short-term learning of a novel sequence, during short-term learning after prior, brief exposure to the sequence, and over long-term (3 wk) training in the task. Short-term learning was associated with decreases in activity in the dorsolateral prefrontal, anterior cingulate, posterior parietal, primary motor, and cerebellar cortex, and with increased activation in the right cerebellar dentate nucleus, the left putamen, and left thalamus. Prefrontal, parietal, and cerebellar cortical changes were not apparent with short-term learning after prior exposure to the sequence. With long-term learning, increases in activity were found in the left primary somatosensory and motor cortex and in the right putamen. Our observations extend previous work suggesting that distinguishable networks are recruited during the different phases of motor learning. While short-term motor skill learning seems associated primarily with activation in a cortical network specific for the learned movements, long-term learning involves increased activation of a bihemispheric cortical-subcortical network in a pattern suggesting "plastic" development of new representations for both motor output and somatosensory afferent information.

Adult↗
PubMed2006

A dual-process account of digit invariance learning.

Performance in the McGeorge and Burton (1990) digit invariance task was originally thought to be mediated by unconscious abstraction of a "rule" that identified the invariant feature across all study items. Subsequent explanations have suggested explicit strategy use or similarity-to-exemplar matching rather than abstraction. This paper presents data that suggest that both similarity and abstraction can be used under different task demands. Delay between study and test afforded abstraction of the invariant knowledge whereas reducing the pool of study exemplars enhanced responding based on specific similarity. These results parallel effects found in the categorization literature. Rule abstraction in this sense may be due to statistical learning of feature frequency rather than abstraction of a central tendency or a complex/conceptual rule. Categorizing responses into subjective memory states (remember/know/guess) demonstrates that neither the similarity matching nor the abstraction mechanism uses information from episodic memory. Confidence measures show that participants are more confident of responses when the prototypical representation is used but not specific similarity. Taken together, these data suggest that abstracted knowledge is not held consciously but that participants have meta-awareness of when they are using the abstracted representation.

Analysis of Variance↗
PubMed2001

Exploration, anxiety, and spatial memory in transgenic anophthalmic mice.

Contradictory results are found in the literature concerning the role of vision in the perception of space or in spatial navigation, in part because of the lack of murine models of total blindness used so far. The authors evaluated the spatial abilities of anophthalmic transgenic mice. These mice did not differ qualitatively from their wild-type littermates in general locomotor activity, spontaneous alternation, object exploration, or anxiety, but their level of exploratory activity was generally lower. In the spatial version of the water maze, they displayed persistent thigmotaxic behavior and showed severe spatial learning impairments. However, their performances improved with training, suggesting that they may have acquired a rough representation of the platform position. These results suggest that modalities other than vision enable some degree of spatial processing in proximal and structured spaces but that vision is critical for accurate spatial navigation.

Animals↗
PubMed2004

SINBAD: a neocortical mechanism for discovering environmental variables and regularities hidden in sensory input.

We propose that a top priority of the cerebral cortex must be the discovery and explicit representation of the environmental variables that contribute as major factors to environmental regularities. Any neural representation in which such variables are represented only implicitly (thus requiring extra computing to use them) will make the regularities more complex and therefore more difficult, if not impossible, to learn. The task of discovering such important environmental variables is not an easy one since their existence is only indirectly suggested by the sensory input patterns the cortex receives--these variables are "hidden". We present a candidate computational strategy for (i) discovering regularity-simplifying environmental variables, (ii) learning the regularities, and (iii) using regularities in perceptual and decision-making tasks. The SINBAD computational model discovers useful environmental variables through a search for different, but nevertheless highly correlated, functions of any kind over nonoverlapping subsets of the known variables, this being indicative of some important environmental variable that is responsible for the correlation. We suggest that such a search is performed in the neocortex by the dendritic trees of individual pyramidal cells. According to the SINBAD model, the basic function of each pyramidal cell is to (i) discover and represent one of the regularity-simplifying environmental variables and (ii) learn to infer the state of its variable from the states of other variables, represented by other pyramidal cells. A network of such cells--each cell just attending to representation of its variable--can function as a sophisticated and useful inferential model of the outside world.

Association Learning↗
PubMedDate not supplied

Analysis of respiratory pressure-volume curves in intensive care medicine using inductive machine learning.

