Skill Learning Task Research Articles

Neural Filtering of Physiological Tremor Oscillations to Spinal Motor Neurons Mediates Short-Term Acquisition of a Skill Learning Task.

The acquisition of a motor skill involves adaptations of spinal and supraspinal pathways to alpha motoneurons. In this study, we estimated the shared synaptic contributions of these pathways to understand the neural mechanisms underlying the short-term acquisition of a new force-matching task. High-density surface electromyography (HDsEMG) was acquired from the first dorsal interosseous (FDI; 7 males and 6 females) and tibialis anterior (TA; 7 males and 4 females) during 15 trials of an isometric force-matching task. For two selected trials (pre- and post-skill acquisition), we decomposed the HDsEMG into motor unit spike trains, tracked motor units between trials, and calculated the mean discharge rate and the coefficient of variation of interspike interval (COVISI). We also quantified the post/pre ratio of motor units' coherence within delta, alpha, and beta bands. Force-matching improvements were accompanied by increased mean discharge rate and decreased COVISI for both muscles. Moreover, the area under the curve within alpha band decreased by ∼22% (TA) and ∼13% (FDI), with no delta or beta bands changes. These reductions correlated significantly with increased coupling between force/neural drive and target oscillations. These results suggest that short-term force-matching skill acquisition is mediated by attenuation of physiological tremor oscillations in the shared synaptic inputs. Supported by simulations, a plausible mechanism for alpha band reductions may involve spinal interneuron phase-cancelling descending oscillations. Therefore, during skill learning, the central nervous system acts as a matched filter, adjusting synaptic weights of shared inputs to suppress neural components unrelated to the specific task.

Association Between Physical Activity and Performance in Skill Learning Among Older Adults Based on Cognitive Function.

Most older adults experience cognitive and physical functioning problems; however, they require the ability to learn skills in response to age-related or social environmental changes for independent living. This study aimed to clarify the associations between age-related physical activity and performance in skill learning tasks based on cognitive function. Fifty-eight adults participated in this study and were divided into two groups: the control group (aged under 65years) and older adult group (aged over 65). All the participants performed two-skill learning exercises based on cognitive function. Habitual exercise was measured using an accelerometer and a self-reported questionnaire. At baseline, the scores on skill tasks were lower in the older adult group than in the control group and were associated with habitual exercise and motor performance. Skill acquisition, observed in both groups, was associated with age and self-reported physical activity. Retention of the acquired skill was not associated with habitual exercise, and it declined significantly in the older group. Skill acquisition was maintained regardless of age; however, the ability to retain the acquired skills decreased among the older adults. Habitual physical activity was associated with skill acquisition but not the retention of the acquired skill. Significance/Implications: The study findings highlight the association between habitual exercise and motor skill learning in older adults, providing insight for practitioners in the rehabilitation and health care fields.

Exploratory behavior is associated with the cognitive speed in male chestnut thrushes

Abstract Intra-individual variation in cognitive abilities has been widely reported in animals. Recent studies have found that individual cognitive performance varies with personality traits in a wide range of animal taxa, with a speed-accuracy trade-off between cognition and personality traits. Few studies investigated whether these relationships change depending on different contexts. Here we investigate whether the personality trait (as measured by exploratory behavior in a novel environment) is associated with cognition (novel skill learning and spatial memory) in wild male chestnut thrushes Turdus rubrocanus. Using an experimental novel skill-learning task set-up, we found that fast-exploring individuals explored the experimental device (a cardboard with 8 opaque cups) sooner than slow-exploring individuals. Exploratory behavior was not associated with individual spatial memory performances or an individual’s capacity to learn the novel skill. Learning speed was positively associated with the difficulty of learning phases, and fast-exploring individuals used less trials to meet the learning criterion. In addition, fast-exploring individuals took less time to complete the 24-h spatial memory test, but the accuracy of the test was not significantly different between individuals who were more or less exploratory. We suggest that variation in personality traits associates with individual learning speed in cognitive tasks and that this relationship is context-dependent.

Persistent changes in motor adaptation strategies after perturbations that require exploration of novel muscle activation patterns.

