Spaced Practice and Reactive Inhibition Have Limited or No effects on Motor Sequence Learning

Spaced Practice and Reactive Inhibition Have Limited or No effects on Motor Sequence Learning | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Article Spaced Practice and Reactive Inhibition Have Limited or No effects on Motor Sequence Learning Mohan W. Gupta, Timothy C. Rickard This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-8633674/v1 This work is licensed under a CC BY 4.0 License Status: Under Review Version 1 posted 20 You are reading this latest preprint version Abstract Spaced practice in declarative memory tasks consistently yields greater learning than massed practice, but spacing effects are less consistently observed for motor skills. This study evaluates factors that may determine spacing effects on motor skill learning, including: (1) extant theories of declarative spacing effects, (2) reactive inhibition, which transiently impairs performance and may also impair learning, and (3) the micro-consolidation hypothesis, which posits that motor skill learning takes place exclusively during brief performance breaks. Across two experiments, we varied the number of correct sequences per trial and the length of breaks while keeping the total correct sequence count constant, using a widely studied motor sequence task. A pronounced performance advantage was observed for the spaced groups by the end of training. However, on a later test in which the task conditions were equated, group performance was statistically indistinguishable. Hence, spaced practice yielded no or minimal learning advantage and the large reactive inhibition effect in the massed group appears to be a transient performance phenomenon without consequence for learning. Furthermore, we found no evidence for the most straightforward micro-consolidation account, which predicts greater learning with more breaks. Our results are consistent with a simple account advanced by Gupta and Rickard 1 , 2 , according to which learning occurs entirely online (i.e., concurrently with performance) and is independent of spacing and reactive inhibition. Finally, our findings indicate that proposed mechanisms for declarative spacing effects, such as memory reactivation and contextual variability, do not generalize to motor learning, highlighting fundamental differences between the two learning systems. Biological sciences/Neuroscience Biological sciences/Psychology Social science/Psychology Motor sequence learning reactive inhibition spacing effects Figures Figure 1 Figure 2 Figure 3 Introduction There are many reasons one may choose to take breaks while learning a motor skill (e.g., playing the piano); among them are muscle fatigue, cognitive fatigue, and faltering motivation. In the current work, we investigate whether brief breaks may provide not only respite, but also a boost in motor learning. In the declarative memory literature, there is a positive and potent effect of breaks on learning, a phenomenon known as the spacing effect 3 – 7 . Among the many techniques that have been proposed to enhance declarative learning, spaced practice is one of the more robust. Several of the more promising computational theoretical accounts of the declarative spacing effect may plausibly generalize to motor skill learning, such as accounts based on the ACT-R theory 4 , the Multiscale Context Model 5 , Stimulus Sampling Theory 6 , and the Predictive Performance Eq. 7. Across these models, three core mechanisms of the spacing effect have emerged: encoded memory reactivation, memory contextual diversity, and memory decay. The encoded memory mechanism can in principle be straightforwardly applied to motor skill learning, such that sufficient spacing between bouts of practice causes the retrieval and subsequent reactivation of a stored motor memory, increasing the strength of the memory, whereas massed practice makes long-term memory retrieval less likely due to the information being held in working memory 8 , 9 . Greater contextual diversity has been hypothesized to improve declarative learning by increasing the number and distinctiveness of context–item associations that can later cue retrieval 10 – 13 . A related idea in motor learning is that varying the practice context can promote more flexible, context-independent performance 14 – 16 . However, the extent to which brief rest breaks create meaningfully different contextual states — and therefore distinct retrieval cues — may be limited in motor tasks due to consistency in task, somatosensory inputs, effector outputs. This suggests that any spacing benefit from context diversity during short breaks may be limited in a simple motor sequence task. A third mechanism is memory decay. Declarative memories generally decay relatively rapidly following a power law 17 , 18 . Due to this decay rate, spacing out practice can supply opportune bumps to the strength of a memory 19 . On the other hand, motor memories tend to decay more slowly over comparable intervals 20 – 23 . Thus, brief within-session breaks may produce little forgetting and therefore are likely to benefit from spaced out repeated practice compared to declarative learning. To our knowledge, the Bayesian Learner model proposed by Körding and colleagues 24 is the only framework explicitly applied to spacing effects in both declarative and motor skill learning. The model casts learning as multi-timescale adaptation, separating fleeting, fast fluctuations from slower, meaningful changes. In this formulation, a disturbance is any prediction-error–inducing perturbation to the latent skill (i.e., the process noise that updates the hidden state). Rapid, frequent disturbances — typical of massed practice — are assigned to the fast process and quickly forgotten, whereas more widely spaced disturbances are attributed to the slow process and thus retained. Because timescales are represented as hidden processes in a linear dynamical system, the learner uses temporal structure to decide whether a given disturbance reflects noise (fast) or a stable change (slow). Applied to our experiments, the longer inter-practice intervals in the spaced groups should shift credit to the slow process, enhancing retention and performance. Beyond theories of the declarative spacing effect, two hypotheses with regard to motor learning suggest that spacing may enhance learning. Reactive inhibition (RI) is a well-established empirical phenomenon of worsening motor performance during continuous practice which dissipates during breaks 1 , 2 , 25 – 28 (Fig. 1; for a candidate neurological mechanistic account of RI, see Bächinger and colleagues 29 ). RI is more pronounced when practice is sustained for a long period (massed practice), but less pronounced when practice is short with frequent breaks (spaced practice). Although the classic effect of RI is a transient worsening of performance, several studies have addressed the possibility that RI may also negatively affect motor learning. Across alphabet printing, peg board learning, the stabilometer task, and the Tsai-Partington numbers tasks, every possible result has been found; greater learning with spaced training 30 , 31 , no learning difference between massed and spaced training 32 , 33 , and greater learning with massed training 34 . In a meta-analysis, Lee and Genovese 35 concluded that massed practice impairs both performance and learning. However, given the variety of findings, methodological limitations, and small sample sizes in some cases, we view that literature as inconclusive. If RI does attenuate motor learning, then reducing it through spacing would yield a spacing effect on a delayed test The new theory of micro-consolidation posits that motor learning occurs primarily or exclusively during breaks (offline), driven by hippocampal to neocortical replay of learned sequences on the time scale of seconds 36 – 39 . This theory is akin to the proposed encoded memory reactivation mechanism of the spacing effect in declarative learning. Hence, frequent breaks between brief performance trials (spaced practice) should promote learning due to greater memory reactivations. In apparent support of this theory, motor sequence performance immediately after a break is often better than at the end of the preceding training trial 36 , 37 , 39 . Finally, there is the possibility that spaced practice has no effect at all on motor learning, as proposed by Gupta and Rickard 1 , 2 . They showed that the response time slowing due to accrual of RI across performance trials, along with dissipation of that RI during breaks, provides a viable account for post-break performance improvements without invoking offline micro-consolidation. The question of whether spacing effects occur in the current motor sequence task is also informed by several empirical studies of other motor tasks. Spacing effects have been observed for several tasks, including the rotary pursuit test, typing, discrete linear movement learning, computer game learning, dynamic balancing, and motor sequence learning 40 – 46 . In contrast, investigations into other complex motor skills such as cup stacking and piano learning 47 , 48 and simple sequence learning in both humans and monkeys 49 have shown either a lack of spacing effects or inconsistent benefits across measured variables. Thus, the relatively limited number of studies investigating the spacing effect in motor skill learning do not provide clear guidance on what to expect in the current experiments. Further, there appears to have been no prior studies focused on investigating spacing effects for the finger tapping task investigated here, which has dominated the extensive literature on the hypothesis of sleep-based motor sequence memory consolidation over the last three decades 50 . We investigated whether spacing and RI affect motor sequence learning by manipulating both trial length (the number of sequences within each trial) and the duration of breaks between trials. In Experiment 1, both groups learned to perform a five-item (five keypress) sequence. The spaced group performed five correct sequences during each trial with 30 s breaks between trials. The massed group performed 25 correct sequences during each trial with 10 s breaks. After a 15-minute rest period during which the cumulative effect of RI over training trials is expected to fully resolve, there was a test involving an identical spaced task for both groups, with five correct sequences per trial and 30 s breaks. The total number of training and test sequences was the same for the two groups. Because both groups performed the same task on the test, any learning and RI effects across test trials should be equated. Hence, any performance difference between groups on the test should exclusively reflect differences in the amount of learning during training. In Experiment 2, we increased the sequence length to nine-items. The spaced group performed four correct sequences per trial with 30 s breaks, while the massed group performed 12 correct sequences with 10 s breaks. All other design aspects mirrored those of Experiment 1. Based on prior results 1 , 2 , the effect of RI on performance in the spaced group should be limited primarily to sequences within individual trials, with minimal RI accrual across trials (i.e., each 30 s break should be sufficient to dissipate most of the RI that accrued during the preceding trial). However, for the massed group there should be substantial accrual of unresolved RI across trials. Hence, if RI impairs learning in a dose-response manner, we should see better final test performance in the spaced group (a spacing effect). In addition, because there is a much longer total break time across training in the spaced group (1,020 s, vs 70 s in the massed group), there is more opportunity for offline micro-consolidation in the spaced group, again suggesting a positive spacing effect. In summary, we should observe better test phase performance in the spaced group if RI impairs learning and (or) if spaced training enhances learning through other spacing mechanisms, including micro-consolidation. In contrast, we should observe equivalent test performance in the two groups if each of the following conditions are true: (1) RI is exclusively a performance phenomenon, (2) the greater break frequency and duration for the spaced group in does not promote more offline micro-consolidation, and (3) none of the spacing mechanisms in theories developed for declarative learning generalize to the case of single-session motor sequence learning. The latter possibility is consistent with the hypothesis advanced by Gupta and Rickard 1 , 2 . Results Errors The error rate was calculated for each participant as the number of incorrect keypresses prior to each correct sequence within each trial. In Experiment 1, averaging over sequences, trials, and participants in the training phase, the error rate was 0.38 and 0.41 keypresses in the spaced and massed groups, respectively. A Bayes factor test on the error rate positively favored the null between groups, BF 01 = 5.25, d = -0.056. On the test trials the overall error rate was 0.44 and 0.17 keypresses in the spaced and massed groups, respectively; positively favoring the alternative, BF 10 = 8.28, d = -0.32. In Experiment 2, averaging over sequences, trials, and participants in the training phase, the error rate was 1.26, and 1.73, keypresses in the spaced and massed groups, respectively. A Bayes factor test on the error rate positively favored a difference between groups, BF 10 = 5.08, d = 0.282. On the test, the error rate was 0.99 and 0.6 keypresses in the spaced and massed groups, respectively. A Bayes factor test on the error rate positively favored an effect between groups BF 10 = 8.0, d = -0.3. Within-Trial RI To confirm the accrual of RI over correct sequences within trial in the massed group, we fit a Bayesian linear mixed-effects model with RT predicted by Sequence position its interaction with Group, including random intercepts and slopes by participant and a random intercept for trials within participant. In the massed group, RT increased with sequence position, BF 10 = 3.09e 57 , β = 5.09 ms/sequence. Similarly in the spaced group RT increased with sequence position, BF 10 = 4.64e 8 , β = 23.33 ms/sequence, In Experiment 2, the massed group shows that RT increased robustly with sequence position, BF 10 = 2.85e 13 , β = 58.24 ms/sequence. The spaced group yielded a similarly strong positive slope, BF 10 = 2.09e 8 , β = 62.18 ms/sequence. Training RTs Next, we investigated whether spacing had an effect on performance by the end of training. To test performance differences at the end of training between the massed and spaced groups in Experiment 1, a Bayes factor test very positively favored a group difference between the mean RTs for the last 25 training sequences, performed, BF 1 0 = 6243.1, d = 0.54, indicating greater accrual of cumulative RI in the massed group, as expected (Fig. 3 a). In Experiment 2, a Bayes factor test again very positively favored the alternative hypothesis, yielding BF 10 = 12.66e 76 , d = 0.68, (Fig. 3 c). Post-rest Gain Confirming the interaction between group and experimental phase (end of training vs. test) in Experiment 1 (Fig. 3 a), a Bayes factor test very strongly favored a difference in the mean RT difference scores (mean RT on the last 25 training sequences minus the first 25 test sequences) between the massed and spaced groups BF 10 = 2e 5 , d = 0.68. In Experiment 2 (Fig. 3 c), a Bayes