T-maze Training Research Articles

Gestational choline supplementation ameliorates T-maze reversal learning deficits induced by first trimester binge alcohol exposure in weanling lambs

In rodent models of fetal alcohol spectrum disorders (FASD), cognitive deficits are implicated in impaired T-maze spatial reversal learning. Rat studies have indicated supplemental administration of choline during the developmental period of alcohol exposure can ameliorate spatial reversal deficits. This study tested whether beneficial effects of prenatal choline supplementation could be confirmed in a sheep model of binge exposure in the first trimester equivalent. Two hypotheses were tested: 1) alcohol exposure would produce deficits in reversal of a T-maze position discrimination; and 2) gestational dietary supplementation of choline would ameliorate those deficits. Mated ewes were assigned to one of seven groups—a normal control (NC) group or one of six infusion treatment groups: saline control (SC; isotonic saline), saline control plus choline (SC-CH; isotonic saline plus choline, 10 mg/kg administered orally throughout each day of gestation), binge alcohol (BA; 1.75 g/kg alcohol per infusion day), binge alcohol plus choline (BA-CH; 1.75 g/kg/day alcohol plus choline), heavy binge alcohol (HBA; 2.5 g/kg/day alcohol), or heavy binge alcohol plus choline (HBA-CH; 2.5 g/kg/day alcohol plus choline). The alcohol infusions modeled a weekend binge drinking pattern over the first trimester-equivalent (gestational day 4-41). T-maze training began at 12 weeks of age, with daily sessions occurring 5 days/week. Lambs were given five days of habituation training, followed by five days of position discrimination training (3 trials per daily session, intertrial interval of 3 hours, reinforced side randomly assigned across subjects). Lambs were then given 10 days of training on the reversal task. There was no difference among groups during acquisition. Alcohol impaired reversal learning, and choline supplementation mitigated these deficits in the HBA-CH group. These results suggest that maternal dietary choline supplementation can ameliorate or prevent some impairments of executive function in a sheep model of FASD.

The effect of copper on behaviour, memory, and associative learning ability of zebrafish (Danio rerio)

Copper is an essential element in many biological processes, but may exert toxic effects at levels surplus to metabolic requirements. Herein we assess the effect of copper on zebrafish behaviour using two assays, namely the novel tank diving test and a T-maze test with food reward. Novel tank diving tests were conducted on days 0, 4, and 10 of a 10 day Cu exposure (at concentrations of 0.77 μM (25% of the 240 h LC50) and 1.52 μM (50% of the 240 h LC50) to assess the alterations of behavioural responses in repeating novel tank diving assays and the effect of Cu on these patterns. Results demonstrate habituation to novelty, which is an indicator of spatial memory. Copper exposure had no effect on the latency of entry into the upper zones of the tank, nor on the total time spent therein, but did cause a greater number of freezing bouts in comparison to the control group. Additionally, Cu exposure had no effect on the habituation responses of zebrafish. Using the T-maze assay, we tested the effect of prior exposure to Cu for 10 days on subsequent behavioural trainings. The T-maze protocol was based on associative learning, where a visual stimulus (colour) was linked with a natural stimulus (food). Results of the control group showed that zebrafish are able to perform associative learning tasks. Moreover, Cu was found to negatively affect the associative learning capabilities. Specifically, while zebrafish in the control group achieved a significant number of correct choices (leading to food reward) throughout the T-maze training, such a trend was not observed for Cu exposed fish. Thus at the exposure concentrations and exposure times considered herein, Cu has no determinative impact on instinctual behavioural responses of zebrafish in repeated novel tank diving assays but does limit the associative learning capabilities.

Efficacy of nonselective optogenetic control of the medial septum over hippocampal oscillations: the influence of speed and implications for cognitive enhancement.

