Search PubMed⌕ Search

SEARCH · Search PubMed

Results for “REINFORCEMENT 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 1,657 records · Page 92Linked to original sources
PubMed1984

Trigeminal denervation and operant behavior in the rat.

Selective section of trigeminal sensory and motor nerves was carried out in two experiments designed to examine the contribution of this sensorimotor system to the control of instrumental responses reinforced with food or water. Unilateral section of either V sensory or V motor nerves had no significant effect on lever pressing. Bilateral section significantly reduced lever pressing, and the deficit was greatest in subjects with trigeminal motor nerve section. Thus trigeminal orosensory or oromotor denervation disrupts performance on a food- or water-reinforced task whose execution does not require a trigeminally mediated response. The results are discussed in terms of motivational and reinforcement accounts of instrumental learning.

Animals↗
PubMed1979

Superstitious bar pressing in hippocampal and septal rats.

Unlike normal animals or those with sham lesions, rats with hippocampal and septal lesions behaved in an operant chamber as if a dependency existed between pellet delivery and their behavior, despite the fact that reinforcement was based on time, not behavior, and was therefore free. This superstitious behavior did not result from a general inability to inhibit responding, as responding rapidly ceased when the pellets were discontinued. These findings suggest that the hippocampus integrates information regarding response-reinforcer relations, which in the normal rat permits superfluous operant behavior to be eliminated.

Animals↗
PubMed1992

Learning performances and aging in cuttlefish (Sepia officinalis).

The use of an associative learning protocol with negative reinforcement allowed us to compare the performances of three groups of animals: young, middle-aged, and senescent. Statistical analysis of the results does not enable us to assert that a significant difference exists between the learning speeds of young and senescent animals. On the other hand however, the latter show poor scores in a retention test carried out 24 h after the learning session. These results suggest that the long-term memory processes are affected during the aging of these animals.

Aging↗
PubMed2002

An emerging molecular and cellular framework for memory processing by the hippocampus.

The hippocampus plays a central role in memory consolidation, a process for converting short-term memory into cortically stored, long-lasting memory in the mammalian brain. Here, we review recent data and discuss the 'synaptic re-entry reinforcement' (SRR) hypothesis, which can account for the role of the hippocampus in memory consolidation at both the molecular and systems levels. The central idea of the SRR hypothesis is that reactivation of neural ensembles in the hippocampus during the consolidation period results in multiple rounds of NMDA-receptor-dependent synaptic reinforcement of the hippocampal memory traces created during initial learning. In addition, such reactivation and reinforcement processes permit the hippocampus to act as a 'coincidence regenerator', providing coordinated input that drives the coherent reactivation of cortical neurons, resulting in the progressive strengthening of cortical memory traces through reactivation of cortical NMDA receptors.

Animals↗
PubMed1985

Drosophila mushroom body mutants are deficient in olfactory learning.

Two Drosophila mutants are described in which the connections between the input to and the output from the mushroom bodies is largely interrupted. In all forms of the flies (larva, imago, male, female) showing the structural defect, olfactory conditioning is impaired. Learning is completely abolished when electroshock is used as reinforcement and partially suppressed in reward learning with sucrose. No influence of the mushroom body defect on the perception of the conditioning stimuli or on spontaneous olfactory behavior is observed. The defect seems not to impair learning of color discrimination tasks or operant learning involving visual cues.

Animals↗
PubMed2002

Interactive computer simulations of knee-replacement surgery.

OBJECTIVE: Current surgical training programs in the United States are based on an apprenticeship model. This model is outdated because it does not provide conceptual scaffolding, promote collaborative learning, or offer constructive reinforcement. Our objective was to create a more useful approach by preparing students and residents for operative cases using interactive computer simulations of surgery. Total-knee-replacement surgery (TKR) is an ideal procedure to model on the computer because there is a systematic protocol for the procedure. Also, this protocol is difficult to learn by the apprenticeship model because of the multiple instruments that must be used in a specific order. We designed an interactive computer tutorial to teach medical students and residents how to perform knee-replacement surgery. We also aimed to reinforce the specific protocol of the operative procedure. Our final goal was to provide immediate, constructive feedback. DESCRIPTION: We created a computer tutorial by generating three-dimensional wire-frame models of the surgical instruments. Next, we applied a surface to the wire-frame models using three-dimensional modeling. Finally, the three-dimensional models were animated to simulate the motions of an actual TKR. The tutorial is a step-by-step tutorial that teaches and tests the correct sequence of steps in a TKR. The student or resident must select the correct instruments in the correct order. The learner is encouraged to learn the stepwise surgical protocol through repetitive use of the computer simulation. Constructive feedback is acquired through a grading system, which rates the student's or resident's ability to perform the task in the correct order. The grading system also accounts for the time required to perform the simulated procedure. We evaluated the efficacy of this teaching technique by testing medical students who learned by the computer simulation and those who learned by reading the surgical protocol manual. Both groups then performed TKR on manufactured bone models using real instruments. Their technique was graded with the standard protocol. The students who learned on the computer simulation performed the task in a shorter time and with fewer errors than the control group. They were also more engaged in the learning process. DISCUSSION: Surgical training programs generally lack a consistent approach to preoperative education related to surgical procedures. This interactive computer tutorial has allowed us to make a quantum leap in medical student and resident teaching in our orthopedic department because the students actually participate in the entire process. Our technique provides a linear, sequential method of skill acquisition and direct feedback, which is ideally suited for learning stepwise surgical protocols. Since our initial evaluation has shown the efficacy of this program, we have implemented this teaching tool into our orthopedic curriculum. Our plans for future work with this simulator include modeling procedures involving other anatomic areas of interest, such as the hip and shoulder.