factor test again very strongly favored a difference on the participant-level mean RT difference scores (mean RT on the last 24 training sequences minus first 24 test sequences) between the massed vs. spaced groups was significant, BF 10 = 7.95e 14 , d = 1.0. Final Test RTs The key comparison of this experiment is the grand mean correct sequence RT difference between groups. Critically, all accumulation of RI across training trials is expected to have dissipated during the rest period, and because on the test both groups are performing the spaced task (five sequences / trial with 30 s breaks), any learning and RI effects that occur over test trials should be equivalent. In Experiment 1, the Bayes factor test weakly favored the null, BF 01 = 2.76, d = 0.14 (Fig. 3 b). In Experiment 2, a Bayes factor test weakly favored the null on the group RT means of all the test sequences, BF 01 = 2.51, d = 0.15 (Fig. 3 d). Discussion We investigated whether spaced practice enhances motor skill learning in two experiments. During training in both experiments, a larger amount of RI was induced in the massed group by limiting the number and duration of breaks, and by requiring participants to complete a relatively large number of correct sequences per trial. After a 15-minute rest to dissipate RI, both groups performed the spaced task from the training phase. Despite the large performance differences at the end of training in both experiments, test performance was statistically equivalent in the two groups, suggesting that neither RI, micro-consolidation, nor other candidate mechanisms for spacing effects influenced training phase learning. In both experiments there was a non-significant effect ( d = 0.14; d = 0.15) toward better final test performance in the spaced group, and thus we cannot rule-out a small advantage in learning for that group. It is possible that the 15-minute rest period was insufficient to completely dissipate RI. However, that possibility seems unlikely given that on average the training session lasted around seven minutes, not including breaks, with the 15-minute rest period more than doubling that duration. Further, participants had the same number of total correct keypresses within each experiment. Regardless, any difference in achieved learning between the two groups is at best minimal. That apparent absence of a spacing effect was surprising in light of previous evidence that learning during spaced training is superior to learning during massed training, not only in declarative memory tasks, but also in some types of cognitive skill learning 51 , 52 . The conclusions in the preceding paragraph are drawn solely from the correct sequence RT results. Conceivably, greater error rates in one group may have obfuscated learning differences. In Experiment 1, the error rate was statistically equivalent during training, whereas the spaced group had a statistically higher error rate on the test. In Experiment 2, there were statistically higher error rates during training for the massed group, but no statistical differences in error rate on the test. However, both of those effects were weakly favored the null as indicated by the Bayes factors tests. Previously, Gupta and Rickard 2 speculated that there may be a speed-accuracy tradeoff, wherein the quicker spaced group has more errors than the slower massed group. However, those results do not fully replicate in this study. Hence, across several experiments there is no consistent pattern of error differences between massed and spaced groups. The simplest interpretation of our RT results is that the same amount of learning occurred in both groups across the two experiments, consistent with the online learning account advanced by Gupta and Rickard 1 , 2 . In that account, learning occurs during each executed sequence (and not during breaks) and is independent of the schedule of trials and breaks. Because the same number of sequences were performed in both the massed and spaced groups, the online account predicts equivalent learning by the end of training and hence equivalent performance on the test. In this account, learning is equivalent in the groups and all performance differences during training are solely due to differences in RI accrual and dissipation. Our findings place boundary conditions on the micro-consolidation account of learning. First, if all of the training phase learning occurred during breaks, then the amount of learning that occurred during each break must have been substantially greater in the massed groups (e.g., with eight 10 s breaks) than in the spaced groups (e.g., with twenty-seven 30 s breaks), given the equivalent performance on the test. The micro-consolidation theory, as developed to date, does not specify whether or how the rate of offline consolidation depends on either the number of sequences practiced within each trial or the time available during each break, and it remains to be seen whether it can accommodate the current results. Buch and colleagues 39 observed that hippocampal activity associated with offline micro-consolidation occurred throughout the entire 10 s break periods in their study. If there is a dose-response relationship between break time and amount of micro-consolidation, the spaced groups should have shown greater overall learning. Instead, our results align more closely with recent evidence from Das and colleagues 53 , who showed that short rest periods in a similar finger-tapping task produce micro-offline performance gains during training but no lasting advantage in later tests relative to continuous practice. They further demonstrated that such gains are present even for random, nonrepeating sequences, where no micro-consolidation could occur. Our results further suggest that the mechanisms proposed in extant spacing effect theories in declarative learning do not extend to motor skill learning, at least for the current finger tapping task. In particular, these results are inconsistent with the Bayesian Learner model 24 , which predicts domain-general spacing benefits. Other theories of the declarative spacing effect posit that spacing benefits learning through the reactivation of memories, strengthening the original memory itself, or associations between items and contexts, like in the Multiscale Context Model 5 and the Predictive Performance Equation models 7 . This reactivation may be particularly effective when there is sufficient spacing between repetitions, possibly due to memory traces being primarily mediated by the cortex and out of working memory 11 , 23 . However, reactivation may not benefit motor skill learning in the same way as declarative memories, possibly due to differing neural computations in areas like the basal ganglia and primary motor cortex 54 . Further, motor memories are much more resistant to forgetting 20 , 55 and thus reactivation may only give a marginal benefit to the memory trace strength, resulting in minimal spacing effects. Along with memory reactivations, spaced practice in declarative learning is thought to increase contextual variability, where each spaced repetition occurs in a distinct temporal or environmental context, enhancing memory strength through contextual reinstatement and context diversity 5 , 12 , 56 . In contrast, motor skill learning may be more influenced by physical and sensorimotor context variability, such as changes in movement patterns, speed, or force application 57 , 58 . While spacing might alter the temporal context, we speculate that it does not influence the internal sensorimotor context that drives motor memory formation and retention. As a result, the limited ability of spacing to modify these critical features may constrain its impact on motor learning. Instead, other forms of variability (e.g., interleaved practice schedules) have yielded more robust effects on motor skill retention and transfer 59 – 62 . Our findings mirror earlier studies comparing massed and spaced practice with sleep during the retention period 26 , 27 , 50 , 63 . Once confounding factors such as circadian phase and data averaging are controlled for in those studies, neither massed or spaced training yielded overnight gains — or losses — in performance. In our current study, we likewise observed equivalent post-rest performance across groups. This convergence across studies suggests that, for motor sequence tasks, neither spacing nor sleep enhances learning. Our results and others 49 , 53 suggest that spaced practice confers little or no benefit in relatively simple motor sequence learning tasks. Nevertheless, there is ample evidence that declarative processes contribute to motor learning 64 – 67 . The benefits of spaced practice may depend on the stage of learning it targets (e.g., early declarative-supported learning vs later procedural refinement), and this timing is likely task specific. For example, spacing may confer greater benefit to more complex motor skills that have a protracted declarative learning phase, such as driving or typing. This possibility points to an important direction for future research on spacing: identifying the time course of declarative involvement in motor learning and determining whether and for which types of tasks spacing influences that component of learning. Possibly, the short time frame of the current experiments (~ 30 minutes) was insufficient for spacing to produce a noticeable benefit in performance. For declarative memory, the largest effects are often found when spacing between sessions spans days and the final test is also days after the last training session 5 , 19 . However, in a meta-analysis of that literature by Cepeda and colleagues 11 , modest spacing effects were found for declarative tasks when spacing was similar to that in the current tasks (between 1–59 s between trials with retention intervals between 1 to 10 minutes), suggesting that our conclusions are not idiosyncratic to our timeframe. Further, the possibility that spacing effects for motor tasks are more robust at longer time intervals undermines neither our conclusion that RI is independent of motor skill learning nor diminish our conclusions that pose serious issues for the micro-consolidation account. Our results raise the possibility that massed motor skill training, and the associated RI, is not detrimental to naturalistic motor sequence learning – such as training on a musical instrument or sport – even when the worsening of performance due to RI is palpable. However, there may be factors that limit the generalizability of that conclusion (c.f., Wiseheart and colleagues 47 ). In the current tasks, for example, the error rate was not systematically higher in the massed groups. Longer trial durations, particularly if accompanied by a large error rate, may hamper learning. That possibility hints at an optimization scenario wherein the practical advantage of fewer, relatively long training trials is balanced against the possibility of error-driven impaired learning. That pattern would be consistent with recent findings for a motor skill game in which an optimal error rate for learning was observed 68 . In conclusion, our findings suggest that models of the spacing effects for declarative tasks do not generalize to motor sequence learning. Instead, our results align with the online learning account, wherein motor learning occurs during performance and is unaffected by breaks or RI. This result places important boundary conditions on existing theories, such as the Bayesian Learner model and micro-consolidation accounts, and highlights the need to better understand the unique mechanisms that underlie effective motor skill learning. Methods Participants To determine that sample size for Experiment 1, we conducted an a priori power analysis using G*Power 69 . Based on the final-test difference between spaced and massed practice reported by Gupta and Rickard 2 — with α = .05, power = .80, effect size d = .49, and a two-tailed independent-samples t-test, the required sample was 163 participants. We recruited 171 right-handed participants: 85 in the massed group (M age = 20.8 years, 76.5% female) and 86 in the spaced group (M age = 20.1 years, 81.4% female). Following the data-cleaning procedure described in Data Analysis and Cleaning, eight participants were excluded (six massed, two spaced), yielding a final sample of 163. In Experiment 2, we aimed to recruit a greater number of participants in order to replicate Experiment 1. We recruited 196 right-handed participants, 98 in the massed group (age = 20.5, F = 73.4%) and 98 in the spaced group (age = 21.1, F = 76.5%). Ten participants (six from the spaced group and four from the massed group) were removed using the cleaning procedure described in Data Analysis and Cleaning, leaving 186 participants. The experiment was conducted online and participants were recruited from the University of California, San Diego SONA undergraduate participant pool. Participants provided informed consent via button press prior to participation. All procedures were approved by the Institutional Review Board of the University of California, San Diego, and were conducted in accordance with relevant guidelines and regulations and in compliance with the Declaration of Helsinki. Experimental design and procedure Participants performed a classic finger-tapping-task in which they repeated the sequence, 4-1-3-2-4, as quickly and accurately as possible with their non-dominant left hand 70 (Fig. 2 ). Throughout the experiment, the numbered key sequence was displayed horizontally on the computer screen. Each keypress response was indicated by adding a ‘*’ to a single row underneath the displayed sequence. This did not indicate if the response was correct or incorrect, only that a key was pressed. Between trials, a displayed countdown timer indicated how much break time was left. Participants performed one correct warmup sequence before starting the main task. They had 10 s to do so or had to restart the warmup trial. The warmup trial was not included in the analyses. A between-participant design was used, in which the massed group completed 25 correct sequences per training trial with 10 s breaks between trials and the spaced group completed 5 correct sequences per training trial with 30 s breaks. After 175 completed sequences during the training phase in both groups, there was a 15-minute rest wherein participants in both groups performed a distraction task of double-digit addition. During the subsequent test both groups performed 10 trials with 5 correct sequences per trial and 30 s breaks. In Experiment 2, participants performed the same task as in Experiment 1, however, the sequence length was increased to nine-items, 4-1-3-2-4-2-3-1-4 (we reversed the first four items of the original sequence and then appended it to the end). Participants in the massed group completed 12 correct sequences per training trial with 10 s breaks between trials. The spaced group completed four correct sequences per training trial with 30 s breaks. After 108 completed sequences during training, there was a 15-minute rest wherein participants both groups performed a distraction task of double-digit addition. Before the test trials, participants completed a single warmup sequence and then had a break. Afterwards, both groups performed 15 test trials with four sequences per trial and 30 s breaks. All other aspects of the methods were identical to those of Experiment 1. Data Analysis and Cleaning The dependent measure was time in milliseconds to complete a correct sequence. Keypress