Optogenetics holds great promise for both the dissection of neural circuits and the evaluation of theories centered on the temporal organizing properties of oscillations that underpin cognition. To date, no studies have examined the efficacy of optogenetic stimulation for altering hippocampal oscillations in freely moving wild‐type rats, or how these alterations would affect performance on behavioral tasks. Here, we used an AAV virus to express ChR2 in the medial septum (MS) of wild‐type rats, and optically stimulated septal neurons at 6 Hz and 30 Hz. We measured the corresponding effects of these stimulations on the oscillations of the MS and hippocampal subfields CA1 and CA3 in three different contexts: (1) With minimal movement while the rats sat in a confined chamber; (2) Explored a novel open field; and (3) Learned and performed a T‐maze behavioral task. While control yellow light stimulation did not affect oscillations, 6‐Hz blue light septal stimulations altered hippocampal theta oscillations in a manner that depended on the animal's mobility and speed. While the 30 Hz blue light septal stimulations only altered theta frequency in CA1 while the rat had limited mobility, it robustly increased the amplitude of hippocampal signals at 30 Hz in both regions in all three recording contexts. We found that animals were more likely to make a correct choice during Day 1 of T‐maze training during both MS stimulation protocols than during control stimulation, and that improved performance was independent of theta frequency alterations.

Multimodal Plasticity in Dorsal Striatum While Learning a Lateralized Navigation Task.

Growing evidence supports a critical role for the dorsal striatum in cognitive as well as motor control. Both lesions and in vivo recordings demonstrate a transition in the engaged dorsal striatal subregion, from dorsomedial to dorsolateral, as skill performance shifts from an attentive phase to a more automatic or habitual phase. What are the neural mechanisms supporting the cognitive and behavioral transitions in skill learning? To pursue this question, we used T-maze training during which rats transition from early, attentive (dorsomedial) to late habitual (dorsolateral) performance. Following early or late training, we performed the first direct comparison of bidirectional synaptic plasticity in striatal brain slices, and the first evaluation of striatal synaptic plasticity by hemisphere relative to a learned turn. Consequently, we find that long-term potentiation and long-term depression are independently modulated with learning rather than reciprocally linked as previously suggested. Our results establish that modulation of evoked synaptic plasticity with learning depends on striatal subregion, training stage, and hemisphere relative to the learned turn direction. Exclusive to the contralateral hemisphere, intrinsic excitability is enhanced in dorsomedial relative to dorsolateral medium spiny neurons early in training and population responses are dampened late in training. Neuronal reconstructions indicate dendritic remodeling after training, which may represent a novel form of pruning. In conclusion, we describe region- and hemisphere-specific changes in striatal synaptic, intrinsic, and morphological plasticity which correspond to T-maze learning stages, and which may play a role in the cognitive transition between attentive and habitual strategies. Significance statement: We investigated neural plasticity in dorsal striatum from rats that were briefly or extensively trained on a directional T-maze task. Our results demonstrate that both the extent of training and the direction a rat learns to turn control the location and type of change in synaptic plasticity. In addition, brief training produces changes in neuron excitability only within one striatal subregion, whereas all training produces widespread changes in dendritic morphology. Our results suggest that activity in dorsomedial striatum strengthens the rewarded turn after brief training, whereas activity in dorsolateral striatum suppresses unrewarded turns after extensive training. This study illuminates how plasticity mediates learning using a task recognized for transitioning subjects from attentive to automatic performance.

Working memory deficits and alterations of ERK and CREB phosphorylation following withdrawal from cocaine self-administration.