We present a case study of machine learning and data mining in intensive care medicine. In the study, we compared different methods of measuring pressure-volume curves in artificially ventilated patients suffering from the adult respiratory distress syndrome (ARDS). Our aim was to show that inductive machine learning can be used to gain insights into differences and similarities among these methods. We defined two tasks: the first one was to recognize the measurement method producing a given pressure-volume curve. This was defined as the task of classifying pressure-volume curves (the classes being the measurement methods). The second was to model the curves themselves, that is, to predict the volume given the pressure, the measurement method and the patient data. Clearly, this can be defined as a regression task. For these two tasks, we applied C5.0 and CUBIST, two inductive machine learning tools, respectively. Apart from medical findings regarding the characteristics of the measurement methods, we found some evidence showing the value of an abstract representation for classifying curves: normalization and high-level descriptors from curve fitting played a crucial role in obtaining reasonably accurate models. Another useful feature of algorithms for inductive machine learning is the possibility of incorporating background knowledge. In our study, the incorporation of patient data helped to improve regression results dramatically, which might open the door for the individual respiratory treatment of patients in the future.

Adult↗
PubMed2004

Implicit sequence learning: effects of level of structure, adult age, and extended practice.

The influence of structure and age on sequence learning was investigated by testing 24 young and 24 older participants for 10 sessions in an alternating serial response time task in which pattern trials alternated with random trials. Individuals encountered lag-2 or lag-3 structure, and learning was measured by the difference (in response time and accuracy) between pattern and random trials. Both ages learned lag-2 structure, but the young learned more than the older participants. Only the young people learned lag-3 structure, and they did so more slowly and to a lesser degree than they learned lag-2 structure. These age deficits in higher order sequence learning after extended practice are consistent with simultaneity theory and with theories positing that age-related deficits in neuromodulation lead to less distinctive representations.

Adult↗
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↗
PubMed2000

Object identification as a function of discriminability and learning presentations: the effect of stimulus similarity and canonical frame alignment on aircraft identification.

Aircraft that were relatively similar (homogeneous) and relatively dissimilar (heterogeneous) in appearance were studied at orientations either consistent (canonical) or inconsistent (noncanonical) with the environmental frame of reference. At test, participants' identification performance was measured with stimuli rotated to novel orientations within the picture plane. During learning and testing, identification of heterogeneous aircraft was better than that of homogeneous aircraft. At test, only identification of homogeneous aircraft revealed a strong linear degradation of performance as angular disparity between the novel test orientations and the original learning orientations increased. During learning and testing, identification was better for aircraft studied at canonical orientations than for those studied at noncanonical orientations. The results are discussed in terms of object identification, aircraft recognition training, categorization, mental representations, and visual mental rotation.

Adolescent↗
PubMed2002

Dynamic cortical and subcortical networks in learning and delayed recall of timed motor sequences.

We used positron emission tomography to examine learning and retention of timed motor sequences. Subjects were scanned during learning (LRN) and baseline (ISO) on 3 d: day 1, after 5 d of practice (day 5) and after a 4 week delay (recall). Blood flow was compared across days of learning and between the LRN and ISO conditions. Overall, significant changes in activity were seen across days for the LRN condition, but not the ISO baseline. Day 1 results revealed extensive activation in the cerebellar cortex, particularly lobules III/IV and VI. Day 5 results showed increased activity in the basal ganglia (BG) and frontal lobe, with no significant cerebellar activity. At recall, significantly greater activity was seen in M1, premotor, and parietal cortex. Blood flow in the cerebellum decreased significantly between day 1 and recall. These results reveal a dynamic network of motor structures that are differentially active during different phases of learning and delayed recall. For the first time our findings show that recall of motor sequences in humans is mediated by a predominantly cortical network. Based on these results, we suggest that during early learning cerebellar mechanisms are involved in adjusting movement kinematics according to sensory input to produce accurate motor output. Thereafter, the cerebellar mechanisms required for early learning are no longer called into play. During late learning, the BG may be involved in automatization. At delayed recall, movement parameters appear to be encoded in a distributed representation mediated by M1, premotor, and parietal cortex.

Adult↗
PubMed1995

Exploring memory functions by means of brain electrical topography: a review.

A series of experiments is reviewed which explored whether the functional brain state of long-term memory retrieval is correlated with specific changes in slow, DC-like event-related brain potentials. The main results are: (1) Retrieving associations from long-term memory is accompanied by a slow negative shift of 5-10 microV which prevails about as long as the retrieval process lasts, i.e., in our experiments, for a period of several seconds: (2) When different types of representations have to be reactivated in memory the slow negative wave shows a clearly distinct topography. The maximum was found in a verbal condition over the left frontal, in a spatial condition over the parietal, and in a color condition over the right occipital to temporal cortex. All these conditions were completely equivalent with respect to the established associative structure, the learning procedure, and the performance criterion. (3) The amplitude of the topographic maximum increases with the number of representations which have to be reactivated. This effect is not due to a non-specific increase of effort but specifically related to the number of activated episodic memory contents which had been experimentally established. In contrast, the reactivation of a priori given semantic association did not become manifest in a specific slow wave effect. These findings are compatible with the idea that memory retrieval implies a reactivation of those cortical cell assemblies in the cortex in which the constituting features of a mnestic entity had originally been processed during perception and learning. The results are also discussed with respect to the possible advantages of EEG and MEG recordings for a cognitive psychophysiology in comparison to other brain imaging techniques as PET or fMRI.

