A5073735097- Memory and Neural Mechanisms
- Neuroscience and Neuropharmacology Research
- Neural dynamics and brain function
- Neural and Behavioral Psychology Studies
- Sleep and Wakefulness Research
- Child and Animal Learning Development
- Functional Brain Connectivity Studies
- Dementia and Cognitive Impairment Research
- Alzheimer's disease research and treatments
- Neural Networks and Applications
- Memory Processes and Influences
- Schizophrenia research and treatment
- Neurotransmitter Receptor Influence on Behavior
- Neuroinflammation and Neurodegeneration Mechanisms
- Traumatic Brain Injury and Neurovascular Disturbances
- Parkinson's Disease Mechanisms and Treatments
- Cognitive Science and Mapping
- EEG and Brain-Computer Interfaces
- Neuroscience and Music Perception
- Neurological disorders and treatments
- Autism Spectrum Disorder Research
- Sleep and related disorders
- Glaucoma and retinal disorders
- Receptor Mechanisms and Signaling
- Olfactory and Sensory Function Studies
Rutgers, The State University of New Jersey
2015-2024
Rutgers Sexual and Reproductive Health and Rights
1991-2013
Stanford University
1984-2013
University of Szeged
2008
Columbia University
2007
University of California, Los Angeles
2005
Rütgers (Germany)
2004
National Institute of Advanced Industrial Science and Technology
2002
University of California, San Diego
1994
Amnesic patients and control subjects participated in a study of probabilistic classification learning. In each three tasks, four different cues were probabilistically associated with one two outcomes. On trial, the could appear alone or combination other selected outcome they thought was correct. Feedback provided after trial. task, amnesic learned gradually to associate appropriate at same rate as subjects, improving from 50% correct approximately 65% Presumably because cue-outcome...
Abstract The authors propose a computational theory of the hippocampal region's function in mediating stimulus representations. assumes that region develops new representations enhance discriminability differentially predictive cues while compressing representation redundant cues. Other brain regions, including cerebral and cerebellar cortices, are presumed to use these recode their own In absence an intact hippocmpal region, implies other regions will attempt learn associations using...
The striatum has been widely implicated in cognition, but a precise understanding of its role remains elusive. Here we present converging evidence for the feedback-based learning. In prior functional imaging study, healthy controls showed striatal activity during learning task, which was decreased when same task learned without feedback. show that individuals with dysfunction due to Parkinson's disease are impaired on not non-feedback version task. patients and also used different strategies...
Parkinson's disease is characterized by the degeneration of dopaminergic pathways projecting to striatum. These are implicated in reward prediction. In this study, we investigated and punishment processing young, never-medicated patients, recently medicated patients receiving dopamine receptor agonists pramipexole ropinirole healthy controls. The were also re-evaluated after 12 weeks treatment with agonists. Reward was assessed a feedback-based probabilistic classification task. Personality...
Mesencephalic dopaminergic system (MDS) neurons may participate in learning by providing a prediction error signal to their targets, which include ventral striatal, orbital, and medial frontal regions, as well showing sensitivity the degree of uncertainty associated with individual stimuli. We investigated mechanisms probabilistic classification humans using functional magnetic resonance imaging examine effects feedback uncertainty. The design was optimized for separating neural responses...
Probabilistic category learning is often assumed to be an incrementally learned cognitive skill, dependent on nondeclarative memory systems. One paradigm in particular, the weather prediction task, has been used over half a dozen neuropsychological and neuroimaging studies date. Because of growing interest using this task others like it as behavioral tools for studying neuroscience skill learning, becomes especially important understand how subjects solve kind whether all learn same way. We...
Making appropriate choices often requires the ability to learn value of available options from experience. Parkinson's disease is characterized by a loss dopamine neurons in substantia nigra, hypothesized play role reinforcement learning. Although previous studies have shown that patients are impaired tasks involving learning feedback, they not directly tested widely held hypothesis neuron activity specifically encodes reward prediction error signal used models. To test key this hypothesis,...
Abstract Based on prior animal and computational models, we propose a double dissociation between the associative learning deficits observed in patients with medial temporal (hippocampal) damage versus Parkinson's disease (basal ganglia dysfunction). Specifically, expect that basal dysfunction may result slowed learning, while individuals hippocampal learn at normal speed. However, when challenged transfer task where previously learned information is presented novel recombinations, will...
The category utility hypothesis holds that categories are useful because they can be used to predict the features of instances and tend survive become preferred in a culture (basic-level categories) those best improve users' ability perform this function. Starting from hypothesis, quantitative measure is derived. Application special case substitutive attributes described. successfully basic level applications data hierarchies natural artificial category-learning experiments. relationship...
The purpose of the present study was to gain a deeper understanding role basal ganglia in learning and memory by examining strategies among patients with dysfunction. Using probabilistic category task (the "weather prediction" task) previously shown be sensitive function, authors examined patterns performance during used mathematical models capture different strategies. Results showed that Parkinson's disease exhibit strategy use. Specifically, most controls initially simple, but suboptimal,...
A common misperception within the neural network community is that even with nonlinearities in their hidden layer, autoassociators trained backpropagation are equivalent to linear methods such as principal component analysis (PCA). Our purpose demonstrate nonlinear actually behave differently from and they can outperform these when used for latent extraction, projection, classification. While emulate PCA, thus exhibit a flat or unimodal reconstruction error surface, layer learn domains by...