The mechanisms underlying memory functions during withdrawal from the chronic drug use are poorly understood. We assessed learning and spatial working memory using the delayed alternation assay (T-maze) in rats, previously subjected to cocaine self-administration. The T-maze training was conducted 1-5 weeks after cocaine cessation; working memory efficacy was assessed at 5-8 weeks of drug withdrawal. After behavioral training and testing, the rats were sacrificed and the levels of p-CREB/CREB and p-ERK2/ERK2 in several brain areas were measured. The same molecular assessment was performed in rats with cocaine injections, but forced to drug abstinence in home cages. After 5 weeks of cocaine withdrawal from self-administration, a significant impairment of working memory under increased working memory load (inter-trial delay extended to 30s), with no changes at baseline conditions (inter-trial delay 10s), was noticed. Neither acquisition phase nor working memory performance measured 6-8 weeks after the last drug intake differed between cocaine or saline pretreated rats. Upon T-maze training and 8-week withdrawal, cocaine-pretreated rats had higher levels of p-CREB/CREB in prefrontal cortex and dorsal striatum and lower in hippocampus compared to saline rats. Increased levels of p-ERK2/ERK2 were observed in dorsal striatum, hippocampus and decreased in nucleus accumbens. In cocaine-pretreated caged rats no changes in p-CREB/CREB levels were observed, while ERK2 levels either decreased (frontal cortex) or increased (nucleus accumbens). Our results suggest that cocaine self-administration results in cognitive impairments and alterations in ERK/CREB signaling pathway long after discontinuation of drug use.

Learning-induced alterations in prefrontal cortical dendritic morphology

The influences of complex housing, T-maze, and Grice box training on dendritic morphology of the prefrontal cortex (PFC) and primary somatosensory (Par 1) were investigated in the rat. Golgi–Cox analyses demonstrated that all learning paradigms produced alterations in PFC connectivity, albeit differently. Furthermore, the effects of experience on dendritic morphology varied by region, hemisphere, and lamina and often resulted in opposing changes within each. For example, complex housing produced a time-dependent decrease in the dendritic fields of layer (L) V neurons in region 3 of the cingulate cortex (Cg3) in the medial PFC (mPFC) neurons and increases in LIII of the dorsal agranular insular (AID) region of the orbitofrontal cortex. In contrast, T-maze training produced increases in Cg3 LV and decreases in AID LIII spine density. Of interest, the influence of experience was not reflected equally between hemispheres. For example, T-maze training produced an increase in Cg3 LV and AID LIII branch order and length in the right hemisphere, and an increase in Cg3 LIII branch order in the left hemisphere. Additionally, correlation analysis of task performance and dendritic morphology indicated an opposing influence of experience within the different laminae of the Cg3. For example, whereas, performance was negatively correlated with Cg3 LV dendritic branch order/length, there was a positive correlation with Cg3 LIII dendrites. Complex housing induced changes in the Par 1were also time -dependent but were only apparent after a prolonged period of exposure. In addition to demonstrating learning-specific modifications in dendritic connectivity within the PFC, the results illustrate varying patterns of change that likely reflect task-dependent requirements on the PFC.

Memory‐related changes in L‐citrulline and agmatine in the rat brain

L-citrulline, L-ornithine, and agmatine are the metabolites of L-arginine by nitric oxide synthase (NOS), arginase, and arginine decarboxylase (ADC), respectively. In contrast to the NOS and arginase pathways, the role of the ADC-agmatine pathway in learning and memory has only been paid attention lately. Recent evidence suggests a potential involvement of agmatine in learning and memory processing. The present study further addressed this issue by comparing the levels of agmatine, as well as L-arginine, L-citrulline, and L-ornithine, in the hippocampus, parahippocampal region, prefrontal cortex, vestibular nucleus, and cerebellum in rats that were trained in the delayed nonmatch to position task in the T-maze with their yoked controls. There were significantly increased agmatine levels in the prefrontal, entorhinal, and perirhinal cortices and increased L-citrulline concentrations in the dentate gyrus (DG) and prefrontal cortex in the T-maze training group relative to the control one. L-arginine and L-ornithine levels were not significantly different between groups in the brain regions examined. These results demonstrate T-maze training-induced region-specific increases in L-citrulline and agmatine. Significant positive correlations between prefrontal and perirhinal agmatine levels and animals' performance in the T-maze further suggest the direct involvement of agmatine in learning and memory processing.