Motor skill learning requires the acquisition of novel muscle patterns and a new control policy-a process that requires time. In contrast, motor adaptation often requires only the adjustment of existing muscle patterns-a fast process. By altering the mapping of muscle activations onto cursor movements in a myoelectrically controlled virtual environment, we are able to create perturbations that require either the recombination of existing muscle synergies (compatible virtual surgery) or the learning of novel muscle patterns (incompatible virtual surgery). We investigated whether adaptation to a compatible surgery is affected by prior exposure to an incompatible surgery, i.e., a motor skill learning task. We found that adaptation to a compatible surgery was characterized by a decrease in the quality of muscle pattern reconstructions using the original synergies and an increase in reaction times only after exposure to an incompatible surgery. In contrast, prior exposure to a compatible surgery did not affect the learning process required to overcome an incompatible surgery. The fact that exposure to an incompatible surgery had a profound effect on the muscle patterns during the adaptation to a subsequent compatible surgery and not vice versa suggests that null space exploration, possibly combined with an explicit exploration strategy, is engaged during exposure to an incompatible surgery and remains enhanced during a new adaptation episode. We conclude that motor skill learning, requiring novel muscle activation patterns, leads to changes in the exploration strategy employed during a subsequent perturbation.NEW & NOTEWORTHY Motor skill learning requires the acquisition of novel muscle patterns, whereas motor adaptation requires adjusting existing ones. We wondered whether training a new motor skill affects motor adaptation strategies. We show that learning an incompatible perturbation, a complex skill requiring new muscle synergies, affects the muscle patterns observed during adaption to a compatible perturbation, which requires adjusting the existing synergies. Our results suggest that motor skill learning results in persistent changes in the exploration strategy.

Initial motor skill performance predicts future performance, but not learning

People show vast variability in skill performance and learning. What determines a person's individual performance and learning ability? In this study we explored the possibility to predict participants’ future performance and learning, based on their behavior during initial skill acquisition. We recruited a large online multi-session sample of participants performing a sequential tapping skill learning task. We used machine learning to predict future performance and learning from raw data acquired during initial skill acquisition, and from engineered features calculated from the raw data. Strong correlations were observed between initial and final performance, and individual learning was not predicted. While canonical experimental tasks developed and selected to detect average effects may constrain insights regarding individual variability, development of novel tasks may shed light on the underlying mechanism of individual skill learning, relevant for real-life scenarios.

Individuals with larger head volume have better learning ability in wild chestnut thrushes

Abstract Recent comparative studies on fish and mammals have suggested that brain size is associated with cognitive ability, such as problem solving and self-control; larger-brained individuals are assumed to have better cognitive ability. However, it remains largely unexplored whether larger-brained birds perform better in learning and spatial memory tests. Here, we used two experiments (novel skill learning task and spatial memory task) to examine whether head volume (proxy for brain size) could predict the learning performance and spatial performance in wild chestnut thrushes (Turdus rubrocanus). First, we found that chestnut thrushes with larger head volume were more likely to learn the novel skill than individuals with smaller head volume. Second, individuals with larger head volume relative to their tarsus (relative head volume) learned the novel skill faster. Age, sex, and exploratory tendency of individuals were not related to learning performance or learning speed. Finally, head volume, relative head volume and age was not associated with spatial performance in the spatial memory tasks. Together, these findings provided empirical evidences for the relationship between head volume and the ability of novel skill learning, and also provide a non-invasive method for studying the relationship between cognitive ability and head volume in other bird species.

EEG variability: Task-driven or subject-driven signal of interest?

Neurons in the brain are seldom perfectly quiet. They continually receive input and generate output, resulting in highly variable patterns of ongoing activity. Yet the functional significance of this variability is not well understood. If brain signal variability is functionally relevant and serves as an important indicator of cognitive function, then it should be highly sensitive to the precise manner in which a cognitive system is engaged and/or relate strongly to differences in behavioral performance. To test this, we examined EEG activity in younger adults as they performed a cognitive skill learning task and during rest. Several measures of EEG variability and signal strength were calculated in overlapping time windows that spanned the trial interval. We performed a systematic examination of the factors that most strongly influenced the variability and strength of EEG activity. First, we examined the relative sensitivity of each measure to across-subject variation (within blocks) and across-block variation (within subjects). We found that the across-subject variation in EEG variability and signal strength was much stronger than the across-block variation. Second, we examined the sensitivity of each measure to different sources of across-block variation during skill acquisition. We found that key task-driven changes in EEG activity were best reflected in changes in the strength, rather than the variability, of EEG activity. Lastly, we examined across-subject variation in each measure and its relationship with behavior. We found that individual differences in response time measures were best reflected in individual differences in the variability, rather than the strength, of EEG activity. Importantly, we found that individual differences in EEG variability related strongly to stable indicators of subject identity rather than dynamic indicators of subject performance. We therefore suggest that EEG variability may provide a more sensitive subject-driven measure of individual differences than task-driven signal of interest.