latency within sequence was measured as the time (in milliseconds) between consecutive keypresses. To reduce noise in the data, we log-transformed the keypresses latencies. The mean of the logged keypresses was then calculated for each sequence and participant. We then exponentiated those means and multiplied them by the number of correct keypresses in the sequence (five for Experiment 1 and nine for Experiment 2) to obtain a measure of sequence RT in milliseconds. The first completed sequence was removed from each trial prior to further analysis due to the consistently longer RTs on those sequences, indicative of warmup 1 , 2 . Data were analyzed in R 71 (version 4.2.1) using the tidyverse 72 (version 2.0.0) and BayesFactor 73 (version 0.9.12–4.2) packages. For Bayes factors, the Cauchy prior width was set to r = 0.707 74 . Interpretation followed Raftery’s guidelines 75 : 1–3 = weak, 3–20 = positive, 20–150 = strong, > 150 = very strong evidence for the null or alternative hypothesis. BF₀₁ denotes evidence for the null hypothesis; BF₁₀ denotes evidence for the alternative. To fit Bayesian mixed-effects models, we used brms 76 with sequence RT as the outcome. For numerical stability, all continuous variables were z-scored prior to modeling; reported effects are back-transformed to milliseconds for interpretability. Bayes factors comparing nested models were obtained by bridge sampling the marginal likelihoods of the scaled models, taking the median of multiple repetitions. We conducted an unbiased data cleaning procedure to remove poor performing participants. Although our task design only allowed participants to advance after correctly executing the sequence multiple times in a trial, brief lapses in attention - such as stepping away from their keyboard - could still occur, inflating error rates and RT. Therefore, we summed each participant’s total incorrect keypresses and excluded anyone whose error count exceeded 2.5 standard deviations above the group mean. In parallel, we calculated each participant’s average RT across all trials and removed those whose mean RT fell above 2.5 standard deviations of the group mean. Data availability All data are available at https://osf.io/ukwf9/ . Further information and requests for resources should be directed to and will be fulfilled by the corresponding author, TCR ( [email protected] ). Code Availability All stimuli, and analyses are available at https://osf.io/ukwf9/ . Further information and requests for resources should be directed to and will be fulfilled by the corresponding author, TCR ( [email protected] ). Declarations Author Note This work was not part of a preregistered study. Materials for this study are archived at the Open Science Framework at: https://osf.io/ukwf9/ Acknowledgements None to declare. Author Contributions MWG conducted the experiments and analyzed the data. Both MWG and TCR designed the experiments and wrote the manuscript. Funding None. Competing Interest Statement We have no competing interests to report. References Gupta, M. W. & Rickard, T. C. Dissipation of reactive inhibition is sufficient to explain post-rest improvements in motor sequence learning. Npj Sci. Learn. 7 , 1–4 (2022). Gupta, M. W. & Rickard, T. C. Comparison of online, offline, and hybrid hypotheses of motor sequence learning using a quantitative model that incorporate reactive inhibition. Sci. Rep. 14 , 4661 (2024). Melton, A. W. The situation with respect to the spacing of repetitions and memory. J. Verbal Learn. Verbal Behav. 9 , 596–606 (1970). Pavlik, P. I. Jr & Anderson, J. R. Practice and Forgetting Effects on Vocabulary Memory: An Activation-Based Model of the Spacing Effect. Cogn. Sci. 29 , 559–586 (2005). Mozer, M. C., Pashler, H., Cepeda, N., Lindsey, R. V. & Vul, E. Predicting the Optimal Spacing of Study: A Multiscale Context Model of Memory. in Advances in Neural Information Processing Systems vol. 22 (Curran Associates, Inc., (2009). Tullis, J. G. & Benjamin, A. S. On the effectiveness of self-paced learning. J. Mem. Lang. 64 , 109–118 (2011). Walsh, M. M., Gluck, K. A., Gunzelmann, G., Jastrzembski, T. & Krusmark, M. Evaluating the Theoretic Adequacy and Applied Potential of Computational Models of the Spacing Effect. Cogn. Sci. 42 , 644–691 (2018). Braun, K. & Rubin, D. C. The Spacing Effect Depends on an Encoding Deficit, Retrieval, and Time in Working Memory: Evidence. Memory 6 , 37–66 (1998). Thios, S. J. & D’Agostino, P. R. Effects of repetition as a function of study-phase retrieval. J. Verbal Learn. Verbal Behav. 15 , 529–536 (1976). Hintzman, D. L., Summers, J. J. & Block, R. A. What causes the spacing effect? Some effects of repetition, duration, and spacing on memory for pictures. Mem. Cognit . 3 , 287–294 (1975). Cepeda, N. J., Pashler, H., Vul, E., Wixted, J. T. & Rohrer, D. Distributed practice in verbal recall tasks: A review and quantitative synthesis. Psychol. Bull. 132 , 354–380 (2006). Polyn, S. M., Norman, K. A. & Kahana, M. J. A context maintenance and retrieval model of organizational processes in free recall. Psychol. Rev. 116 , 129–156 (2009). Maddox, G. B. Understanding the underlying mechanism of the spacing effect in verbal learning: a case for encoding variability and study-phase retrieval. J. Cogn. Psychol. 28 , 684–706 (2016). Wright, D. L. & Shea, C. H. Contextual dependencies in motor skills. Mem. Cognit . 19 , 361–370 (1991). Ruitenberg, M. F. L., Abrahamse, E. L., De Kleine, E. & Verwey, W. B. Context-dependent motor skill: perceptual processing in memory-based sequence production. Exp. Brain Res. 222 , 31–40 (2012). Apolinário-Souza, T. et al. The effect of context variability on motor learning. Hum. Mov. Sci. 77 , 102794 (2021). Wixted, J. T. & Ebbesen, E. B. On the Form of Forgetting. Psychol. Sci. 2 , 409–415 (1991). Rubin, D. C. & Wenzel, A. E. One Hundred Years of Forgetting: A Quantitative Description of Retention. Psychol. Rev. 103 , 734–760 (1996). Cepeda, N. J., Vul, E., Rohrer, D., Wixted, J. T. & Pashler, H. Spacing Effects in Learning: A Temporal Ridgeline of Optimal Retention. Psychol. Sci. 19 , 1095–1102 (2008). Smith, C. D. et al. Memories that last in old age: motor skill learning and memory preservation. Neurobiol. Aging . 26 , 883–890 (2005). Squire, L. R. & Wixted, J. T. The Cognitive Neuroscience of Human Memory Since H.M. Annu. Rev. Neurosci. 34 , 259–288 (2011). Winocur, G. & Moscovitch, M. Memory Transformation and Systems Consolidation. J. Int. Neuropsychol. Soc. 17 , 766–780 (2011). Tallman, C. W., Clark, R. E. & Smith, C. N. Human brain activity and functional connectivity as memories age from one hour to one month. Cogn. Neurosci. 13 , 115–133 (2022). Kording, K. P., Tenenbaum, J. B. & Shadmehr, R. The dynamics of memory as a consequence of optimal adaptation to a changing body. Nat. Neurosci. 10 , 779–786 (2007). Hull, C. L. Principles of Behavior. J. Nerv. Ment Dis. 101 , 396 (1945). Rickard, T. C., Cai, D. J., Rieth, C. A., Jones, J. & Ard, M. C. Sleep does not enhance motor sequence learning. J. Exp. Psychol. Learn. Mem. Cogn. 34 , 834–842 (2008). Brawn, T. P., Fenn, K. M., Nusbaum, H. C. & Margoliash, D. Consolidating the Effects of Waking and Sleep on Motor-Sequence Learning. J. Neurosci. 30 , 13977–13982 (2010). Török, B., Janacsek, K., Nagy, D. G., Orbán, G. & Nemeth, D. Measuring and filtering reactive inhibition is essential for assessing serial decision making and learning. J. Exp. Psychol. Gen. 146 , 529–542 (2017). Bächinger, M. et al. Human motor fatigability as evoked by repetitive movements results from a gradual breakdown of surround inhibition. eLife 8 , e46750 (2019). Ammons, R. B. Acquisition of motor skill: I. Quantitative analysis and theoretical formulation. Psychol. Rev. 54 , 263–281 (1947). Kimble, G. A. An experimental test of a two-factor theory of inhibition. J. Exp. Psychol. 39 , 15–23 (1949). Carron, A. V. Performance and Learning in a Discrete Motor Task under Massed Vs. Distributed Practice. Res. Q. Am. Assoc. Health Phys. Educ. Recreat . 40 , 481–489 (1969). Carron, A. V. Physical Fatigue and Motor Learning. Res. Q. Am. Assoc. Health Phys. Educ. Recreat . 40 , 682–686 (1969). Archer, E. J. Postrest performance in motor learning as a function of prerest degree of distribution of practice. J. Exp. Psychol. 47 , 47–51 (1954). Lee, T. D. & Genovese, E. D. Distribution of Practice in Motor Skill Acquisition: Learning and Performance Effects Reconsidered. Res. Q. Exerc. Sport . 59 , 277–287 (1988). Bönstrup, M. et al. A Rapid Form of Offline Consolidation in Skill Learning. Curr. Biol. 29 , 1346–1351e4 (2019). Bönstrup, M., Iturrate, I., Hebart, M. N., Censor, N. & Cohen, L. G. Mechanisms of offline motor learning at a microscale of seconds in large-scale crowdsourced data. Npj Sci. Learn. 5 , 7 (2020). Jacobacci, F. et al. Rapid hippocampal plasticity supports motor sequence learning. Proc. Natl. Acad. Sci. 117, 23898–23903 (2020). Buch, E. R., Claudino, L., Quentin, R., Bönstrup, M. & Cohen, L. G. Consolidation of human skill linked to waking hippocampo-neocortical replay. Cell. Rep. 35 , 109193 (2021). Marshall, P. H., Jones, M. T. & Sheehan, E. M. The Spacing Effect in Short-Term Motor Memory: The Differential Attention Hypothesis. J. Mot Behav. 9 , 119–126 (1977). Baddeley, A. D. & Longman, D. J. A. The Influence of Length and Frequency of Training Session on the Rate of Learning to Type. Ergonomics 21 , 627–635 (1978). Ball, L. J. & Payne, R. B. Comparison of Inhibition and Consolidation Theories of Spaced Practice Effects. Percept. Mot Skills . 66 , 219–225 (1988). Shebilske, W. L., Goettl, B. P., Corrington, K. & Day, E. A. Interlesson spacing and task-related processing during complex skill acquisition. J. Exp. Psychol. Appl. 5 , 413–437 (1999). Shea, C. H., Lai, Q., Black, C. & Park, J. H. Spacing practice sessions across days bene®ts the learning of motor skills. Hum Mov. Sci (2000). Savion-Lemieux, T. & Penhune, V. B. The effects of practice and delay on motor skill learning and retention. Exp. Brain Res. 161 , 423–431 (2005). Krigolson, O. E., Ferguson, T. D., Colino, F. L. & Binsted, G. Distribution of Practice Combined with Observational Learning Has Time Dependent Effects on Motor Skill Acquisition. Percept. Mot Skills . 128 , 885–899 (2021). Wiseheart, M., D’Souza, A. A. & Chae, J. Lack of spacing effects during piano learning. PLOS ONE . 12 , e0182986 (2017). Cecilio-Fernandes, D., Cnossen, F., Jaarsma, D. A. D. C. & Tio, R. A. Avoiding Surgical Skill Decay: A Systematic Review on the Spacing of Training Sessions. J. Surg. Educ. 75 , 471–480 (2018). Hikosaka, O. et al. Long-term retention of motor skill in macaque monkeys and humans. Exp. Brain Res. 147 , 494–504 (2002). Rickard, T. C., Pan, S. C. & Gupta, M. W. Severe publication bias contributes to illusory sleep consolidation in the motor sequence learning literature. J. Exp. Psychol. Learn. Mem. Cogn. 48 , 1787–1796 (2022). Rohrer, D. & Taylor, K. The effects of overlearning and distributed practise on the retention of mathematics knowledge. Appl. Cogn. Psychol. 20 , 1209–1224 (2006). Rickard, T. C., Lau, J. S. H. & Pashler, H. Spacing and the transition from calculation to retrieval. Psychon Bull. Rev. 15 , 656–661 (2008). Das, A., Karagiorgis, A., Diedrichsen, J., Stenner, M. P. & Azañón, E. Micro-offline gains do not reflect offline learning during early motor skill acquisition in humans. Proc. Natl. Acad. Sci. 122, e2509233122 (2025). Hardwick, R. M., Forrence, A. D., Krakauer, J. W. & Haith, A. M. Time-dependent competition between goal-directed and habitual response preparation. Nat. Hum. Behav. 3 , 1252–1262 (2019). Luft, A. R. & Buitrago, M. M. Stages of Motor Skill Learning. Mol. Neurobiol. 32 , 205–216 (2005). Howard, M. W. & Kahana, M. J. A Distributed Representation of Temporal Context. J. Math. Psychol. 46 , 269–299 (2002). Shea, J. & Morgan, R. Contextual interference effects on the acquisition, retention, and transfer of a motor skill. J. Exp. Psychol. [Hum Learn. ] 5 , 179 (1979). Goode, S. & Magill, R. A. Contextual Interference Effects in Learning Three Badminton Serves. Res. Q. Exerc. Sport . 57 , 308–314 (1986). Schmidt, R. A. & Bjork, R. A. New Conceptualizations of Practice: Common Principles in Three Paradigms Suggest New Concepts for Training. Psychol. Sci. 3 , 207–218 (1992). Schmidt, R. A. & Young, D. E. Transfer of Movement Control in Motor Skill Learning. in Transfer of Learning 47–79Elsevier, (1987). 10.1016/B978-0-12-188950-0.50009-6 Brady, F. Contextual Interference: A Meta-Analytic Study. Percept. Mot Skills . 99 , 116–126 (2004). Schorn, J. M. & Knowlton, B. J. Interleaved practice benefits implicit sequence learning and transfer. Mem. Cognit . 49 , 1436–1452 (2021). Nettersheim, A., Hallschmid, M., Born, J. & Diekelmann, S. The Role of Sleep in Motor Sequence Consolidation: Stabilization Rather Than Enhancement. J. Neurosci. 35 , 6696–6702 (2015). Keisler, A. & Shadmehr, R. A. Shared Resource between Declarative Memory and Motor Memory. J. Neurosci. 30 , 14817–14823 (2010). Taylor, J. A., Krakauer, J. W. & Ivry, R. B. Explicit and Implicit Contributions to Learning in a Sensorimotor Adaptation Task. J. Neurosci. 34 , 3023–3032 (2014). McDougle, S. D., Bond, K. M. & Taylor, J. A. Explicit and Implicit Processes Constitute the Fast and Slow Processes of Sensorimotor Learning. J. Neurosci. 35 , 9568–9579 (2015). Krakauer, J. W., Hadjiosif, A. M., Xu, J., Wong, A. L. & Haith, A. M. Motor Learning. in Comprehensive Physiology (ed. Terjung, R.) 613–663Wiley, (2019). 10.1002/cphy.c170043 Al-Fawakhiri, N., Kayani, S. & McDougle, S. D. Evidence of an optimal error rate for motor skill learning. Preprint at. https://doi.org/10.1101/2023.07.19.549705 (2023). Faul, F., Erdfelder, E., Lang, A. G. & Buchner, A. G*Power 3: A flexible statistical power analysis program for the social, behavioral, and biomedical sciences. Behav. Res. Methods . 39 , 175–191 (2007). Nissen, M. J. & Bullemer, P. Attentional requirements of learning: Evidence from performance measures. Cognit Psychol. 19 , 1–32 (1987). R Core Team. R: A Language and Environment for Statistical Computing . (2024). https://www.R-project.org/ Wickham, H. et al. Welcome to the tidyverse. J. Open. Source Softw. 4 , 1686 (2019). Morey, R. D. & Rouder, J. N. BayesFactor: Computation of Bayes Factors for Common Designs . (2024). https://CRAN.R-project.org/package=BayesFactor Wagenmakers, E. J. et al. Bayesian inference for psychology. Part II: Example applications with JASP. Psychon Bull. Rev. 25 , 58–76 (2018). Raftery, A. E. Bayesian Model Selection in Social Research. Sociol. Methodol. 25 , 111–163 (1995). Bürkner, P. C. Bayesian item response modeling in R with brms and Stan. J. Stat. Softw. 