Long‐term deficits on a foraging task after bilateral vestibular deafferentation in rats

Animal studies have shown that bilateral vestibular deafferentation (BVD) causes deficits in spatial memory that may be related to electrophysiological and neurochemical changes in the hippocampus. Recently, human studies have also indicated that human patients can exhibit spatial memory impairment and hippocampal atrophy even 8-10 yr following BVD. Our previous studies have shown that rats with unilateral vestibular deafferentation (UVD) showed an impairment at 3 months after the surgery on a food foraging task that relies on hippocampal integration of egocentric cues, such as vestibular information; however, by 6 months postop, they showed a recovery of function. By contrast, the long-term effects of BVD on spatial navigation have never been well studied. In this study, we tested BVD or sham rats on a food foraging task at 5 months postop. Under light conditions, BVD rats were able to use visual cues to guide themselves home, but did so with a significantly longer homing time. However, in darkness, BVD rats were severely impaired in the foraging task, as indicated by a significantly longer homing distance and homing time, with more errors and larger heading angles when compared with sham rats. These results suggest that, unlike UVD, BVD causes long-term deficits in spatial navigation that are unlikely to recover, even with repeated T-maze training.

Microdialysis measures of functional increases in ACh release in the hippocampus with and without inclusion of acetylcholinesterase inhibitors in the perfusate

Because brain extracellular acetylcholine (ACh) levels are near detection limits in microdialysis samples, an acetylcholinesterase (AChE) inhibitor such as neostigmine is often added to microdialysis perfusates to increase ACh levels in the dialysate, a practice that raises concerns that the inhibitor might alter the results. Two experiments compared functional differences in ACh release with and without neostigmine. In the first experiment, 30-60% increases in extracellular ACh concentrations in the hippocampus were evident during food-rewarded T-maze training with 20-500 nm neostigmine in the perfusate but no increases were seen without neostigmine. In the second experiment, 78% increases in ACh release in the hippocampus were seen after injections of the GABA(A) receptor antagonist, bicuculline, into medial septum only if neostigmine (50 nm) was included in the perfusate. These findings suggest that, in the hippocampus, endogenous brain AChEs are very efficient at removing extracellular ACh, obscuring differences in ACh release in these experiments. Therefore, inclusion of AChE inhibitors in the microdialysis perfusate may be necessary under some conditions for observations of functional changes in release of ACh in the hippocampus.

Effects of leptin on memory processing

Leptin is a peptide hormone secreted by adipose tissue. Studies have shown that leptin crosses the blood–brain barrier (BBB) by a saturable transport system where it acts within the hypothalamus to regulate food intake and energy expenditure. Leptin also acts in the hippocampus where it facilitates the induction of long-term potentiation and enhances NMDA receptor-mediated transmission. This suggests that leptin plays a role in learning and memory. Obese mice and rats, which have leptin receptor deficiency, have impaired spatial learning. In disease states such as diabetes, humans and animals develop leptin resistance at the BBB. This suggests that low leptin levels in the brain may be involved in cognitive deficits associated with diabetes. In the current study, the effects of leptin on post-training memory processing in CD-1 mice were examined. Mice were trained in T-maze footshock avoidance and step down inhibitory avoidance. Immediately after training, mice received bilateral injections of leptin into the hippocampus. Retention was tested 1 week later in the T-maze and 1 day later in step down inhibitory avoidance. Leptin administration improved retention of T-maze footshock avoidance and step down inhibitory avoidance. Leptin administered 24 h after T-maze training did not improve retention when tested 1 week after training. SAMP8 mice at 12 months of age have elevated amyloid-beta protein and impaired learning and memory. We examined the effect of leptin on memory processing in the hippocampus of 4 and 12 months old SAMP8 mice. Leptin improved retention in both 4 and 12 months old SAMP8 mice; 12 month SAMP8 mice required a lower dose to improve memory compared to 4 months SAMP8 mice. The current results indicate that leptin in the hippocampus is involved in memory processing and suggests that low levels of leptin may be involved in cognitive deficits seen in disease states where leptin transport into the CNS is compromised.