Arthroplasty, Replacement, Knee↗
PubMed2005

Antennal tactile learning in the honeybee: effect of nicotinic antagonists on memory dynamics.

Restrained worker honeybees (Apis mellifera L.) are able to learn to associate antennal-scanning of a metal plate with a sucrose reinforcement delivered to the mouthparts. Learning occurs reliably in a single association of the two sensory stimuli. The involvement of nicotinic pathways in memory formation and retrieval processes was tested by injecting, into the whole brain through the median ocellus, either mecamylamine (0.6 microg per bee) or alpha-bungarotoxin (2.4 ng per bee). Saline served as a control. Mecamylamine injected 10 min before the retrieval test impairs the retention level tested 3 h and 24 h after single- or multi-trial learning. Retrieval tests performed at various times after the injection show that the blocking effect of mecamylamine lasts about 1 h. The drug has no effect on the reconsolidation or extinction processes. Mecamylamine injected 10 min before conditioning impairs single-trial learning but has no effect on five-trial learning and on the consolidation process. By contrast, alpha-bungarotoxin only impairs the formation of long-term memory (24 h) induced by the five-trial learning and has no effect on medium-term memory (3 h), on single-trial learning or on the retrieval process. Hence, owing to previous data, at least two kinds of nicotinic receptors seem to be involved in honeybee memory, an alpha-bungarotoxin-sensitive and an alpha-bungarotoxin-insensitive receptor. Our results extend to antennal mechanosensory conditioning the role of the cholinergic system that we had previously described for olfactory conditioning in the honeybee. Moreover, we describe here in this insect a pharmacological dissociation between alpha-bungarotoxin sensitive long-term memory and alpha-bungarotoxin insensitive medium-term memory, the last one being affected by mecamylamine.

Animals↗
PubMed2004

Reinforcement-related neurons in the primate basal forebrain respond to the learned significance of task events rather than to the hedonic attributes of reward.

The objective of this study was to determine if the responses of basal forebrain neurons are related to the cognitive processes necessary for the performance of behavioural tasks, or to the hedonic attributes of the reinforcers delivered to the monkey as a consequence of task performance. In all cases, it was found that the primary neuronal responses were to visual stimuli that required analysis important for the selection of a behavioural response and not to the delivery of reinforcement per se. Indeed, reinforcement-related neurons that responded only to the delivery of juice or of saline were never encountered. In additional experiments, it was found that abstract visual cues-specific gestures of the experimenter-that signaled the impending delivery of reinforcement were able to activate these neurons. These data are consistent with the view that reinforcement-related basal forebrain neurons influence the sensory and motor processes in the cerebral cortex, providing control signals that optimise the processing of complex sensory stimuli and/or the generation of appropriate behavioural responses.

Animals↗
PubMed1986

Deficits in spatial-memory tasks following lesions of septal dopaminergic terminals in the rat.

The behavioral effects of 6-hydroxydopamine, injected bilaterally into the lateral septum, were investigated in two tests of spatial memory (radial 8-arm and T-maze). Three different experiments were conducted in the radial maze. In experiment I, rats were permitted to learn the task with food reinforcement in all arms of the maze. In experiment II, retention of the spatial information (working memory) learned in experiment I was tested by interposing various time intervals between choice 4 and 5 of each trial. In experiment III, reference and working memory were simultaneously assessed by only reinforcing 4 choices in the radial maze. Performances were compared in spaced versus massed trials. In the T-maze, the rats were first tested for learning a spatial discrimination between the two arms of the maze, and subsequently for reversal of the previously learned response. The results showed that the rats with lesions were impaired in all experiments. This impairment was particularly marked in some aspects of the procedures used: (1) in the search for the last 4 pellets in experiment I, (2) in the first presentations of various intervals interposed between choices 4 and 5, (3) in the search for food in the baited arms when the trials were massed in experiment III and (4) in the reversal of previously learned spatial discrimination in the T-maze. These behavioral deficits in the rats with septal dopaminergic lesions were interpreted as an increased susceptibility to interference. The lesions were shown to have selectively depleted dopamine concentrations in the septum without damaging noradrenergic terminals or cholinergic cell bodies. It was concluded that dopaminergic neurons could have a modulatory influence on memory processes.