Procognitive profiling of a serotonin 5-HT6 receptor antagonist in a complex model system in rats: A novel translational approach for clinical prediction

IntroductionThe serial clinical failures of novel cognitive enhancer candidates point out the lack of predictive power in the preceding animal experimentation. For a more predictive profiling of putative procognitive drugs in rodents, we recently elaborated a methodical approach which consists of three fundamental steps: 1. teaching various learning tasks representing different cognitive domains to the same cohort of animals with the aim to create a population with ‘widespread knowledge’. 2. Applying a cognitive deficit-inducing intervention to transform this cohort of animals to a ‘patient population’. 3. Testing putative procognitive drugs with a ‘clinical trial-like’ design on the wide spectrum of cognitive (dys)functions in the actual ‘patient population’. The present study has been the first trial to test the feasibility and utility of the proposed system. MethodsThe population with ‘widespread knowledge’ consisted of 2 year old male Long-Evans rats with a learning history in five-choice serial reaction time task (5-CSRTT, attentional paradigm), Morris water maze (MWM, spatial learning), a cooperative task carried out in pairs (social learning), and a skill-learning task, „pot-jumping”. For inducing cognitive deficit, thus creating a ‘patient population’ we increased the difficulty of the tasks. For the cognitive enhancer mechanism to test in the system we chose a serotonin 5-HT6 receptor antagonist compound, RO4368554. Animals were randomly assigned to vehicle- and drug treated groups based on their baseline learning performance and their response in a pilot test of increase in task difficulty. During the 13-day long treatment with 3 mg/kg ip. RO4368554 all the learning paradigms were repeatedly run with increased difficulty supplemented with a novel object recognition test (NOR, episodic memory). ResultsIn the 5-CSRTT, reducing the stimulus duration from 1 s to 0.25 s caused a significant decrease in the percentage of correct responses (from 52 % to 31 % in the control group) which was not affected by the 5-HT6 receptor antagonist treatment (correct responses decreased from 58 % to 31 %). In the MWM, replacing the escape platform to a new location did not mean a hard challenge for the rats. Members of both groups could find it within a relatively short time: mean escape latencies were 83 s and 65 s at the first replacement trial and 58 s and 74 s at the second one in the control and drug-treated groups, respectively. In the cooperation paradigm, where the rats had to perform simultaneous nose-pokes to get a reward, task difficulty was increased by requiring two consecutive simultaneous nose-poking from the animals. This caused a fall in the percentage of successful trials in both groups (from 48 % to 12 % and from 50 % to 20 % in the saline – and drug-treated group, respectively), however, by the end of the treatment RO4368554-treated animals showed significantly higher performance (29 %) than saline treated rats (2%). The NOR test, carried out with a 5 -h delay, revealed poor recognition memory in both groups (discrimination index (DI) values were 0.13 and 0.06 for saline and RO4368554, respectively). Performance in the pot jumping test was also not improved by the drug-treatment. ConclusionsThe applied study design allowed parallel measurements of the action of the test compound on several cognitive functions and to follow its time course. RO4368554 did not show notable effects on impaired attention and visual recognition; nor did it affect spatial and procedural learning, but it exerted beneficial effect on cooperative behaviour. The revealed activity pattern highlight the cognitive domain most sensitive to the particular drug effect and may give hints for further target validating and clinical studies.

Learning new gait patterns is enhanced by specificity of training rather than progression of task difficulty

The use of motor learning strategies may enhance rehabilitation outcomes of individuals with neurological injuries (e.g., stroke or cerebral palsy). A common strategy to facilitate learning of challenging tasks is to use sequential progression – i.e., initially reduce task difficulty and slowly increase task difficulty until the desired difficulty level is reached. However, the evidence related to the use of such sequential progressions to improve learning is mixed for functional skill learning tasks, especially considering situations where practice duration is limited. Here, we studied the benefits of sequential progression using a functional motor learning task that has been previously used in gait rehabilitation. Three groups of participants (N = 43) learned a novel motor task during treadmill walking using different learning strategies. Participants in the specific group (n = 21) practiced only the criterion task (i.e., matching a target template that was scaled-up by 30%) throughout the training. Participants in the sequential group (n = 11) gradually progressed to the criterion task (from 3% to 30% in increments of 3%), whereas participants in the random group (n = 11) started at 3% and progressed in random increments (involving both increases and decreases in task difficulty) to the criterion task. At the end of training, kinematic tracking performance on the criterion task was evaluated in all participants both with and without visual feedback. Results indicated that the tracking error was significantly lower in the specific group, and no differences were observed between the sequential and the random progression groups. The findings indicate that the amount of practice in the criterion task is more critical than the difficulty and variations of task practice when learning new gait patterns during treadmill walking.