Glucose injections into the dorsal hippocampus or dorsolateral striatum of rats prior to T-maze training: Modulation of learning rates and strategy selection

The present experiments examined the effects of injecting glucose into the dorsal hippocampus or dorsolateral striatum on learning rates and on strategy selection in rats trained on a T-maze that can be solved by using either a hippocampus-sensitive place or striatum-sensitive response strategy. Percentage strategy selection on a probe trial (P(crit)) administered after rats achieved criterion (nine of 10 correct choices) varied by group. All groups predominately exhibited a response strategy on a probe trial administered after overtraining, i.e., after 90 trials. In experiment 1, rats that received intrahippocampal glucose injections showed enhanced acquisition of the T-maze and showed increased use of response solutions at P(crit) compared with that of unimplanted and artificial cerebral spinal fluid (aCSF)-treated groups. These findings suggest that glucose enhanced hippocampal functions to accelerate the rate of learning and the early adoption of a response strategy. In experiment 2, rats that received intrastriatal glucose injections exhibited place solutions early in training and reached criterion more slowly than did aCSF controls, with learning rates comparable to those of unoperated and operated-uninjected controls. Relative to unoperated, operated-uninjected and glucose-injected rats, rats that received intrastriatal aCSF injections showed enhanced acquisition of the T-maze and increased use of response solutions at P(crit). The unexpected enhanced acquisition seen after striatal aCSF injections suggests at least two possible interpretations: (1) aCSF impaired striatal function, thereby releasing competition with the hippocampus and ceding control over learning to the hippocampus during early training trials; and (2) aCSF enhanced striatal functioning to facilitate striatal-sensitive learning. With either interpretation, the results indicate that intrastriatal glucose injections compensated for the aCSF-induced effect. Finally, enhanced acquisition regardless of treatment was accompanied by rapid adoption of a response solution for the T-maze.

Specific features of the erythroid hemopoietic stem in CBA/CaLac mice with neuroses demonstrating good and poor learning capacities

We studied changes in the erythroid hemopoietic stem in CBA/CaLac mice with experimental neuroses demonstrating good and poor learning capacities (conflict situation and paradoxical sleep deprivation followed by training in a 3-arm T-maze). The animals with different learning capacities exhibited pronounced hyperplasia of the erythroid hemopoietic stem in response to neurosis. Activation of the erythron in good learners was related to acceleration of division and maturation of erythroid precursors and enhanced formation of cell complexes containing central macrophage. In poor learners hyperplasia of the erythroid hemopoietic stem under conditions of conflict situation manifested in activation of proliferation and differentiation (against the background of decreased count of erythroid and mixed complexes in the bone marrow), while after paradoxical sleep deprivation followed by T-maze training this hyperplasia was associated with increased formation of additional hemopoietic islets (against the background of desynchronization of division and maturation of erythroid precursor cells).

Late behavioural and neuropathological effects of local brain irradiation in the rat

The delayed consequences of radiation damage on learning and memory in rats were assessed over a period of 44 weeks, commencing 26 weeks after local irradiation of the brain with single doses of X-rays. Doses were set at levels known to produce vascular changes alone (20 Gy) or vascular changes followed by necrosis (25 Gy). Following T-maze training, 29 weeks after irradiation, irradiated and sham control groups performed equally well on the forced choice alternation task. When tested 35 weeks after irradiation, treated rats achieved a much lower percentage of correct choices than controls in T-maze alternation, with no difference between the two irradiated groups. At 38–40 weeks after irradiation, rats receiving both doses showed marked deficits in water maze place learning compared with age-matched controls; performance was more adversely affected by the higher dose. The extent of impairment was equivalent in the two groups of rats irradiated with 25 Gy, those trained or not previously trained in the T-maze, suggesting that water maze acquisition deficits were not influenced by prior experience in a different spatial task. In contrast to water maze acquisition, rats irradiated with 20 Gy showed no deficits in working memory assessed in the water maze 44 weeks after irradiation, whereas rats receiving 25 Gy showed substantial impairment. Rats receiving 25 Gy irradiation showed marked necrosis of the fimbria and degeneration of the corpus callosum, damage to the callosum occurring in animals examined histologically 46 weeks after irradiation, but in only a third of the animals examined at 41 weeks. However, there was no evidence of white matter necrosis in rats irradiated with 20 Gy, examined 46 weeks after irradiation. These findings demonstrated that local cranial irradiation with single doses of 20 and 25 Gy of X-rays produced delayed impairment of spatial learning and working memory in the rat. The extent of these deficits appears to be task- and dose-related, since rats treated with 25 Gy showed marked impairments in all measures, whereas rats treated with the lower dose showed less impairment in water maze learning and no deficits water maze working memory, despite significant disruption of working memory in the T-maze. The findings further suggest that although high dose irradiation-induced white matter necrosis is associated with substantial impairment, cognitive deficits may also be detected after a lower dose, not associated with the development of necrosis.