100 , 1–54 (2021). Additional Declarations No competing interests reported. Cite Share Download PDF Status: Under Review Version 1 posted Editorial decision: Revision requested 24 Feb, 2026 Reviews received at journal 23 Feb, 2026 Reviews received at journal 22 Feb, 2026 Reviews received at journal 21 Feb, 2026 Reviews received at journal 20 Feb, 2026 Reviews received at journal 12 Feb, 2026 Reviewers agreed at journal 04 Feb, 2026 Reviewers agreed at journal 02 Feb, 2026 Reviewers agreed at journal 01 Feb, 2026 Reviewers agreed at journal 01 Feb, 2026 Reviewers agreed at journal 31 Jan, 2026 Reviewers agreed at journal 30 Jan, 2026 Reviewers agreed at journal 30 Jan, 2026 Reviewers agreed at journal 30 Jan, 2026 Reviewers agreed at journal 30 Jan, 2026 Reviewers invited by journal 30 Jan, 2026 Editor assigned by journal 30 Jan, 2026 Editor invited by journal 30 Jan, 2026 Submission checks completed at journal 28 Jan, 2026 First submitted to journal 28 Jan, 2026 You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-8633674","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Article","associatedPublications":[],"authors":[{"id":583665396,"identity":"1f225560-c843-4c04-b1ff-a414fb1513fe","order_by":0,"name":"Mohan W. Gupta","email":"","orcid":"","institution":"Princeton University","correspondingAuthor":false,"prefix":"","firstName":"Mohan","middleName":"W.","lastName":"Gupta","suffix":""},{"id":583665397,"identity":"83bf0d00-b91d-45cb-9ea9-0fc96828f99b","order_by":1,"name":"Timothy C. Rickard","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAAv0lEQVRIiWNgGAWjYFACxgbGBiDZD2LzgLiENPDAtMxsIF4LRBXjhgPEarGXPtz4cUbNPdnNN5Iff3jDYCML1ovXFr7EZskNx4qNt91IM5Ocw5BmTFgLD2Mb4wO2hMRtN3LYmHkYDicSqeVfQuLmGTnMn3kY/hOpZWNbQuIGiRwGaR6GA0RoOcPYLDmzL8F4xplnQL8YJBvPJKSFvYf94ceebwmy/e2gEKuwk+0jpAUNGJCmfBSMglEwCkYBDgAAc7tEPJAJzBYAAAAASUVORK5CYII=","orcid":"","institution":"University of California San Diego","correspondingAuthor":true,"prefix":"","firstName":"Timothy","middleName":"C.","lastName":"Rickard","suffix":""}],"badges":[],"createdAt":"2026-01-18 21:53:17","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-8633674/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-8633674/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":101881131,"identity":"8b84026d-c343-49e4-bf8a-4a71780b2076","added_by":"auto","created_at":"2026-02-04 15:10:04","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":37501,"visible":true,"origin":"","legend":"\u003cp\u003eLegend not included with this version\u003c/p\u003e","description":"","filename":"Figure1.png","url":"https://assets-eu.researchsquare.com/files/rs-8633674/v1/4494f98b0c3f03ae97eb9de6.png"},{"id":101754181,"identity":"a2965dab-79fd-4bd0-931d-8906682351f3","added_by":"auto","created_at":"2026-02-03 10:41:54","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":321651,"visible":true,"origin":"","legend":"\u003cp\u003eIn both experiments, participants learned to type a motor sequence over one session with their non-dominant left hand. The spaced groups performed fewer sequences per trial and had longer breaks than the massed groups. The overall number of sequences performed was equated between groups, but break time was significantly longer in the spaced group. After training, participants performed 15 minutes of double-digit addition during the rest period. After, both groups were tested on the trained sequence with the spaced design for sequences per trial and break time.\u003c/p\u003e","description":"","filename":"Figure2.png","url":"https://assets-eu.researchsquare.com/files/rs-8633674/v1/91350ea965fce1c24817ea50.png"},{"id":101685429,"identity":"7c07e582-a3c8-4443-a744-c32b29ef3c57","added_by":"auto","created_at":"2026-02-02 15:17:20","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":114301,"visible":true,"origin":"","legend":"\u003cp\u003ea) Experiment 1: Each dot is the mean RT for a correctly completed sequence; line segments connect sequences within the same trial. Because the spaced group completed more trials, more first-sequence warmup observations were removed, so fewer sequences are shown. Gray error bars indicate SE. b) Experiment 1: Final test mean RT averaged across test trials. Dots show participant means; error bars are 95% CIs. c) Experiment 2: Same plotting conventions as in (a). d) Experiment 2: Same plotting conventions as in (b).\u003c/p\u003e","description":"","filename":"figure3.png","url":"https://assets-eu.researchsquare.com/files/rs-8633674/v1/ed76062771d314a031bf6075.png"},{"id":101884317,"identity":"14d49d97-46b8-4065-b4e5-cf7eea2ab62d","added_by":"auto","created_at":"2026-02-04 15:31:02","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":1135111,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-8633674/v1/e8310542-b0d7-40f1-abd7-f7d7e4636574.pdf"}],"financialInterests":"No competing interests reported.","formattedTitle":"Spaced Practice and Reactive Inhibition Have Limited or No effects on Motor Sequence Learning","fulltext":[{"header":"Introduction","content":"\u003cp\u003eThere are many reasons one may choose to take breaks while learning a motor skill (e.g., playing the piano); among them are muscle fatigue, cognitive fatigue, and faltering motivation. In the current work, we investigate whether brief breaks may provide not only respite, but also a boost in motor learning. In the declarative memory literature, there is a positive and potent effect of breaks on learning, a phenomenon known as the \u003cem\u003espacing effect\u003c/em\u003e\u003csup\u003e\u003cspan additionalcitationids=\"CR4 CR5 CR6\" citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e7\u003c/span\u003e\u003c/sup\u003e. Among the many techniques that have been proposed to enhance declarative learning, spaced practice is one of the more robust.\u003c/p\u003e \u003cp\u003eSeveral of the more promising computational theoretical accounts of the declarative spacing effect may plausibly generalize to motor skill learning, such as accounts based on the ACT-R theory\u003csup\u003e\u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e\u003c/sup\u003e, the Multiscale Context Model\u003csup\u003e\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e\u003c/sup\u003e, Stimulus Sampling Theory\u003csup\u003e\u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e\u003c/sup\u003e, and the Predictive Performance Eq.\u0026nbsp;7. Across these models, three core mechanisms of the spacing effect have emerged: encoded memory reactivation, memory contextual diversity, and memory decay. The encoded memory mechanism can in principle be straightforwardly applied to motor skill learning, such that sufficient spacing between bouts of practice causes the retrieval and subsequent reactivation of a stored motor memory, increasing the strength of the memory, whereas massed practice makes long-term memory retrieval less likely due to the information being held in working memory\u003csup\u003e\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e,\u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e\u003c/sup\u003e.\u003c/p\u003e \u003cp\u003eGreater contextual diversity has been hypothesized to improve declarative learning by increasing the number and distinctiveness of context\u0026ndash;item associations that can later cue retrieval\u003csup\u003e\u003cspan additionalcitationids=\"CR11 CR12\" citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e13\u003c/span\u003e\u003c/sup\u003e. A related idea in motor learning is that varying the practice context can promote more flexible, context-independent performance\u003csup\u003e\u003cspan additionalcitationids=\"CR15\" citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e16\u003c/span\u003e\u003c/sup\u003e. However, the extent to which brief rest breaks create meaningfully different contextual states \u0026mdash; and therefore distinct retrieval cues \u0026mdash; may be limited in motor tasks due to consistency in task, somatosensory inputs, effector outputs. This suggests that any spacing benefit from context diversity during short breaks may be limited in a simple motor sequence task.\u003c/p\u003e \u003cp\u003eA third mechanism is memory decay. Declarative memories generally decay relatively rapidly following a power law\u003csup\u003e\u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e,\u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e\u003c/sup\u003e. Due to this decay rate, spacing out practice can supply opportune bumps to the strength of a memory\u003csup\u003e\u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e19\u003c/span\u003e\u003c/sup\u003e. On the other hand, motor memories tend to decay more slowly over comparable intervals\u003csup\u003e\u003cspan additionalcitationids=\"CR21 CR22\" citationid=\"CR20\" class=\"CitationRef\"\u003e20\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e23\u003c/span\u003e\u003c/sup\u003e. Thus, brief within-session breaks may produce little forgetting and therefore are likely to benefit from spaced out repeated practice compared to declarative learning.\u003c/p\u003e \u003cp\u003eTo our knowledge, the Bayesian Learner model proposed by K\u0026ouml;rding and colleagues\u003csup\u003e\u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e24\u003c/span\u003e\u003c/sup\u003e is the only framework explicitly applied to spacing effects in both declarative and motor skill learning. The model casts learning as multi-timescale adaptation, separating fleeting, fast fluctuations from slower, meaningful changes. In this formulation, a disturbance is any prediction-error\u0026ndash;inducing perturbation to the latent skill (i.e., the process noise that updates the hidden state). Rapid, frequent disturbances \u0026mdash; typical of massed practice \u0026mdash; are assigned to the fast process and quickly forgotten, whereas more widely spaced disturbances are attributed to the slow process and thus retained. Because timescales are represented as hidden processes in a linear dynamical system, the learner uses temporal structure to decide whether a given disturbance reflects noise (fast) or a stable change (slow). Applied to our experiments, the longer inter-practice intervals in the spaced groups should shift credit to the slow process, enhancing retention and performance.\u003c/p\u003e \u003cp\u003eBeyond theories of the declarative spacing effect, two hypotheses with regard to motor learning suggest that spacing may enhance learning. Reactive inhibition (RI) is a well-established empirical phenomenon of worsening motor performance during continuous practice which dissipates during breaks\u003csup\u003e\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e,\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e,\u003cspan additionalcitationids=\"CR26 CR27\" citationid=\"CR25\" class=\"CitationRef\"\u003e25\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e28\u003c/span\u003e\u003c/sup\u003e (Fig.\u0026nbsp;1; for a candidate neurological mechanistic account of RI, see B\u0026auml;chinger and colleagues\u003csup\u003e\u003cspan citationid=\"CR29\" class=\"CitationRef\"\u003e29\u003c/span\u003e\u003c/sup\u003e). RI is more pronounced when practice is sustained for a long period (massed practice), but less pronounced when practice is short with frequent breaks (spaced practice). Although the classic effect of RI is a transient worsening of performance, several studies have addressed the possibility that RI may also negatively affect motor learning. Across alphabet printing, peg board learning, the stabilometer task, and the Tsai-Partington numbers tasks, every possible result has been found; greater learning with spaced training\u003csup\u003e\u003cspan citationid=\"CR30\" class=\"CitationRef\"\u003e30\u003c/span\u003e,\u003cspan citationid=\"CR31\" class=\"CitationRef\"\u003e31\u003c/span\u003e\u003c/sup\u003e, no learning difference between massed and spaced training\u003csup\u003e\u003cspan citationid=\"CR32\" class=\"CitationRef\"\u003e32\u003c/span\u003e,\u003cspan citationid=\"CR33\" class=\"CitationRef\"\u003e33\u003c/span\u003e\u003c/sup\u003e, and greater learning with massed training\u003csup\u003e\u003cspan citationid=\"CR34\" class=\"CitationRef\"\u003e34\u003c/span\u003e\u003c/sup\u003e. In a meta-analysis, Lee and Genovese\u003csup\u003e\u003cspan citationid=\"CR35\" class=\"CitationRef\"\u003e35\u003c/span\u003e\u003c/sup\u003e concluded that massed practice impairs both performance and learning. However, given the variety of findings, methodological limitations, and small sample sizes in some cases, we view that literature as inconclusive. If RI does attenuate motor learning, then reducing it through spacing would yield a spacing effect on a delayed test\u003c/p\u003e \u003cp\u003eThe new theory of \u003cem\u003emicro-consolidation\u003c/em\u003e posits that motor learning occurs primarily or exclusively during breaks (offline), driven by hippocampal to neocortical replay of learned sequences on the time scale of seconds\u003csup\u003e\u003cspan additionalcitationids=\"CR37 CR38\" citationid=\"CR36\" class=\"CitationRef\"\u003e36\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR39\" class=\"CitationRef\"\u003e39\u003c/span\u003e\u003c/sup\u003e. This theory is akin to the proposed encoded memory reactivation mechanism of the spacing effect in declarative learning. Hence, frequent breaks between brief performance trials (spaced practice) should promote learning due to greater memory reactivations. In apparent support of this theory, motor sequence performance immediately after a break is often better than at the end of the preceding training trial\u003csup\u003e\u003cspan citationid=\"CR36\" class=\"CitationRef\"\u003e36\u003c/span\u003e,\u003cspan citationid=\"CR37\" class=\"CitationRef\"\u003e37\u003c/span\u003e,\u003cspan citationid=\"CR39\" class=\"CitationRef\"\u003e39\u003c/span\u003e\u003c/sup\u003e.\u003c/p\u003e \u003cp\u003eFinally, there is the possibility that spaced practice has no effect at all on motor learning, as proposed by Gupta and Rickard\u003csup\u003e\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e,\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e\u003c/sup\u003e. They showed that the response time slowing due to accrual of RI across performance trials, along with dissipation of that RI during breaks, provides a viable account for post-break performance improvements without invoking offline micro-consolidation. The question of whether spacing effects occur in the current motor sequence task is also informed by several empirical studies of other motor tasks. Spacing effects have been observed for several tasks, including the rotary pursuit test, typing, discrete linear movement learning, computer game learning, dynamic balancing, and motor sequence learning\u003csup\u003e\u003cspan additionalcitationids=\"CR41 CR42 CR43 CR44 CR45\" citationid=\"CR40\" class=\"CitationRef\"\u003e40\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR46\" class=\"CitationRef\"\u003e46\u003c/span\u003e\u003c/sup\u003e. In contrast, investigations into other complex motor skills such as cup stacking and piano learning\u003csup\u003e\u003cspan citationid=\"CR47\" class=\"CitationRef\"\u003e47\u003c/span\u003e,\u003cspan citationid=\"CR48\" class=\"CitationRef\"\u003e48\u003c/span\u003e\u003c/sup\u003e and simple sequence learning in both humans and monkeys\u003csup\u003e\u003cspan citationid=\"CR49\" class=\"CitationRef\"\u003e49\u003c/span\u003e\u003c/sup\u003e have shown either a lack of spacing effects or inconsistent benefits across measured variables. Thus, the relatively limited number of studies investigating the spacing effect in motor skill learning do not provide clear guidance on what to expect in the current experiments. Further, there appears to have been no prior studies focused on investigating spacing effects for the finger tapping task investigated here, which has dominated the extensive literature on the hypothesis of sleep-based motor sequence memory consolidation over the last three decades\u003csup\u003e\u003cspan citationid=\"CR50\" class=\"CitationRef\"\u003e50\u003c/span\u003e\u003c/sup\u003e.