Modeling of Basal Ganglia to Incorporate the Procedural Memory

In this research work, a computational model is proposed for the procedural memory, and tried to mimic the functionalities of the Basal Ganglia, which is the brain region responsible for the procedural memory. In this proposed model, the procedural memory is further divided into the two sub memories. First, is the cognitive procedural memory that learns the sequence of decisions/actions, afterward it predicts the action in context of the previous occurred actions and the current sensory activations. Second, is the motor procedural memory which performs the skill learning task in context to the selected action of the cognitive procedural memory. The Learned action sequences in the cognitive procedural memory may have some overlapping which requires the identification of the current context during the sequence learning task, that helps the model in prediction of action in the current context and reduces the ambiguity. Model selects the action corresponding to the semantic activations of the semantic memory and the current context of the previous occurred actions and the gestures get invoked by the selected cognitive actions which then translated by the real working memory parameters to generate the real low-level action.

Impaired sequential and partially compensated probabilistic skill learning in Parkinson's disease.

The striatal dopaminergic dysfunction in Parkinson's disease (PD) has been associated with deficits in skill learning in numerous studies, but some of the findings remain controversial. Our aim was to explore the generality of the learning deficit using two widely reported skill learning tasks in the same group of Parkinson's patients. Thirty‐four patients with PD (mean age: 62.83 years, SD: 7.67) were compared to age‐matched healthy adults. Two tasks were employed: the Serial Reaction Time Task (SRT), testing the learning of motor sequences, and the Weather Prediction (WP) task, testing non‐sequential probabilistic category learning. On the SRT task, patients with PD showed no significant evidence for sequence learning. These results support and also extend previous findings, suggesting that motor skill learning is vulnerable in PD. On the WP task, the PD group showed the same amount of learning as controls, but they exploited qualitatively different strategies in predicting the target categories. While controls typically combined probabilities from multiple predicting cues, patients with PD instead focused on individual cues. We also found moderate to high correlations between the different measures of skill learning. These findings support our hypothesis that skill learning is generally impaired in PD, and can in some cases be compensated by relying on alternative learning strategies.

The Association Between Corticomotor Excitability and Motor Skill Learning in People With Painful Hand Arthritis.

Previous studies have shown a tendency for reduced motor cortex inhibition in chronic pain populations. People with chronic pain also routinely demonstrate motor deficiencies, including skill learning. The goals of the current study were to (1) provide a thorough analysis of corticomotor and intracortical excitability in people with chronic arthritic hand pain, and (2) examine the relationship between these measures and performance on a motor skill learning task. Twenty-three people with arthritic hand pain and 20 pain-free controls participated in a cross-sectional study. Transcranial magnetic stimulation was used to assess corticomotor and intracortical excitability of the first dorsal interosseus muscle. Participants then completed a 30-minute motor skill training task involving the index finger of the same hand. Hand arthritis participants showed evidence of reduced intracortical inhibition and enhanced facilitation, which correlated with duration of hand pain. Arthritis participants were initially poorer at the motor skill task but over the total training time performance was equivalent between groups. There were no associations found between measures of intracortical excitability and motor skill learning. Our findings are the first to provide evidence of cortical disinhibition in people with painful arthritis, as previously demonstrated in other chronic pain populations. Cortical excitability changes may progress the longer pain persists, with increased pain duration being associated with greater cortical disinhibition. There was no evidence that these changes in cortical excitability are related to impaired motor function or skill learning.