Modified T-maze training demonstrates dissociated memory loss in rats with ischemic hippocampal injury

Rats were trained for 20 days on a modified T-maze which required discrimination of a stem choice that was invariant and a goal choice that alternated. Animals were then exposed to 30 min of transient severe forebrain ischemia (postischemic, PI animals), 30 min of less severe ischemia (non-criterion ischemic, NCI animals), or sham operations (controls). After 30 postoperative days, all animals were returned to the maze for an additional 30 trials. PI rats demonstrated a dissociated performance on stem and goal choice. Specifically, the PI animals were significantly impaired on discrimination of a goal choice, yet performed comparably to controls on discrimination of a stem choice. Impaired goal choice for PI animals was not due to differential run times, proactive interference, or interaction between goal and stem choice. Histopathologic analysis of PI animals revealed bilateral destruction of the pyramidal neurons in the dorsal CA1 region of the hippocampus and lesser damage in the anterodorsal subiculum and the dorsolateral caudate. NCI animals had patchy variable ischemic neuronal damage. These data suggest that ischemic-induced bilateral CA1 pyramidal neuron loss is sufficient to cause dissociated performance in trained animals on a T-maze, but that less severe ischemia does not cause reproducible hippocampal neuron damage or functional deficits. Further, the data extends the development of reliable behavioral measures for an animal model of the amnesic syndrome that may occur in humans after cardiac arrest.

The effects of dieldrin and chlordimeform on learning and memory in the cockroach, Periplaneta americana: a study in behavioral toxicology.

A one-session T-maze training procedure for cockroaches in which animals were trained to turn right or left to avoid a shock was utilized to investigate the effects of dieldrin and chlordimeform (two neurotoxic pesticides) on learning and memory. These animals were trained and then tested for retention 5 h later. Three behavioral measures were recorded: choice behavior or direction turned, the time taken to proceed down the runway (runway time), and the time taken to proceed from the runway to either choice arm (choice point time). In control animals the number of correct choices, the runway time, and the choice point time increased with succeeding trials during training. Furthermore, control animals showed retention of correct choice behavior from training to testing. A nontoxic dose (no overt behavioral signs) of dieldrin was injected 2 h before training or 15 min after training. Pretraining injections of dieldrin eliminated correct-choice learning but did not alter the increase with training in runway or choice point times. Posttraining dieldrin administration did not interfere with retention of correct-choice behavior upon testing. These findings indicate that dieldrin induces behavioral alterations in a low acute dose and that detrimental effects are more likely with more complex behavior. Furthermore, behavioral disruption is more likely the closer exposure is to the initial learning of a task. Nontoxic doses of chlordimeform injected 1 h before training eliminated correct-choice learning and facilitated an increase in runway times during training.

Effects of cycloheximide and puromycin on learning and retention in the cockroach, P. americana

A new one-session T-maze training procedure for cockroaches, in which animals were trained to turn right or left to avoid shock, is described. This paradigm was utilized to investigate effects of protein synthesis inhibiting drugs on learning and retention. Cycloheximide (CXM), which inhibited protein synthesis by over 90% during the training period, did not impair acquisition and did not produce retention deficits at any interval up to 1 day after training. Puromycin (PURO), which inhibited protein synthesis by about 70% during the training period, produced amnesia 5 hr after training, while acquisition was not affected. Thus invertebrates, as well as vertebrates, are susceptible to amnesic effects of puromycin. Although PURO-injected animals showed retention deficits as measured by the number of correct turns, no retention deficit occurred for the behavioral modification consisting of an increase in runway time during the training period. Therefore, PURO appears to show specificity for the different types of longer-term memories that are formed in a training situation.