Animals↗
PubMed1992

Context-drug pairings enhance tolerance to ethanol-induced disruption of operant responding.

Much of the research implicating learning in the development of tolerance to ethanol-induced impairment has used an experimental design in which different groups receive drug either before or after an opportunity to perform an instrumental or operant task. The stronger tolerance observed in subjects who perform while intoxicated is most often attributed to the reinforced practice of a learned compensatory response. Using an experimental procedure modeled after Chen (1979), the present study examined an alternative theoretical basis for tolerance in the before-versus-after design. Specifically, the effects of Pavlovian context-drug pairings were assessed under circumstances that precluded reinforced practice of the operant response. Three groups of food-deprived rats were initially trained to barpress for sucrose on an FR15 schedule. After 30 sessions, the bar was retracted and the dipper was covered for a 3-day tolerance acquisition phase. During this phase, each group received an IP injection 15 min before and 45 min after each session. The Paired group received ethanol (1.2 g/kg) before and saline after the session, thus pairing ethanol with cues of the test chamber. The Unpaired group received saline before and ethanol after the session, while the No-Drug group always received saline. During a final test phase, all groups received ethanol (1.5 g/kg) before access to sucrose on the FR schedule. The Paired group completed the first FR15 sequence more rapidly than either control group, indicating that context-ethanol pairings enhanced tolerance to the drug's disruptive effect on the initiation of operant responding.(ABSTRACT TRUNCATED AT 250 WORDS)

Animals↗
PubMed1997

Self-infusion of GABA(A) antagonists directly into the ventral tegmental area and adjacent regions.

This study used an intracerebral self-administration paradigm in rats to determine if blockade of GABA(A) receptors in the ventral tegmental area (VTA) has a reinforcing effect. Rats quickly learned to self-infuse a picrotoxin solution into the anterior VTA; rats discriminated the lever that produced picrotoxin infusions from the lever without consequences; and when the response requirement was increased, rats increased response levels for picrotoxin infusion. The reinforcing effect of picrotoxin was site-specific: Anterior VTA regions supported vigorous self-infusions, but not the posterior VTA, substantia nigra, or lateral hypothalamus. Muscimol, a GABA(A) agonist, disrupted picrotoxin self-infusion, but bicuculline, a GABA(A) antagonist, was self-infused into the VTA. The results suggest that blockade of GABA(A) receptors in the anterior VTA is reinforcing and that functional organization of the GABA systems within the VTA is heterogeneous.

Animals↗
PubMedDate not supplied

Reinforcement concept in investigations on simple nervous systems.

An analysis of the applicability of the concept of reinforcement to the studies of learning in simple nervous systems of invertebrates is made. Analysis of the literature data and my own results suggests that reinforcement cannot be regarded as an independent behavioral phenomenon. A description of reinforcement as a state of the nervous system which precedes long-term changes of behavior is given. Using the example of aversive conditioning to food in gastropod snails it is shown that a state of the network that can be correlated with the state of reinforcement can be elicited in the simple nervous system by activation of serotonergic pedal cells modulating avoidance behavior of the animal. The conclusion is made that with certain limitations the reinforcement concept can be used in studies on simple nervous systems.

Animals↗
PubMedDate not supplied

Improvements in the selective perception and training of rats using an original analog of the C-terminal fragment of vasopressin.

This report presents studies on the effects of intranasal administration of five doses (0.001, 0.01, 0.1, 1, and 10 microg/kg) of a new analog of arginine-vasopressin fragment AVP(6-9), i.e., D-MPRG, on the learning ability of rats with positive and negative reinforcement. All doses of the peptide improved learning. The most effective dose was 0.01 microg/kg, at which the peptide accelerated the acquisition of a conditioned active avoidance reflex both when given 1 h before and when given immediately after training sessions. The peptide had greater effects when animals were trained with negative reinforcement. Analysis of the results suggests that the action of D-MPRG is mainly on perception processes, i.e., extraction of the conditioned stimulus from the environmental surroundings and evaluation and enhancement of its biological significance. In addition, this peptide prevented extinction of the acquired habit and improved the processes of consolidation, though this effect was weaker than its effect on perception.

Administration, Intranasal↗