DTI microstructural abnormalities in adolescent siblings of patients with childhood-onset schizophrenia

DTI microstructural abnormalities in adolescent siblings of patients with childhood-onset schizophrenia

The effect of constant versus varied training on transfer in a cognitive skill learning task: The case of the Tower of Hanoi Puzzle

The differential effect of constant versus variable training conditions on acquisition and transfer has been demonstrated primarily in perceptual motor skills. In the present study, this effect was tested on 84 young adults using a cognitive skill learning task — the Tower of Hanoi Puzzle. The advantage of this task is that it allows testing the effect of the two training protocols on transfer by separately analyzing accuracy of the task solution, speed of reaching the correct solution and time planning before beginning to solve the task. Participants were divided into two groups. The “constant training” group practiced the task for 10 consecutive trials with identical configuration in terms of the “start” and “end” peg; followed by an 11th trial with a new configuration of the task (i.e., transfer). The “varied training” group practiced for 10 consecutive trials with different configurations, followed by a new configuration. As predicted, the constant training group yielded a higher cost when transferring to a new configuration of the task compared with the varied training group. These findings support the notion that varied training leads to the development of a schematic representation of the task solution, thus transfer is facilitated. These results have important implications in terms of the optimal learning conditions for adults while coping with cognitive problem-solving tasks.

Sparse Latent Space Policy Search

Computational agents often need to learn policies that involve many control variables, e.g., a robot needs to control several joints simultaneously. Learning a policy with a high number of parameters, however, usually requires a large number of training samples. We introduce a reinforcement learning method for sample-efficient policy search that exploits correlations between control variables. Such correlations are particularly frequent in motor skill learning tasks. The introduced method uses Variational Inference to estimate policy parameters, while at the same time uncovering a low-dimensional latent space of controls. Prior knowledge about the task and the structure of the learning agent can be provided by specifying groups of potentially correlated parameters. This information is then used to impose sparsity constraints on the mapping between the high-dimensional space of controls and a lower-dimensional latent space. In experiments with a simulated bi-manual manipulator, the new approach effectively identifies synergies between joints, performs efficient low-dimensional policy search, and outperforms state-of-the-art policy search methods.

Sleep benefits consolidation of visuo-motor adaptation learning in older adults.

Sleep is beneficial for performance across a range of memory tasks in young adults, but whether memories are similarly consolidated in older adults is less clear. Performance benefits have been observed following sleep in older adults for declarative learning tasks, but this benefit may be reduced for non-declarative, motor skill learning tasks. To date, studies of sleep-dependent consolidation of motor learning in older adults are limited to motor sequence tasks. To examine whether reduced sleep-dependent consolidation in older adults is generalizable to other forms of motor skill learning, we examined performance changes over intervals of sleep and wake in young (n = 62) and older adults (n = 61) using a mirror-tracing task, which assesses visuo-motor adaptation learning. Participants learned the task either in the morning or in evening, and performance was assessed following a 12-h interval containing overnight sleep or daytime wake. Contrary to our prediction, both young adults and older adults exhibited sleep-dependent gains in visuo-motor adaptation. There was a correlation between performance improvement over sleep and percent of the night in non-REM stage 2 sleep. These results indicate that motor skill consolidation remains intact with increasing age although this relationship may be limited to specific forms of motor skill learning.

Preserved priming effect in individuals with schizophrenia: cues towards rehabilitation

Introduction. Individuals with schizophrenia and affective disorders show relatively intact implicit memory as compared to declarative memory. Implicit memory is usually assessed with skill learning and priming tasks. Whereas priming is thought to involve storage changes in the posterior neocortex, skill learning is thought to rely more on the corticostriatal pathway. Since frontostriatal and frontotemporal dysfunctions are, respectively, found in schizophrenia and affective disorders, we hypothesised that individuals with schizophrenia and first-episode psychosis would exhibit disturbances in skill learning, but not priming.Methods. Thirty-five patients (11 first-episode psychosis; 11 schizophrenia; 13 affective disorders) and 10 controls completed a procedural learning and priming task. Participants had to identify fragmented images throughout five training sessions. The improvement of the threshold at which the images could be identified between the first and last session was used as an index of procedural learning. In a final session, the identification thresholds for old and new images were compared to assess the priming effect.Results. Whereas individuals with schizophrenia and first-episode psychosis showed impaired skill learning, the priming effect was similar in all groups.Conclusion. Even though some aspects of learning and memory are affected in schizophrenia, our results suggest that the posterior cortical pathway remains efficient at modulating the priming effect. This intact ability could be used to guide the elaboration of new rehabilitation programmes.

Learning of a simple grapho-motor task by young children and adults: similar acquisition but age-dependent retention.