Effect of turn preference upon development of discriminated T-maze training ability of infant mice

Mice received 25 trials in a shock-escape T-maze at 7, 9, 11, or 13 days of age. On the first trial, shock-offset occurred upon reaching either goalbox. For the remaining 24 trials, half of the mice at each age were trained to escape to the goal opposite their first-trial choice, while the other half continued to receive shock-offset for reaching either goalbox. Comparisons of direction of choice-point turns indicated little difference between trained and free-choice groups at 7 days of age. In contrast, mice trained to a specific goal at 9, 11, and 13 days of age made more turns in the direction opposite their first-trial choice than did the free-choice groups.

Conventional and reversed partial reinforcement effects in selective learning

A reversed PRE was obtained for choice behavior following T-maze training in which CRF and PR Ss were “yoked” for correct and incorrect responses. Conventional acquisition procedures yielded differences in acquisition performance and a conventional PRE during extinction. Speed data yielded conventional PREs during extinction in all cases.

T-Maze Performance as a Function of Consummatory Activity

Wike and Barrientos ( 1957 ) examined the possibility that selective learning might be positively related to the duration of consummatory activity. Thirsty rats received T-maze training in which both responses were followed by a 5-ml. water reward, on one side through a 0.25-in. diameter tube and on the other through a 0.19-in, tube. The smaller tube required the longer duration consummatory response, and because Ss learned to approach the smaller tube it was concluded that reinforcement strength is posicively related to the duration of the consummatory response. This finding has great relevance for theories of reinforcement to the extent that it is not strictly dependent upon the specific details of this one experiment. The present experiment was performed under similar but slightly different conditions in an attempt to extend the generality of this previous finding.

Long incubation period in kala-azar.

Growing evidence supports a critical role for the dorsal striatum in cognitive as well as motor control. Both lesions and <i>in vivo</i> recordings demonstrate a transition in the engaged dorsal striatal subregion, from dorsomedial to dorsolateral, as skill performance shifts from an attentive phase to a more automatic or habitual phase. What are the neural mechanisms supporting the cognitive and behavioral transitions in skill learning? To pursue this question, we used T-maze training during which rats transition from early, attentive (dorsomedial) to late habitual (dorsolateral) performance. Following early or late training, we performed the first direct comparison of bidirectional synaptic plasticity in striatal brain slices, and the first evaluation of striatal synaptic plasticity by hemisphere relative to a learned turn. Consequently, we find that long-term potentiation and long-term depression are independently modulated with learning rather than reciprocally linked as previously suggested. Our results establish that modulation of evoked synaptic plasticity with learning depends on striatal subregion, training stage, and hemisphere relative to the learned turn direction. Exclusive to the contralateral hemisphere, intrinsic excitability is enhanced in dorsomedial relative to dorsolateral medium spiny neurons early in training and population responses are dampened late in training. Neuronal reconstructions indicate dendritic remodeling after training, which may represent a novel form of pruning. In conclusion, we describe region- and hemisphere-specific changes in striatal synaptic, intrinsic, and morphological plasticity which correspond to T-maze learning stages, and which may play a role in the cognitive transition between attentive and habitual strategies. <b>SIGNIFICANCE STATEMENT</b> We investigated neural plasticity in dorsal striatum from rats that were briefly or extensively trained on a directional T-maze task. Our results demonstrate that both the extent of training and the direction a rat learns to turn control the location and type of change in synaptic plasticity. In addition, brief training produces changes in neuron excitability only within one striatal subregion, whereas all training produces widespread changes in dendritic morphology. Our results suggest that activity in dorsomedial striatum strengthens the rewarded turn after brief training, whereas activity in dorsolateral striatum suppresses unrewarded turns after extensive training. This study illuminates how plasticity mediates learning using a task recognized for transitioning subjects from attentive to automatic performance.