Brain↗
PubMed1986

Simple neural models of classical conditioning.

A systematic study of the necessary and sufficient ingredients of a successful model of classical conditioning is presented. Models are constructed along the lines proposed by Gelperin, Hopfield, and Tank, who showed that many conditioning phenomena could be reproduced in a model using non-trivial distributed representations of the sensory stimuli. The additional phenomena of extinction and blocking are found to be obtainable by generalizing the Hebbian learning algorithm, rather than by additional complications in the hardware. The most successful algorithms have a minimal number of adjustable parameters, and require only local-time information about the level of postsynaptic activity. The proper behavior of these algorithms is verified by both simple analytic arguments and by direct numerical simulation. Certain detailed assumptions concerning the distributed sensory representations are also found to have a surprising degree of importance.

Algorithms↗
PubMedDate not supplied

Cognitive processes in extinction.

Human conditioning research shows that learning is closely related to consciously available contingency knowledge, requires attentional resources, and is influenced by language. This research suggests a cognitive model in which extinction consists of changes in contingency beliefs in long-term memory. Laboratory and clinical evidence on extinction is briefly reviewed, and it is concluded that the evidence supports the cognitive position. There is little evidence for a separate, noncognitive conditioning system. The primary implication for neural analysis is that learning and extinction are unlikely to be reducible to direct connections in which one stimulus simply activates or inhibits the memory representation of another. Rather, an adequate neural model will involve the integration of both low-level and high-level systems, including attention, representation of stimulus relations in long-term memory, and a dynamic performance mechanism based on anticipation, not just activation.

Animals↗
PubMed2005

A refined model of sleep and the time course of memory formation.

Research in the neurosciences continues to provide evidence that sleep plays a role in the processes of learning and memory. There is less of a consensus, however, regarding the precise stages of memory development during which sleep is considered a requirement, simply favorable, or not important. This article begins with an overview of recent studies regarding sleep and learning, predominantly in the procedural memory domain, and is measured against our current understanding of the mechanisms that govern memory formation. Based on these considerations, I offer a new neurocognitive framework of procedural learning, consisting first of acquisition, followed by two specific stages of consolidation, one involving a process of stabilization, the other involving enhancement, whereby delayed learning occurs. Psychophysiological evidence indicates that initial acquisition does not rely fundamentally on sleep. This also appears to be true for the stabilization phase of consolidation, with durable representations, resistant to interference, clearly developing in a successful manner during time awake (or just time, per se). In contrast, the consolidation stage, resulting in additional/enhanced learning in the absence of further rehearsal, does appear to rely on the process of sleep, with evidence for specific sleep-stage dependencies across the procedural domain. Evaluations at a molecular, cellular, and systems level currently offer several sleep specific candidates that could play a role in sleep-dependent learning. These include the upregulation of select plasticity-associated genes, increased protein synthesis, changes in neurotransmitter concentration, and specific electrical events in neuronal networks that modulate synaptic potentiation.

Adaptation, Physiological↗
PubMed1989

Varieties of size-specific visual selection.

Compared time to evaluate stimuli of varying sizes. When Ss expect an upcoming stimulus to be a certain size, response time increases with the disparity between expected and actual size. There are, however, 2 size adjustment processes, and they reflect 2 types of visual selection. In the first, a shape-specific image representation is used to separate a visual object from a superimposed distractor. These representations require the type of slow size scaling demonstrated in imagery experiments. The second size scaling process is faster and not shape-specific. At any given time the visual system is set to process information at a particular scale, and that scale can be adjusted to match an object's size. Because both selection mechanisms depend on size, they probably occur at a relatively low, spatially organized processing level. These findings lead to a new explanation for results that had been taken as evidence for attentional selection at the level of object representations.

Attention↗
PubMed1992

Memory in the chick: multiple cues, distinct brain locations.

Training chicks on a one-trial passive avoidance task results in memory-dependent synaptic remodeling in the intermediate medial hyperstriatum ventrale (IMHV) and lobus parolfactorius (LPO). Because pretraining IMHV lesions are amnestic and posttraining IMHV lesions are not, the functional significance of this remodeling requires explanation. Chicks use various cues to classify and remember objects. If the IMHV were concerned with memory for only one such cue, then posttraining IMHV lesions would not lead to "amnesia" because animals would still avoid the aversive bead using other contextual cues. This hypothesis was tested using a color discrimination task. IMHV lesions, but not LPO lesions, impair color discrimination, suggesting that the IMHV may be involved in classifying and remembering the bitter bead on the basis of color. Thus, even simple associations are stored in the brain in the form of multiple, dispersed representations.

Animals↗