\u003c/p\u003e \u003cp\u003eWe investigated whether spacing and RI affect motor sequence learning by manipulating both trial length (the number of sequences within each trial) and the duration of breaks between trials. In Experiment 1, both groups learned to perform a five-item (five keypress) sequence. The \u003cem\u003espaced\u003c/em\u003e group performed five correct sequences during each trial with 30 s breaks between trials. The \u003cem\u003emassed\u003c/em\u003e group performed 25 correct sequences during each trial with 10 s breaks. After a 15-minute rest period during which the cumulative effect of RI over training trials is expected to fully resolve, there was a test involving an identical spaced task for both groups, with five correct sequences per trial and 30 s breaks. The total number of training and test sequences was the same for the two groups. Because both groups performed the same task on the test, any learning and RI effects across test trials should be equated. Hence, any performance difference between groups on the test should exclusively reflect differences in the amount of learning during training. In Experiment 2, we increased the sequence length to nine-items. The spaced group performed four correct sequences per trial with 30 s breaks, while the massed group performed 12 correct sequences with 10 s breaks. All other design aspects mirrored those of Experiment 1.\u003c/p\u003e \u003cp\u003eBased on prior results\u003csup\u003e\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e,\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e\u003c/sup\u003e, the effect of RI on performance in the spaced group should be limited primarily to sequences within individual trials, with minimal RI accrual across trials (i.e., each 30 s break should be sufficient to dissipate most of the RI that accrued during the preceding trial). However, for the massed group there should be substantial accrual of unresolved RI across trials. Hence, if RI impairs learning in a dose-response manner, we should see better final test performance in the spaced group (a spacing effect). In addition, because there is a much longer total break time across training in the spaced group (1,020 s, vs 70 s in the massed group), there is more opportunity for offline micro-consolidation in the spaced group, again suggesting a positive spacing effect.\u003c/p\u003e \u003cp\u003eIn summary, we should observe better test phase performance in the spaced group if RI impairs learning and (or) if spaced training enhances learning through other spacing mechanisms, including micro-consolidation. In contrast, we should observe equivalent test performance in the two groups if each of the following conditions are true: (1) RI is exclusively a performance phenomenon, (2) the greater break frequency and duration for the spaced group in does not promote more offline micro-consolidation, and (3) none of the spacing mechanisms in theories developed for declarative learning generalize to the case of single-session motor sequence learning. The latter possibility is consistent with the hypothesis advanced by Gupta and Rickard\u003csup\u003e\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e,\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e\u003c/sup\u003e.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e"},{"header":"Results","content":"\u003cdiv id=\"Sec3\" class=\"Section2\"\u003e \u003ch2\u003eErrors\u003c/h2\u003e \u003cp\u003eThe error rate was calculated for each participant as the number of incorrect keypresses prior to each correct sequence within each trial. In Experiment 1, averaging over sequences, trials, and participants in the training phase, the error rate was 0.38 and 0.41 keypresses in the spaced and massed groups, respectively. A Bayes factor test on the error rate positively favored the null between groups, \u003cem\u003eBF\u003c/em\u003e\u003csub\u003e\u003cem\u003e01\u003c/em\u003e\u003c/sub\u003e\u0026thinsp;=\u0026thinsp;5.25, \u003cem\u003ed\u003c/em\u003e = -0.056. On the test trials the overall error rate was 0.44 and 0.17 keypresses in the spaced and massed groups, respectively; positively favoring the alternative, \u003cem\u003eBF\u003c/em\u003e\u003csub\u003e\u003cem\u003e10\u003c/em\u003e\u003c/sub\u003e\u0026thinsp;=\u0026thinsp;8.28, \u003cem\u003ed\u003c/em\u003e = -0.32.\u003c/p\u003e \u003cp\u003eIn Experiment 2, averaging over sequences, trials, and participants in the training phase, the error rate was 1.26, and 1.73, keypresses in the spaced and massed groups, respectively. A Bayes factor test on the error rate positively favored a difference between groups, \u003cem\u003eBF\u003c/em\u003e\u003csub\u003e\u003cem\u003e10\u003c/em\u003e\u003c/sub\u003e\u0026thinsp;=\u0026thinsp;5.08, \u003cem\u003ed\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.282. On the test, the error rate was 0.99 and 0.6 keypresses in the spaced and massed groups, respectively. A Bayes factor test on the error rate positively favored an effect between groups \u003cem\u003eBF\u003c/em\u003e\u003csub\u003e\u003cem\u003e10\u003c/em\u003e\u003c/sub\u003e\u0026thinsp;=\u0026thinsp;8.0, \u003cem\u003ed\u003c/em\u003e = -0.3.\u003c/p\u003e \u003c/div\u003e\n\u003ch3\u003eWithin-Trial RI\u003c/h3\u003e\n\u003cp\u003eTo confirm the accrual of RI over correct sequences within trial in the massed group, we fit a Bayesian linear mixed-effects model with RT predicted by Sequence position its interaction with Group, including random intercepts and slopes by participant and a random intercept for trials within participant. In the massed group, RT increased with sequence position, BF\u003csub\u003e10\u003c/sub\u003e\u0026thinsp;=\u0026thinsp;3.09e\u003csup\u003e57\u003c/sup\u003e, \u003cem\u003eβ\u003c/em\u003e\u0026thinsp;=\u0026thinsp;5.09 ms/sequence. Similarly in the spaced group RT increased with sequence position, BF\u003csub\u003e10\u003c/sub\u003e\u0026thinsp;=\u0026thinsp;4.64e\u003csup\u003e8\u003c/sup\u003e, \u003cem\u003eβ\u003c/em\u003e\u0026thinsp;=\u0026thinsp;23.33 ms/sequence, In Experiment 2, the massed group shows that RT increased robustly with sequence position, BF\u003csub\u003e10\u003c/sub\u003e\u0026thinsp;=\u0026thinsp;2.85e\u003csup\u003e13\u003c/sup\u003e, \u003cem\u003eβ\u003c/em\u003e\u0026thinsp;=\u0026thinsp;58.24 ms/sequence. The spaced group yielded a similarly strong positive slope, BF\u003csub\u003e10\u003c/sub\u003e\u0026thinsp;=\u0026thinsp;2.09e\u003csup\u003e8\u003c/sup\u003e, \u003cem\u003eβ\u003c/em\u003e\u0026thinsp;=\u0026thinsp;62.18 ms/sequence.\u003c/p\u003e\n\u003ch3\u003eTraining RTs\u003c/h3\u003e\n\u003cp\u003eNext, we investigated whether spacing had an effect on performance by the end of training. To test performance differences at the end of training between the massed and spaced groups in Experiment 1, a Bayes factor test very positively favored a group difference between the mean RTs for the last 25 training sequences, performed, \u003cem\u003eBF\u003c/em\u003e\u003csub\u003e\u003cem\u003e1\u003c/em\u003e0\u003c/sub\u003e\u0026thinsp;=\u0026thinsp;6243.1, \u003cem\u003ed\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.54, indicating greater accrual of cumulative RI in the massed group, as expected (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e3\u003c/span\u003ea). In Experiment 2, a Bayes factor test again very positively favored the alternative hypothesis, yielding \u003cem\u003eBF\u003c/em\u003e\u003csub\u003e\u003cem\u003e10\u003c/em\u003e\u003c/sub\u003e\u0026thinsp;=\u0026thinsp;12.66e\u003csup\u003e76\u003c/sup\u003e, \u003cem\u003ed\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.68, (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e3\u003c/span\u003ec).\u003c/p\u003e\n\u003ch3\u003ePost-rest Gain\u003c/h3\u003e\n\u003cp\u003eConfirming the interaction between group and experimental phase (end of training vs. test) in Experiment 1 (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e3\u003c/span\u003ea), a Bayes factor test very strongly favored a difference in the mean RT difference scores (mean RT on the last 25 training sequences minus the first 25 test sequences) between the massed and spaced groups \u003cem\u003eBF\u003c/em\u003e\u003csub\u003e\u003cem\u003e10\u003c/em\u003e\u003c/sub\u003e\u0026thinsp;=\u0026thinsp;2e\u003csup\u003e5\u003c/sup\u003e, \u003cem\u003ed\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.68. In Experiment 2 (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e3\u003c/span\u003ec), a Bayes factor test again very strongly favored a difference on the participant-level mean RT difference scores (mean RT on the last 24 training sequences minus first 24 test sequences) between the massed vs. spaced groups was significant, \u003cem\u003eBF\u003c/em\u003e\u003csub\u003e\u003cem\u003e10\u003c/em\u003e\u003c/sub\u003e\u0026thinsp;=\u0026thinsp;7.95e\u003csup\u003e14\u003c/sup\u003e, \u003cem\u003ed\u003c/em\u003e\u0026thinsp;=\u0026thinsp;1.0.\u003c/p\u003e\n\u003ch3\u003eFinal Test RTs\u003c/h3\u003e\n\u003cp\u003eThe key comparison of this experiment is the grand mean correct sequence RT difference between groups. Critically, all accumulation of RI across training trials is expected to have dissipated during the rest period, and because on the test both groups are performing the spaced task (five sequences / trial with 30 s breaks), any learning and RI effects that occur over test trials should be equivalent. In Experiment 1, the Bayes factor test weakly favored the null, \u003cem\u003eBF\u003c/em\u003e\u003csub\u003e\u003cem\u003e01\u003c/em\u003e\u003c/sub\u003e\u0026thinsp;=\u0026thinsp;2.76, \u003cem\u003ed\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.14 (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e3\u003c/span\u003eb). In Experiment 2, a Bayes factor test weakly favored the null on the group RT means of all the test sequences, \u003cem\u003eBF\u003c/em\u003e\u003csub\u003e\u003cem\u003e01\u003c/em\u003e\u003c/sub\u003e\u0026thinsp;=\u0026thinsp;2.51, \u003cem\u003ed\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.15 (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e3\u003c/span\u003ed).\u003c/p\u003e \u003cp\u003e \u003c/p\u003e"},{"header":"Discussion","content":"\u003cp\u003eWe investigated whether spaced practice enhances motor skill learning in two experiments. During training in both experiments, a larger amount of RI was induced in the massed group by limiting the number and duration of breaks, and by requiring participants to complete a relatively large number of correct sequences per trial. After a 15-minute rest to dissipate RI, both groups performed the spaced task from the training phase. Despite the large performance differences at the end of training in both experiments, test performance was statistically equivalent in the two groups, suggesting that neither RI, micro-consolidation, nor other candidate mechanisms for spacing effects influenced training phase learning.\u003c/p\u003e \u003cp\u003eIn both experiments there was a non-significant effect (\u003cem\u003ed\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.14; d\u0026thinsp;=\u0026thinsp;0.15) toward better final test performance in the spaced group, and thus we cannot rule-out a small advantage in learning for that group. It is possible that the 15-minute rest period was insufficient to completely dissipate RI. However, that possibility seems unlikely given that on average the training session lasted around seven minutes, not including breaks, with the 15-minute rest period more than doubling that duration. Further, participants had the same number of total correct keypresses within each experiment. Regardless, any difference in achieved learning between the two groups is at best minimal. That apparent absence of a spacing effect was surprising in light of previous evidence that learning during spaced training is superior to learning during massed training, not only in declarative memory tasks, but also in some types of cognitive skill learning\u003csup\u003e\u003cspan citationid=\"CR51\" class=\"CitationRef\"\u003e51\u003c/span\u003e,\u003cspan citationid=\"CR52\" class=\"CitationRef\"\u003e52\u003c/span\u003e\u003c/sup\u003e.\u003c/p\u003e \u003cp\u003eThe conclusions in the preceding paragraph are drawn solely from the correct sequence RT results. Conceivably, greater error rates in one group may have obfuscated learning differences. In Experiment 1, the error rate was statistically equivalent during training, whereas the spaced group had a statistically higher error rate on the test. In Experiment 2, there were statistically higher error rates during training for the massed group, but no statistical differences in error rate on the test. However, both of those effects were weakly favored the null as indicated by the Bayes factors tests. Previously, Gupta and Rickard\u003csup\u003e\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e\u003c/sup\u003e speculated that there may be a speed-accuracy tradeoff, wherein the quicker spaced group has more errors than the slower massed group. However, those results do not fully replicate in this study. Hence, across several experiments there is no consistent pattern of error differences between massed and spaced groups. The simplest interpretation of our RT results is that the same amount of learning occurred in both groups across the two experiments, consistent with the online learning account advanced by Gupta and Rickard\u003csup\u003e\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e,\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e\u003c/sup\u003e. In that account, learning occurs during each executed sequence (and not during breaks) and is independent of the schedule of trials and breaks. Because the same number of sequences were performed in both the massed and spaced groups, the online account predicts equivalent learning by the end of training and hence equivalent performance on the test. In this account, learning is equivalent in the groups and all performance differences during training are solely due to differences in RI accrual and dissipation.\u003c/p\u003e \u003cp\u003eOur findings place boundary conditions on the micro-consolidation account of learning. First, if all of the training phase learning occurred during breaks, then the amount of learning that occurred during each break must have been substantially greater in the massed groups (e.g., with eight 10 s breaks) than in the spaced groups (e.g., with twenty-seven 30 s breaks), given the equivalent performance on the test. The micro-consolidation theory, as developed to date, does not specify whether or how the rate of offline consolidation depends on either the number of sequences practiced within each trial or the time available during each break, and it remains to be seen whether it can accommodate the current results. Buch and colleagues\u003csup\u003e\u003cspan citationid=\"CR39\" class=\"CitationRef\"\u003e39\u003c/span\u003e\u003c/sup\u003e observed that hippocampal activity associated with offline micro-consolidation occurred throughout the entire 10 s break periods in their study. If there is a dose-response relationship between break time and amount of micro-consolidation, the spaced groups should have shown greater overall learning. Instead, our results align more closely with recent evidence from Das and colleagues\u003csup\u003e\u003cspan citationid=\"CR53\" class=\"CitationRef\"\u003e53\u003c/span\u003e\u003c/sup\u003e, who showed that short rest periods in a similar finger-tapping task produce micro-offline performance gains during training but no lasting advantage in later tests relative to continuous practice. They further demonstrated that such gains are present even for random, nonrepeating sequences, where no micro-consolidation could occur.