Many new skills are acquired during early childhood. Typical laboratory skill learning tasks are not applicable for developmental studies that involve children younger than 8 years of age. It is not clear whether young children and adults share a basic underlying skill learning mechanism. In the present study, the learning and retention of a simple grapho-motor pattern were studied in three age groups: 5–6, 7–8, and 19–29 years. Each block of the task consists of identical patterns arranged in a spaced writing array. Progression across the block involves on-page movements while producing the pattern, and off-page movements between patterns. The participants practiced the production of the pattern using a digitizing tablet and were tested at 24 h and 2 weeks post-practice. All age groups produced the task blocks more quickly with practice, and the learning rate was inversely related to the initial production time. All groups exhibited additional gains 24 h post-practice that were well-retained 2 weeks later. The accuracy of the participants was maintained throughout the 2-weeks period. These findings suggest that young children and young adults use a similar mechanism when learning the task. Nevertheless, the 6-years-old spent more time off-page during retention testing than when tested at 24 h post-practice, thus supporting the notion that an age advantage may exists in the long-term retention of skills due to planning-dependent aspects.

Neural substrates underlying stimulation-enhanced motor skill learning after stroke.

Motor skill learning is one of the key components of motor function recovery after stroke, especially recovery driven by neurorehabilitation. Transcranial direct current stimulation can enhance neurorehabilitation and motor skill learning in stroke patients. However, the neural mechanisms underlying the retention of stimulation-enhanced motor skill learning involving a paretic upper limb have not been resolved. These neural substrates were explored by means of functional magnetic resonance imaging. Nineteen chronic hemiparetic stroke patients participated in a double-blind, cross-over randomized, sham-controlled experiment with two series. Each series consisted of two sessions: (i) an intervention session during which dual transcranial direct current stimulation or sham was applied during motor skill learning with the paretic upper limb; and (ii) an imaging session 1 week later, during which the patients performed the learned motor skill. The motor skill learning task, called the 'circuit game', involves a speed/accuracy trade-off and consists of moving a pointer controlled by a computer mouse along a complex circuit as quickly and accurately as possible. Relative to the sham series, dual transcranial direct current stimulation applied bilaterally over the primary motor cortex during motor skill learning with the paretic upper limb resulted in (i) enhanced online motor skill learning; (ii) enhanced 1-week retention; and (iii) superior transfer of performance improvement to an untrained task. The 1-week retention's enhancement driven by the intervention was associated with a trend towards normalization of the brain activation pattern during performance of the learned motor skill relative to the sham series. A similar trend towards normalization relative to sham was observed during performance of a simple, untrained task without a speed/accuracy constraint, despite a lack of behavioural difference between the dual transcranial direct current stimulation and sham series. Finally, dual transcranial direct current stimulation applied during the first session enhanced continued learning with the paretic limb 1 week later, relative to the sham series. This lasting behavioural enhancement was associated with more efficient recruitment of the motor skill learning network, that is, focused activation on the motor-premotor areas in the damaged hemisphere, especially on the dorsal premotor cortex. Dual transcranial direct current stimulation applied during motor skill learning with a paretic upper limb resulted in prolonged shaping of brain activation, which supported behavioural enhancements in stroke patients.

Mirror reversal and visual rotation are learned and consolidated via separate mechanisms: recalibrating or learning de novo?

Motor learning tasks are often classified into adaptation tasks, which involve the recalibration of an existing control policy (the mapping that determines both feedforward and feedback commands), and skill-learning tasks, requiring the acquisition of new control policies. We show here that this distinction also applies to two different visuomotor transformations during reaching in humans: Mirror-reversal (left-right reversal over a mid-sagittal axis) of visual feedback versus rotation of visual feedback around the movement origin. During mirror-reversal learning, correct movement initiation (feedforward commands) and online corrections (feedback responses) were only generated at longer latencies. The earliest responses were directed into a nonmirrored direction, even after two training sessions. In contrast, for visual rotation learning, no dependency of directional error on reaction time emerged, and fast feedback responses to visual displacements of the cursor were immediately adapted. These results suggest that the motor system acquires a new control policy for mirror reversal, which initially requires extra processing time, while it recalibrates an existing control policy for visual rotations, exploiting established fast computational processes. Importantly, memory for visual rotation decayed between sessions, whereas memory for mirror reversals showed offline gains, leading to better performance at the beginning of the second session than in the end of the first. With shifts in time-accuracy tradeoff and offline gains, mirror-reversal learning shares common features with other skill-learning tasks. We suggest that different neuronal mechanisms underlie the recalibration of an existing versus acquisition of a new control policy and that offline gains between sessions are a characteristic of latter.