\u003c/p\u003e \u003cp\u003eOur results further suggest that the mechanisms proposed in extant spacing effect theories in declarative learning do not extend to motor skill learning, at least for the current finger tapping task. In particular, these results are inconsistent with the Bayesian Learner model\u003csup\u003e\u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e24\u003c/span\u003e\u003c/sup\u003e, which predicts domain-general spacing benefits. Other theories of the declarative spacing effect posit that spacing benefits learning through the reactivation of memories, strengthening the original memory itself, or associations between items and contexts, like in the Multiscale Context Model\u003csup\u003e\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e\u003c/sup\u003e and the Predictive Performance Equation models\u003csup\u003e\u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e7\u003c/span\u003e\u003c/sup\u003e. This reactivation may be particularly effective when there is sufficient spacing between repetitions, possibly due to memory traces being primarily mediated by the cortex and out of working memory\u003csup\u003e\u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e,\u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e23\u003c/span\u003e\u003c/sup\u003e. However, reactivation may not benefit motor skill learning in the same way as declarative memories, possibly due to differing neural computations in areas like the basal ganglia and primary motor cortex\u003csup\u003e\u003cspan citationid=\"CR54\" class=\"CitationRef\"\u003e54\u003c/span\u003e\u003c/sup\u003e. Further, motor memories are much more resistant to forgetting\u003csup\u003e\u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e20\u003c/span\u003e,\u003cspan citationid=\"CR55\" class=\"CitationRef\"\u003e55\u003c/span\u003e\u003c/sup\u003e and thus reactivation may only give a marginal benefit to the memory trace strength, resulting in minimal spacing effects.\u003c/p\u003e \u003cp\u003eAlong with memory reactivations, spaced practice in declarative learning is thought to increase contextual variability, where each spaced repetition occurs in a distinct temporal or environmental context, enhancing memory strength through contextual reinstatement and context diversity\u003csup\u003e\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e,\u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e,\u003cspan citationid=\"CR56\" class=\"CitationRef\"\u003e56\u003c/span\u003e\u003c/sup\u003e. In contrast, motor skill learning may be more influenced by physical and sensorimotor context variability, such as changes in movement patterns, speed, or force application\u003csup\u003e\u003cspan citationid=\"CR57\" class=\"CitationRef\"\u003e57\u003c/span\u003e,\u003cspan citationid=\"CR58\" class=\"CitationRef\"\u003e58\u003c/span\u003e\u003c/sup\u003e. While spacing might alter the temporal context, we speculate that it does not influence the internal sensorimotor context that drives motor memory formation and retention. As a result, the limited ability of spacing to modify these critical features may constrain its impact on motor learning. Instead, other forms of variability (e.g., interleaved practice schedules) have yielded more robust effects on motor skill retention and transfer\u003csup\u003e\u003cspan additionalcitationids=\"CR60 CR61\" citationid=\"CR59\" class=\"CitationRef\"\u003e59\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR62\" class=\"CitationRef\"\u003e62\u003c/span\u003e\u003c/sup\u003e.\u003c/p\u003e \u003cp\u003eOur findings mirror earlier studies comparing massed and spaced practice with sleep during the retention period\u003csup\u003e\u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e26\u003c/span\u003e,\u003cspan citationid=\"CR27\" class=\"CitationRef\"\u003e27\u003c/span\u003e,\u003cspan citationid=\"CR50\" class=\"CitationRef\"\u003e50\u003c/span\u003e,\u003cspan citationid=\"CR63\" class=\"CitationRef\"\u003e63\u003c/span\u003e\u003c/sup\u003e. Once confounding factors such as circadian phase and data averaging are controlled for in those studies, neither massed or spaced training yielded overnight gains \u0026mdash; or losses \u0026mdash; in performance. In our current study, we likewise observed equivalent post-rest performance across groups. This convergence across studies suggests that, for motor sequence tasks, neither spacing nor sleep enhances learning.\u003c/p\u003e \u003cp\u003eOur results and others\u003csup\u003e\u003cspan citationid=\"CR49\" class=\"CitationRef\"\u003e49\u003c/span\u003e,\u003cspan citationid=\"CR53\" class=\"CitationRef\"\u003e53\u003c/span\u003e\u003c/sup\u003e suggest that spaced practice confers little or no benefit in relatively simple motor sequence learning tasks. Nevertheless, there is ample evidence that declarative processes contribute to motor learning\u003csup\u003e\u003cspan additionalcitationids=\"CR65 CR66\" citationid=\"CR64\" class=\"CitationRef\"\u003e64\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR67\" class=\"CitationRef\"\u003e67\u003c/span\u003e\u003c/sup\u003e. The benefits of spaced practice may depend on the stage of learning it targets (e.g., early declarative-supported learning vs later procedural refinement), and this timing is likely task specific. For example, spacing may confer greater benefit to more complex motor skills that have a protracted declarative learning phase, such as driving or typing. This possibility points to an important direction for future research on spacing: identifying the time course of declarative involvement in motor learning and determining whether and for which types of tasks spacing influences that component of learning.\u003c/p\u003e \u003cp\u003ePossibly, the short time frame of the current experiments (~\u0026thinsp;30 minutes) was insufficient for spacing to produce a noticeable benefit in performance. For declarative memory, the largest effects are often found when spacing between sessions spans days and the final test is also days after the last training session\u003csup\u003e\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e,\u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e19\u003c/span\u003e\u003c/sup\u003e. However, in a meta-analysis of that literature by Cepeda and colleagues\u003csup\u003e\u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e\u003c/sup\u003e, modest spacing effects were found for declarative tasks when spacing was similar to that in the current tasks (between 1\u0026ndash;59 s between trials with retention intervals between 1 to 10 minutes), suggesting that our conclusions are not idiosyncratic to our timeframe. Further, the possibility that spacing effects for motor tasks are more robust at longer time intervals undermines neither our conclusion that RI is independent of motor skill learning nor diminish our conclusions that pose serious issues for the micro-consolidation account.\u003c/p\u003e \u003cp\u003eOur results raise the possibility that massed motor skill training, and the associated RI, is not detrimental to naturalistic motor sequence learning \u0026ndash; such as training on a musical instrument or sport \u0026ndash; even when the worsening of performance due to RI is palpable. However, there may be factors that limit the generalizability of that conclusion (c.f., Wiseheart and colleagues\u003csup\u003e\u003cspan citationid=\"CR47\" class=\"CitationRef\"\u003e47\u003c/span\u003e\u003c/sup\u003e). In the current tasks, for example, the error rate was not systematically higher in the massed groups. Longer trial durations, particularly if accompanied by a large error rate, may hamper learning. That possibility hints at an optimization scenario wherein the practical advantage of fewer, relatively long training trials is balanced against the possibility of error-driven impaired learning. That pattern would be consistent with recent findings for a motor skill game in which an optimal error rate for learning was observed\u003csup\u003e\u003cspan citationid=\"CR68\" class=\"CitationRef\"\u003e68\u003c/span\u003e\u003c/sup\u003e.\u003c/p\u003e \u003cp\u003eIn conclusion, our findings suggest that models of the spacing effects for declarative tasks do not generalize to motor sequence learning. Instead, our results align with the online learning account, wherein motor learning occurs during performance and is unaffected by breaks or RI. This result places important boundary conditions on existing theories, such as the Bayesian Learner model and micro-consolidation accounts, and highlights the need to better understand the unique mechanisms that underlie effective motor skill learning.\u003c/p\u003e"},{"header":"Methods","content":"\u003cdiv id=\"Sec10\" class=\"Section2\"\u003e \u003ch2\u003eParticipants\u003c/h2\u003e \u003cp\u003eTo determine that sample size for Experiment 1, we conducted an a priori power analysis using G*Power\u003csup\u003e69\u003c/sup\u003e. Based on the final-test difference between spaced and massed practice reported by Gupta and Rickard\u003csup\u003e\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e\u003c/sup\u003e \u0026mdash; with α\u0026thinsp;=\u0026thinsp;.05, power = .80, effect size d = .49, and a two-tailed independent-samples t-test, the required sample was 163 participants. We recruited 171 right-handed participants: 85 in the massed group (M age\u0026thinsp;=\u0026thinsp;20.8 years, 76.5% female) and 86 in the spaced group (M age\u0026thinsp;=\u0026thinsp;20.1 years, 81.4% female). Following the data-cleaning procedure described in Data Analysis and Cleaning, eight participants were excluded (six massed, two spaced), yielding a final sample of 163.\u003c/p\u003e \u003cp\u003eIn Experiment 2, we aimed to recruit a greater number of participants in order to replicate Experiment 1. We recruited 196 right-handed participants, 98 in the massed group (age\u0026thinsp;=\u0026thinsp;20.5, F\u0026thinsp;=\u0026thinsp;73.4%) and 98 in the spaced group (age\u0026thinsp;=\u0026thinsp;21.1, F\u0026thinsp;=\u0026thinsp;76.5%). Ten participants (six from the spaced group and four from the massed group) were removed using the cleaning procedure described in Data Analysis and Cleaning, leaving 186 participants.\u003c/p\u003e \u003cp\u003eThe experiment was conducted online and participants were recruited from the University of California, San Diego SONA undergraduate participant pool. Participants provided informed consent via button press prior to participation. All procedures were approved by the Institutional Review Board of the University of California, San Diego, and were conducted in accordance with relevant guidelines and regulations and in compliance with the Declaration of Helsinki.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec11\" class=\"Section2\"\u003e \u003ch2\u003eExperimental design and procedure\u003c/h2\u003e \u003cp\u003eParticipants performed a classic finger-tapping-task in which they repeated the sequence, 4-1-3-2-4, as quickly and accurately as possible with their non-dominant left hand\u003csup\u003e\u003cspan citationid=\"CR70\" class=\"CitationRef\"\u003e70\u003c/span\u003e\u003c/sup\u003e (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e2\u003c/span\u003e). Throughout the experiment, the numbered key sequence was displayed horizontally on the computer screen. Each keypress response was indicated by adding a \u0026lsquo;*\u0026rsquo; to a single row underneath the displayed sequence. This did not indicate if the response was correct or incorrect, only that a key was pressed. Between trials, a displayed countdown timer indicated how much break time was left. Participants performed one correct warmup sequence before starting the main task. They had 10 s to do so or had to restart the warmup trial. The warmup trial was not included in the analyses.\u003c/p\u003e \u003cp\u003eA between-participant design was used, in which the massed group completed 25 correct sequences per training trial with 10 s breaks between trials and the spaced group completed 5 correct sequences per training trial with 30 s breaks. After 175 completed sequences during the training phase in both groups, there was a 15-minute rest wherein participants in both groups performed a distraction task of double-digit addition. During the subsequent test both groups performed 10 trials with 5 correct sequences per trial and 30 s breaks.\u003c/p\u003e \u003cp\u003eIn Experiment 2, participants performed the same task as in Experiment 1, however, the sequence length was increased to nine-items, 4-1-3-2-4-2-3-1-4 (we reversed the first four items of the original sequence and then appended it to the end). Participants in the massed group completed 12 correct sequences per training trial with 10 s breaks between trials. The spaced group completed four correct sequences per training trial with 30 s breaks. After 108 completed sequences during training, there was a 15-minute rest wherein participants both groups performed a distraction task of double-digit addition. Before the test trials, participants completed a single warmup sequence and then had a break. Afterwards, both groups performed 15 test trials with four sequences per trial and 30 s breaks. All other aspects of the methods were identical to those of Experiment 1.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec12\" class=\"Section2\"\u003e \u003ch2\u003eData Analysis and Cleaning\u003c/h2\u003e \u003cp\u003eThe dependent measure was time in milliseconds to complete a correct sequence. Keypress latency within sequence was measured as the time (in milliseconds) between consecutive keypresses. To reduce noise in the data, we log-transformed the keypresses latencies. The mean of the logged keypresses was then calculated for each sequence and participant. We then exponentiated those means and multiplied them by the number of correct keypresses in the sequence (five for Experiment 1 and nine for Experiment 2) to obtain a measure of sequence RT in milliseconds. The first completed sequence was removed from each trial prior to further analysis due to the consistently longer RTs on those sequences, indicative of warmup\u003csup\u003e\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e,\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e\u003c/sup\u003e.\u003c/p\u003e \u003cp\u003eData were analyzed in R\u003csup\u003e71\u003c/sup\u003e (version 4.2.1) using the \u003cem\u003etidyverse\u003c/em\u003e\u003csup\u003e\u003cspan citationid=\"CR72\" class=\"CitationRef\"\u003e72\u003c/span\u003e\u003c/sup\u003e (version 2.0.0) and \u003cem\u003eBayesFactor\u003c/em\u003e\u003csup\u003e\u003cspan citationid=\"CR73\" class=\"CitationRef\"\u003e73\u003c/span\u003e\u003c/sup\u003e (version 0.9.12\u0026ndash;4.2) packages. For Bayes factors, the Cauchy prior width was set to \u003cem\u003er\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.707\u003csup\u003e74\u003c/sup\u003e. Interpretation followed Raftery\u0026rsquo;s guidelines\u003csup\u003e\u003cspan citationid=\"CR75\" class=\"CitationRef\"\u003e75\u003c/span\u003e\u003c/sup\u003e: 1\u0026ndash;3\u0026thinsp;=\u0026thinsp;weak, 3\u0026ndash;20\u0026thinsp;=\u0026thinsp;positive, 20\u0026ndash;150\u0026thinsp;=\u0026thinsp;strong, \u0026gt; 150\u0026thinsp;=\u0026thinsp;very strong evidence for the null or alternative hypothesis. BF₀₁ denotes evidence for the null hypothesis; BF₁₀ denotes evidence for the alternative.\u003c/p\u003e \u003cp\u003eTo fit Bayesian mixed-effects models, we used \u003cem\u003ebrms\u003c/em\u003e\u003csup\u003e\u003cspan citationid=\"CR76\" class=\"CitationRef\"\u003e76\u003c/span\u003e\u003c/sup\u003e with sequence RT as the outcome. For numerical stability, all continuous variables were z-scored prior to modeling; reported effects are back-transformed to milliseconds for interpretability. Bayes factors comparing nested models were obtained by bridge sampling the marginal likelihoods of the scaled models, taking the median of multiple repetitions.\u003c/p\u003e \u003cp\u003e We conducted an unbiased data cleaning procedure to remove poor performing participants. Although our task design only allowed participants to advance after correctly executing the sequence multiple times in a trial, brief lapses in attention - such as stepping away from their keyboard - could still occur, inflating error rates and RT. Therefore, we summed each participant\u0026rsquo;s total incorrect keypresses and excluded anyone whose error count exceeded 2.5 standard deviations above the group mean. In parallel, we calculated each participant\u0026rsquo;s average RT across all trials and removed those whose mean RT fell above 2.5 standard deviations of the group mean.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec13\" class=\"Section2\"\u003e \u003ch2\u003eData availability\u003c/h2\u003e \u003cp\u003eAll data are available at \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://osf.io/ukwf9/\u003c/span\u003e\u003cspan address=\"https://osf.io/ukwf9/\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e. Further information and requests for resources should be directed to and will be fulfilled by the corresponding author, TCR (\u003cspan type=\"Underline\" class=\"Underline\" name=\"Emphasis\"\[email protected]\u003c/span\u003e).\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec14\" class=\"Section2\"\u003e \u003ch2\u003eCode Availability\u003c/h2\u003e \u003cp\u003eAll stimuli, and analyses are available at \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://osf.io/ukwf9/\u003c/span\u003e\u003cspan address=\"https://osf.io/ukwf9/\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e. Further information and requests for resources should be directed to and will be fulfilled by the corresponding author, TCR (\u003cspan type=\"Underline\" class=\"Underline\" name=\"Emphasis\"\[email protected]\u003c/span\u003e).\u003c/p\u003e \u003c/div\u003e"},{"header":"Declarations","content":"\u003cp\u003e\u003cstrong\u003eAuthor Note\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThis work was not part of a preregistered study. Materials for this study are archived at the Open Science Framework at: https://osf.io/ukwf9/\u003c/p\u003e\n\u003cp\u003e\u0026nbsp;\u003cstrong\u003eAcknowledgements\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eNone to declare.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAuthor Contributions\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eMWG conducted the experiments and analyzed the data. Both MWG and TCR designed the experiments and wrote the manuscript.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eFunding\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eNone.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eCompeting Interest Statement\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eWe have no competing interests to report.\u0026nbsp;\u003c/p\u003e\n"},{"header":"References","content":"\u003col\u003e\u003cli\u003e\u003cspan\u003eGupta, M. W. \u0026amp; Rickard, T. C. Dissipation of reactive inhibition is sufficient to explain post-rest improvements in motor sequence learning. \u003cem\u003eNpj Sci. Learn.\u003c/em\u003e \u003cb\u003e7\u003c/b\u003e, 1\u0026ndash;4 (2022).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGupta, M. W. \u0026amp; Rickard, T. C. Comparison of online, offline, and hybrid hypotheses of motor sequence learning using a quantitative model that incorporate reactive inhibition. \u003cem\u003eSci. Rep.\u003c/em\u003e \u003cb\u003e14\u003c/b\u003e, 4661 (2024).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMelton, A. W. The situation with respect to the spacing of repetitions and memory. \u003cem\u003eJ. Verbal Learn. Verbal Behav.\u003c/em\u003e \u003cb\u003e9\u003c/b\u003e, 596\u0026ndash;606 (1970).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003ePavlik, P. I. Jr \u0026amp; Anderson, J. R. Practice and Forgetting Effects on Vocabulary Memory: An Activation-Based Model of the Spacing Effect. \u003cem\u003eCogn. Sci.\u003c/em\u003e \u003cb\u003e29\u003c/b\u003e, 559\u0026ndash;586 (2005).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMozer, M. C., Pashler, H., Cepeda, N., Lindsey, R. V. \u0026amp; Vul, E. Predicting the Optimal Spacing of Study: A Multiscale Context Model of Memory. in Advances in Neural Information Processing Systems vol. 22 (Curran Associates, Inc., (2009).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eTullis, J. G. \u0026amp; Benjamin, A. S. On the effectiveness of self-paced learning. \u003cem\u003eJ. Mem. Lang.\u003c/em\u003e \u003cb\u003e64\u003c/b\u003e, 109\u0026ndash;118 (2011).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWalsh, M. M., Gluck, K. A., Gunzelmann, G., Jastrzembski, T. \u0026amp; Krusmark, M. Evaluating the Theoretic Adequacy and Applied Potential of Computational Models of the Spacing Effect. \u003cem\u003eCogn. Sci.\u003c/em\u003e \u003cb\u003e42\u003c/b\u003e, 644\u0026ndash;691 (2018).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBraun, K. \u0026amp; Rubin, D. C. The Spacing Effect Depends on an Encoding Deficit, Retrieval, and Time in Working Memory: Evidence. \u003cem\u003eMemory\u003c/em\u003e \u003cb\u003e6\u003c/b\u003e, 37\u0026ndash;66 (1998).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eThios, S. J. \u0026amp; D\u0026rsquo;Agostino, P. R. Effects of repetition as a function of study-phase retrieval. \u003cem\u003eJ. Verbal Learn. Verbal Behav.\u003c/em\u003e \u003cb\u003e15\u003c/b\u003e, 529\u0026ndash;536 (1976).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHintzman, D. L., Summers, J. J. \u0026amp; Block, R. A. What causes the spacing effect? Some effects of repetition, duration, and spacing on memory for pictures. \u003cem\u003eMem. Cognit\u003c/em\u003e. \u003cb\u003e3\u003c/b\u003e, 287\u0026ndash;294 (1975).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eCepeda, N. J., Pashler, H., Vul, E., Wixted, J. T. \u0026amp; Rohrer, D. Distributed practice in verbal recall tasks: A review and quantitative synthesis. \u003cem\u003ePsychol. Bull.\u003c/em\u003e \u003cb\u003e132\u003c/b\u003e, 354\u0026ndash;380 (2006).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003ePolyn, S. M., Norman, K. A. \u0026amp; Kahana, M. J. A context maintenance and retrieval model of organizational processes in free recall. \u003cem\u003ePsychol. Rev.\u003c/em\u003e \u003cb\u003e116\u003c/b\u003e, 129\u0026ndash;156 (2009).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMaddox, G. B. Understanding the underlying mechanism of the spacing effect in verbal learning: a case for encoding variability and study-phase retrieval. \u003cem\u003eJ. Cogn. Psychol.\u003c/em\u003e \u003cb\u003e28\u003c/b\u003e, 684\u0026ndash;706 (2016).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWright, D. L. \u0026amp; Shea, C. H. Contextual dependencies in motor skills. \u003cem\u003eMem. Cognit\u003c/em\u003e. \u003cb\u003e19\u003c/b\u003e, 361\u0026ndash;370 (1991).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eRuitenberg, M. F. L., Abrahamse, E. L., De Kleine, E. \u0026amp; Verwey, W. B. Context-dependent motor skill: perceptual processing in memory-based sequence production. \u003cem\u003eExp. Brain Res.\u003c/em\u003e \u003cb\u003e222\u003c/b\u003e, 31\u0026ndash;40 (2012).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eApolin\u0026aacute;rio-Souza, T. et al. The effect of context variability on motor learning. \u003cem\u003eHum. Mov. Sci.\u003c/em\u003e \u003cb\u003e77\u003c/b\u003e, 102794 (2021).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWixted, J. T. \u0026amp; Ebbesen, E. B. On the Form of Forgetting. \u003cem\u003ePsychol. Sci.\u003c/em\u003e \u003cb\u003e2\u003c/b\u003e, 409\u0026ndash;415 (1991).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eRubin, D. C. \u0026amp; Wenzel, A. E. One Hundred Years of Forgetting: A Quantitative Description of Retention. \u003cem\u003ePsychol. Rev.\u003c/em\u003e \u003cb\u003e103\u003c/b\u003e, 734\u0026ndash;760 (1996).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eCepeda, N. J., Vul, E., Rohrer, D., Wixted, J. T. \u0026amp; Pashler, H. Spacing Effects in Learning: A Temporal Ridgeline of Optimal Retention. \u003cem\u003ePsychol. Sci.\u003c/em\u003e \u003cb\u003e19\u003c/b\u003e, 1095\u0026ndash;1102 (2008).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSmith, C. D. et al. Memories that last in old age: motor skill learning and memory preservation. \u003cem\u003eNeurobiol. Aging\u003c/em\u003e. \u003cb\u003e26\u003c/b\u003e, 883\u0026ndash;890 (2005).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSquire, L. R. \u0026amp; Wixted, J. T. The Cognitive Neuroscience of Human Memory Since H.M. \u003cem\u003eAnnu. Rev. Neurosci.\u003c/em\u003e \u003cb\u003e34\u003c/b\u003e, 259\u0026ndash;288 (2011).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWinocur, G. \u0026amp; Moscovitch, M. Memory Transformation and Systems Consolidation. \u003cem\u003eJ. Int. Neuropsychol. Soc.\u003c/em\u003e \u003cb\u003e17\u003c/b\u003e, 766\u0026ndash;780 (2011).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eTallman, C. W., Clark, R. E. \u0026amp; Smith, C. N. Human brain activity and functional connectivity as memories age from one hour to one month. \u003cem\u003eCogn. Neurosci.\u003c/em\u003e \u003cb\u003e13\u003c/b\u003e, 115\u0026ndash;133 (2022).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKording, K. P., Tenenbaum, J. B. \u0026amp; Shadmehr, R. The dynamics of memory as a consequence of optimal adaptation to a changing body. \u003cem\u003eNat. Neurosci.\u003c/em\u003e \u003cb\u003e10\u003c/b\u003e, 779\u0026ndash;786 (2007).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHull, C. L. Principles of Behavior. \u003cem\u003eJ. Nerv. Ment Dis.\u003c/em\u003e \u003cb\u003e101\u003c/b\u003e, 396 (1945).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eRickard, T. C., Cai, D. J., Rieth, C. A., Jones, J. \u0026amp; Ard, M. C. Sleep does not enhance motor sequence learning. \u003cem\u003eJ. Exp. Psychol. Learn. Mem. Cogn.\u003c/em\u003e \u003cb\u003e34\u003c/b\u003e, 834\u0026ndash;842 (2008).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBrawn, T. P., Fenn, K. M., Nusbaum, H. C. \u0026amp; Margoliash, D. Consolidating the Effects of Waking and Sleep on Motor-Sequence Learning. \u003cem\u003eJ. Neurosci.\u003c/em\u003e \u003cb\u003e30\u003c/b\u003e, 13977\u0026ndash;13982 (2010).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eT\u0026ouml;r\u0026ouml;k, B., Janacsek, K., Nagy, D. G., Orb\u0026aacute;n, G. \u0026amp; Nemeth, D. Measuring and filtering reactive inhibition is essential for assessing serial decision making and learning. \u003cem\u003eJ. Exp. Psychol. Gen.\u003c/em\u003e \u003cb\u003e146\u003c/b\u003e, 529\u0026ndash;542 (2017).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eB\u0026auml;chinger, M. et al. Human motor fatigability as evoked by repetitive movements results from a gradual breakdown of surround inhibition. \u003cem\u003eeLife\u003c/em\u003e \u003cb\u003e8\u003c/b\u003e, e46750 (2019).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eAmmons, R. B. Acquisition of motor skill: I. Quantitative analysis and theoretical formulation. \u003cem\u003ePsychol. Rev.\u003c/em\u003e \u003cb\u003e54\u003c/b\u003e, 263\u0026ndash;281 (1947).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKimble, G. A. An experimental test of a two-factor theory of inhibition. \u003cem\u003eJ. Exp. Psychol.\u003c/em\u003e \u003cb\u003e39\u003c/b\u003e, 15\u0026ndash;23 (1949).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eCarron, A. V. Performance and Learning in a Discrete Motor Task under Massed Vs. Distributed Practice. \u003cem\u003eRes. Q. Am. Assoc. Health Phys. Educ. Recreat\u003c/em\u003e. \u003cb\u003e40\u003c/b\u003e, 481\u0026ndash;489 (1969).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eCarron, A. V. Physical Fatigue and Motor Learning. \u003cem\u003eRes. Q. Am. Assoc. Health Phys. Educ. Recreat\u003c/em\u003e. \u003cb\u003e40\u003c/b\u003e, 682\u0026ndash;686 (1969).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eArcher, E. J. Postrest performance in motor learning as a function of prerest degree of distribution of practice. \u003cem\u003eJ. Exp. Psychol.\u003c/em\u003e \u003cb\u003e47\u003c/b\u003e, 47\u0026ndash;51 (1954).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLee, T. D. \u0026amp; Genovese, E. D. Distribution of Practice in Motor Skill Acquisition: Learning and Performance Effects Reconsidered. \u003cem\u003eRes. Q. Exerc. Sport\u003c/em\u003e. \u003cb\u003e59\u003c/b\u003e, 277\u0026ndash;287 (1988).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eB\u0026ouml;nstrup, M. et al. A Rapid Form of Offline Consolidation in Skill Learning. \u003cem\u003eCurr. Biol.\u003c/em\u003e \u003cb\u003e29\u003c/b\u003e, 1346\u0026ndash;1351e4 (2019).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eB\u0026ouml;nstrup, M., Iturrate, I., Hebart, M. N., Censor, N. \u0026amp; Cohen, L. G. Mechanisms of offline motor learning at a microscale of seconds in large-scale crowdsourced data. \u003cem\u003eNpj Sci. Learn.\u003c/em\u003e \u003cb\u003e5\u003c/b\u003e, 7 (2020).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eJacobacci, F. et al. Rapid hippocampal plasticity supports motor sequence learning. \u003cem\u003eProc. Natl. Acad. Sci.\u003c/em\u003e 117, 23898\u0026ndash;23903 (2020).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBuch, E. R., Claudino, L., Quentin, R., B\u0026ouml;nstrup, M. \u0026amp; Cohen, L. G. Consolidation of human skill linked to waking hippocampo-neocortical replay. \u003cem\u003eCell. Rep.\u003c/em\u003e \u003cb\u003e35\u003c/b\u003e, 109193 (2021).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMarshall, P. H., Jones, M. T. \u0026amp; Sheehan, E. M. The Spacing Effect in Short-Term Motor Memory: The Differential Attention Hypothesis. \u003cem\u003eJ. Mot Behav.\u003c/em\u003e \u003cb\u003e9\u003c/b\u003e, 119\u0026ndash;126 (1977).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBaddeley, A. D. \u0026amp; Longman, D. J. A. The Influence of Length and Frequency of Training Session on the Rate of Learning to Type. \u003cem\u003eErgonomics\u003c/em\u003e \u003cb\u003e21\u003c/b\u003e, 627\u0026ndash;635 (1978).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBall, L. J. \u0026amp; Payne, R. B. Comparison of Inhibition and Consolidation Theories of Spaced Practice Effects. \u003cem\u003ePercept. Mot Skills\u003c/em\u003e. \u003cb\u003e66\u003c/b\u003e, 219\u0026ndash;225 (1988).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eShebilske, W. L., Goettl, B. P., Corrington, K. \u0026amp; Day, E. A. Interlesson spacing and task-related processing during complex skill acquisition. \u003cem\u003eJ. Exp. Psychol. Appl.\u003c/em\u003e \u003cb\u003e5\u003c/b\u003e, 413\u0026ndash;437 (1999).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eShea, C. H., Lai, Q., Black, C. \u0026amp; Park, J. H. Spacing practice sessions across days bene\u0026reg;ts the learning of motor skills. \u003cem\u003eHum Mov. Sci\u003c/em\u003e (2000).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSavion-Lemieux, T. \u0026amp; Penhune, V. B. The effects of practice and delay on motor skill learning and retention. \u003cem\u003eExp. Brain Res.\u003c/em\u003e \u003cb\u003e161\u003c/b\u003e, 423\u0026ndash;431 (2005).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKrigolson, O. E., Ferguson, T. D., Colino, F. L. \u0026amp; Binsted, G. Distribution of Practice Combined with Observational Learning Has Time Dependent Effects on Motor Skill Acquisition. \u003cem\u003ePercept. Mot Skills\u003c/em\u003e. \u003cb\u003e128\u003c/b\u003e, 885\u0026ndash;899 (2021).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWiseheart, M., D\u0026rsquo;Souza, A. A. \u0026amp; Chae, J. Lack of spacing effects during piano learning. \u003cem\u003ePLOS ONE\u003c/em\u003e. \u003cb\u003e12\u003c/b\u003e, e0182986 (2017).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eCecilio-Fernandes, D., Cnossen, F., Jaarsma, D. A. D. C. \u0026amp; Tio, R. A. Avoiding Surgical Skill Decay: A Systematic Review on the Spacing of Training Sessions. \u003cem\u003eJ. Surg. Educ.\u003c/em\u003e \u003cb\u003e75\u003c/b\u003e, 471\u0026ndash;480 (2018).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHikosaka, O. et al. Long-term retention of motor skill in macaque monkeys and humans. \u003cem\u003eExp. Brain Res.\u003c/em\u003e \u003cb\u003e147\u003c/b\u003e, 494\u0026ndash;504 (2002).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eRickard, T. C., Pan, S. C. \u0026amp; Gupta, M. W. Severe publication bias contributes to illusory sleep consolidation in the motor sequence learning literature. \u003cem\u003eJ. Exp. Psychol. Learn. Mem. Cogn.\u003c/em\u003e \u003cb\u003e48\u003c/b\u003e, 1787\u0026ndash;1796 (2022).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eRohrer, D. \u0026amp; Taylor, K. The effects of overlearning and distributed practise on the retention of mathematics knowledge. \u003cem\u003eAppl. Cogn. Psychol.\u003c/em\u003e \u003cb\u003e20\u003c/b\u003e, 1209\u0026ndash;1224 (2006).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eRickard, T. C., Lau, J. S. H. \u0026amp; Pashler, H. Spacing and the transition from calculation to retrieval. \u003cem\u003ePsychon Bull. Rev.\u003c/em\u003e \u003cb\u003e15\u003c/b\u003e, 656\u0026ndash;661 (2008).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDas, A., Karagiorgis, A., Diedrichsen, J., Stenner, M. P. \u0026amp; Aza\u0026ntilde;\u0026oacute;n, E. Micro-offline gains do not reflect offline learning during early motor skill acquisition in humans. \u003cem\u003eProc. Natl. Acad. Sci.\u003c/em\u003e 122, e2509233122 (2025).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHardwick, R. M., Forrence, A. D., Krakauer, J. W. \u0026amp; Haith, A. M. Time-dependent competition between goal-directed and habitual response preparation. \u003cem\u003eNat. Hum. Behav.\u003c/em\u003e \u003cb\u003e3\u003c/b\u003e, 1252\u0026ndash;1262 (2019).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLuft, A. R. \u0026amp; Buitrago, M. M. Stages of Motor Skill Learning. \u003cem\u003eMol. Neurobiol.\u003c/em\u003e \u003cb\u003e32\u003c/b\u003e, 205\u0026ndash;216 (2005).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHoward, M. W. \u0026amp; Kahana, M. J. A Distributed Representation of Temporal Context. \u003cem\u003eJ. Math. Psychol.\u003c/em\u003e \u003cb\u003e46\u003c/b\u003e, 269\u0026ndash;299 (2002).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eShea, J. \u0026amp; Morgan, R. Contextual interference effects on the acquisition, retention, and transfer of a motor skill. \u003cem\u003eJ. Exp. Psychol. [Hum Learn.\u003c/em\u003e \u003cb\u003e] 5\u003c/b\u003e, 179 (1979).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGoode, S. \u0026amp; Magill, R. A. Contextual Interference Effects in Learning Three Badminton Serves. \u003cem\u003eRes. Q. Exerc. Sport\u003c/em\u003e. \u003cb\u003e57\u003c/b\u003e, 308\u0026ndash;314 (1986).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSchmidt, R. A. \u0026amp; Bjork, R. A. New Conceptualizations of Practice: Common Principles in Three Paradigms Suggest New Concepts for Training. \u003cem\u003ePsychol. Sci.\u003c/em\u003e \u003cb\u003e3\u003c/b\u003e, 207\u0026ndash;218 (1992).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSchmidt, R. A. \u0026amp; Young, D. E. Transfer of Movement Control in Motor Skill Learning. in \u003cem\u003eTransfer of Learning\u003c/em\u003e 47\u0026ndash;79Elsevier, (1987). \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1016/B978-0-12-188950-0.50009-6\u003c/span\u003e\u003cspan address=\"10.1016/B978-0-12-188950-0.50009-6\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBrady, F. Contextual Interference: A Meta-Analytic Study. \u003cem\u003ePercept. Mot Skills\u003c/em\u003e. \u003cb\u003e99\u003c/b\u003e, 116\u0026ndash;126 (2004).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSchorn, J. M. \u0026amp; Knowlton, B. J. Interleaved practice benefits implicit sequence learning and transfer. \u003cem\u003eMem. Cognit\u003c/em\u003e. \u003cb\u003e49\u003c/b\u003e, 1436\u0026ndash;1452 (2021).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eNettersheim, A., Hallschmid, M., Born, J. \u0026amp; Diekelmann, S. The Role of Sleep in Motor Sequence Consolidation: Stabilization Rather Than Enhancement. \u003cem\u003eJ. Neurosci.\u003c/em\u003e \u003cb\u003e35\u003c/b\u003e, 6696\u0026ndash;6702 (2015).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKeisler, A. \u0026amp; Shadmehr, R. A. Shared Resource between Declarative Memory and Motor Memory. \u003cem\u003eJ. Neurosci.\u003c/em\u003e \u003cb\u003e30\u003c/b\u003e, 14817\u0026ndash;14823 (2010).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eTaylor, J. A., Krakauer, J. W. \u0026amp; Ivry, R. B. Explicit and Implicit Contributions to Learning in a Sensorimotor Adaptation Task. \u003cem\u003eJ. Neurosci.\u003c/em\u003e \u003cb\u003e34\u003c/b\u003e, 3023\u0026ndash;3032 (2014).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMcDougle, S. D., Bond, K. M. \u0026amp; Taylor, J. A. Explicit and Implicit Processes Constitute the Fast and Slow Processes of Sensorimotor Learning. \u003cem\u003eJ. Neurosci.\u003c/em\u003e \u003cb\u003e35\u003c/b\u003e, 9568\u0026ndash;9579 (2015).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKrakauer, J. W., Hadjiosif, A. M., Xu, J., Wong, A. L. \u0026amp; Haith, A. M. Motor Learning. in \u003cem\u003eComprehensive Physiology\u003c/em\u003e (ed. Terjung, R.) 613\u0026ndash;663Wiley, (2019). \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1002/cphy.c170043\u003c/span\u003e\u003cspan address=\"10.1002/cphy.c170043\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eAl-Fawakhiri, N., Kayani, S. \u0026amp; McDougle, S. D. Evidence of an optimal error rate for motor skill learning. \u003cem\u003ePreprint at.\u003c/em\u003e \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1101/2023.07.19.549705\u003c/span\u003e\u003cspan address=\"10.1101/2023.07.19.549705\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2023).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eFaul, F., Erdfelder, E., Lang, A. G. \u0026amp; Buchner, A. G*Power 3: A flexible statistical power analysis program for the social, behavioral, and biomedical sciences. \u003cem\u003eBehav. Res. Methods\u003c/em\u003e. \u003cb\u003e39\u003c/b\u003e, 175\u0026ndash;191 (2007).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eNissen, M. J. \u0026amp; Bullemer, P. Attentional requirements of learning: Evidence from performance measures. \u003cem\u003eCognit Psychol.\u003c/em\u003e \u003cb\u003e19\u003c/b\u003e, 1\u0026ndash;32 (1987).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eR Core Team. \u003cem\u003eR: A Language and Environment for Statistical Computing\u003c/em\u003e. (2024). \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://www.R-project.org/\u003c/span\u003e\u003cspan address=\"https://www.R-project.org/\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWickham, H. et al. Welcome to the tidyverse. \u003cem\u003eJ. Open. Source Softw.\u003c/em\u003e \u003cb\u003e4\u003c/b\u003e, 1686 (2019).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMorey, R. D. \u0026amp; Rouder, J. N. \u003cem\u003eBayesFactor: Computation of Bayes Factors for Common Designs\u003c/em\u003e. (2024). \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://CRAN.R-project.org/package=BayesFactor\u003c/span\u003e\u003cspan address=\"https://CRAN.R-project.org/package=BayesFactor\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWagenmakers, E. J. et al. Bayesian inference for psychology. Part II: Example applications with JASP. \u003cem\u003ePsychon Bull. Rev.\u003c/em\u003e \u003cb\u003e25\u003c/b\u003e, 58\u0026ndash;76 (2018).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eRaftery, A. E. Bayesian Model Selection in Social Research. \u003cem\u003eSociol. Methodol.\u003c/em\u003e \u003cb\u003e25\u003c/b\u003e, 111\u0026ndash;163 (1995).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eB\u0026uuml;rkner, P. C. Bayesian item response modeling in R with brms and Stan. \u003cem\u003eJ. Stat. Softw.\u003c/em\u003e \u003cb\u003e100\u003c/b\u003e, 1\u0026ndash;54 (2021).\u003c/span\u003e\u003c/li\u003e\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":false,"highlight":"","institution":"","isAcceptedByJournal":true,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"scientific-reports","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"scirep","sideBox":"Learn more about [Scientific Reports](http://www.nature.com/srep/)","snPcode":"","submissionUrl":"","title":"Scientific Reports","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"stoa","reportingPortfolio":"Scientific Reports","inReviewEnabled":true,"inReviewRevisionsEnabled":true},"keywords":"Motor sequence learning, reactive inhibition, spacing effects","lastPublishedDoi":"10.21203/rs.3.rs-8633674/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-8633674/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eSpaced practice in declarative memory tasks consistently yields greater learning than massed practice, but spacing effects are less consistently observed for motor skills. This study evaluates factors that may determine spacing effects on motor skill learning, including: (1) extant theories of declarative spacing effects, (2) reactive inhibition, which transiently impairs performance and may also impair learning, and (3) the micro-consolidation hypothesis, which posits that motor skill learning takes place exclusively during brief performance breaks. Across two experiments, we varied the number of correct sequences per trial and the length of breaks while keeping the total correct sequence count constant, using a widely studied motor sequence task. A pronounced performance advantage was observed for the spaced groups by the end of training. However, on a later test in which the task conditions were equated, group performance was statistically indistinguishable. Hence, spaced practice yielded no or minimal learning advantage and the large reactive inhibition effect in the massed group appears to be a transient performance phenomenon without consequence for learning. Furthermore, we found no evidence for the most straightforward micro-consolidation account, which predicts greater learning with more breaks. Our results are consistent with a simple account advanced by Gupta and Rickard\u003csup\u003e\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e,\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e\u003c/sup\u003e, according to which learning occurs entirely online (i.e., concurrently with performance) and is independent of spacing and reactive inhibition. Finally, our findings indicate that proposed mechanisms for declarative spacing effects, such as memory reactivation and contextual variability, do not generalize to motor learning, highlighting fundamental differences between the two learning systems.\u003c/p\u003e","manuscriptTitle":"Spaced Practice and Reactive Inhibition Have Limited or No effects on Motor Sequence Learning","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2026-02-02 15:17:13","doi":"10.21203/rs.3.rs-8633674/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"decision","content":"Revision requested","date":"2026-02-24T08:52:58+00:00","index":"","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2026-02-23T22:50:16+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2026-02-22T21:12:26+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2026-02-21T17:42:37+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2026-02-20T10:06:59+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2026-02-12T23:09:39+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"268391867230451195425686796644863213574","date":"2026-02-04T13:58:05+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"71682721979543015911943415465421590217","date":"2026-02-02T06:51:06+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"209900226031682689852715473816548999803","date":"2026-02-01T20:50:38+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"285622601490928645680481699933315941808","date":"2026-02-01T19:37:35+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"255588999818908472079140054461599602176","date":"2026-01-31T11:06:13+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"155175978641027737252016112731288364643","date":"2026-01-30T19:49:31+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"280864433009948326161466360620296157159","date":"2026-01-30T18:22:43+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"26499711872270825903742094505648115751","date":"2026-01-30T18:14:48+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"301632408722244542154803921809962541202","date":"2026-01-30T18:03:28+00:00","index":"hide","fulltext":""},{"type":"reviewersInvited","content":"","date":"2026-01-30T17:20:07+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2026-01-30T17:07:45+00:00","index":"","fulltext":""},{"type":"editorInvited","content":"","date":"2026-01-30T12:52:41+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2026-01-28T15:39:17+00:00","index":"","fulltext":""},{"type":"submitted","content":"Scientific Reports","date":"2026-01-28T14:32:53+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"scientific-reports","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"scirep","sideBox":"Learn more about [Scientific Reports](http://www.nature.com/srep/)","snPcode":"","submissionUrl":"","title":"Scientific Reports","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"stoa","reportingPortfolio":"Scientific Reports","inReviewEnabled":true,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"3b806d5a-1003-4e3c-afba-6171d70abffc","owner":[],"postedDate":"February 2nd, 2026","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"under-review","subjectAreas":[{"id":62097517,"name":"Biological sciences/Neuroscience"},{"id":62097518,"name":"Biological sciences/Psychology"},{"id":62097519,"name":"Social science/Psychology"}],"tags":[],"updatedAt":"2026-05-07T05:23:36+00:00","versionOfRecord":[],"versionCreatedAt":"2026-02-02 15:17:13","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-8633674","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-8633674","identity":"rs-8633674","version":["v1"]},"buildId":"XKTyCvWXoU3ODBz1xrDgd","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}