1894 revisited: Cross-education of skilled muscular control in women and the importance of representation.

In 1894 foundational work showed that training one limb for "muscular power" (i.e. strength) or "muscular control" (i.e. skill) improves performance in both limbs. Despite that the original data were exclusively from two female participants ("Miss Smith" and "Miss Brown"), in the decades that followed, such "cross-education" training interventions have focused predominantly on improving strength in men. Here, in a female cohort, we revisit that early research to underscore that training a task that requires precise movements in a timely fashion (i.e. "muscular control") on one side of the body is transferred to the contralateral untrained limb. With unilateral practice, women reduced time to completion and the number of errors committed during the commercially available game of Operation® Iron Man 2 with both limbs. Modest reductions in bilateral Hoffmann (H-) reflex excitability evoked in the wrist flexors suggest that alterations in the spinal cord circuitry may be related to improvements in performance of a fine motor task. These findings provide a long overdue follow-up to the efforts of Miss Theodate L. Smith from more than 125 years ago, highlight the need to focus on female participants, and advocate more study of cross-education of skilled tasks.

Introduction

In the year 1894, Edward Wheeler Scripture, Miss Theodate L. Smith and Miss Emily Brown [1] published “On the education of muscular control and power.” This study had only two participants, both women, who were also co-authors. Miss Brown (as identified in the paper) trained for ‘muscular power’ (strength training) and Miss Smith for ‘muscular control’ (skill training). Their training was performed with one arm for 9 and 10 days, respectively, and both their trained arm and the contralateral untrained arm improved strength and accuracy by 40% and 25% with ‘muscular power’ training and ‘muscular control’ training, respectively. Their work was the first to formally identify ‘cross-education’, which is the bilateral improvement in performance with unilateral training. Since that time, an entire field of research [2] has studied various aspects of cross-education (sometimes referred to as inter-limb/inter-manual transfer or the cross-transfer effect). Early experiments focused on acute (within session) learning with one hand, and showed that learning novel tasks transferred quite well to the contralateral untrained limb. Many studies have since characterized the cross-education of ‘muscular power’ described by Scripture, Smith and Brown, however only one study has closely resembled the cross-education of ‘muscular control’. In this study, Schulze et al. [3] examined the effects of unilateral training of a timed pegboard task and found bilateral improvements in time to completion after training. Although indirectly assessed (i.e. the timer kept running when a peg was dropped), there was no direct assessment of accuracy that coincided with improved time to completion. Therefore, this task did not directly assess what Scripture, Smith, and Brown originally coined as the cross-education of ‘muscular control’.

Underlying mechanisms of cross-education have been identified within cortical and subcortical regions, and within the spinal cord [4-6]. For example, reductions in interhemispheric inhibition from the ipsi- to contralateral motor cortex is well-correlated with the intermanual transfer of a serial reaction time task after ~30 minutes of motor sequence training [7]. This transcallosal pathway seems to dominate whichever model is brought forward as the leading framework for the neural processes underlying cross-education, whether “cross-activation” or “bilateral access”. Although not mutually exclusive, the two models commend modest subtleties that differentiate themselves from one another. The “cross-activation” hypothesis suggests that both cortices are activated to a certain extent during unilateral behaviours, whereas the “bilateral access” hypothesis suggests that motor plans that are formed through practice of one limb may be accessed in the future for the performance with the contralateral limb [5]. In both cases, the transcallosal pathway appears essential to the cross-education of performance, as evidenced by the increases in cortical activation in the ipsilateral sensorimotor cortex with unilateral handgrip training [8]. Below the cortical and sub-cortical regions, cross-education of strength causes alterations in reciprocal inhibition and Hoffmann (H-) reflex excitability that accompany improvements in strength suggesting that cross-education also causes plasticity of spinal origin [9]. Whether this occurs in response to cross-education of skill is less clear, even though it is well known that H-reflex excitability within the trained limb is reduced immediately following (within 10 minutes) acquisition of a novel fine motor skill [10].

The clinical relevance of cross-education of strength has recently been highlighted, such that strength training the less affected limb can facilitate strength gains and functional improvements in the untrained, more-affected limb in chronic stroke participants with accompanying neurophysiological changes at the spinal and supraspinal level [11, 12]. Although cross-education of strength at the ankle joint has shown functional improvements in walking [11], functional improvements in hand function (i.e. clinical tests of hand function) did not accompany the observed improvements in strength as a results of cross-education in the upper limb [12]. To target functional improvements in upper limb function, we suggest that the cross-education of skill in a rehabilitation setting could be an alternative to the cross-education of strength.

Cross-education spares muscle strength and size when a limb is immobilized and the opposite limb is trained [13]. Restoring symmetry of strength after musculoskeletal or neurological impairment through cross-education of strength has therefore gained substantial traction lately [14], given the vast literature available to support its efficacy (for reviews see [14, 15]). Cross-education of skill has received less attention in the clinical realm, possibly because of the inadequate support it has received in the literature since the majority of studies have focused on more acute transfer effects (within session to < 2 weeks) (for examples see [7, 16–23]). In only a couple studies, more long-term training effects were examined, but were limited to the fifth digit adduction/abduction visuomotor tracking [24] and visuomotor tracking of simple elbow flexion/extension [25]. Although these tasks are great experimental paradigms to assess changes in fine motor control, they are not very functional tasks. Therefore, we set out to closely replicate the experiment of Scripture, Smith, and Brown [1] and examine whether 5 weeks of unilateral training for highly functional ‘muscular control’ results in bilateral improvements in functional motor performance. True to the original work and in direct contrast to the vast majority of cross-education studies specifically and strength training studies generally, we included a women-only sample of participants.

To replicate in modern day the approach taken in 1894 that used an electrified drillboard, we had participants play the game of Operation® Iron Man 2 (furthermore referred to simply as “the game”) and used a combination of the time to completion and number of contacts (i.e. errors) as an index of task performance. This game requires people to use tweezers to reach into small holes and pull out plastic objects without touching the metallic edges because a contact will cause a harsh buzzer to sound. Therefore, it is a highly functional task that requires precise movements in a timely fashion (i.e. high muscular control) and closely resembles the electrified needle and peg board with bell used in 1894. We hypothesized that, similar to acute transfer found in other skilled tasks, bilateral improvements in task performance would manifest from unilateral training. Furthermore, we hypothesized that improved ‘muscular control’ would be accompanied by reductions in spinal reflex excitability, which would provide novel insights about the neural mechanisms contributing to the cross-education of skill.

Materials and methods

Participants

Nine neurologically intact young women (aged 22–24) were recruited for this study. Eight of the participants were right handed, while the other was left-handed as assessed by the Edinburgh handedness inventory [26]. The sample size was determined based on prior related work in the laboratory using similar designs and outcome measures, and were sufficient to achieve significant cross-education effects of strength training with moderate effect sizes [9, 12, 27]. Participants signed a written consent form that adhered to the protocol approved by the University of Victoria Human Research Ethics Committee.

Experimental design

The experimental timeline is in Fig 1B. This study used a multiple baseline within subject repeated-measures design, where participants completed three baseline measures and one post-test measure [12, 27, 28]. Multiple baseline tests were used to enable participants to act as their own controls. These baseline measures were obtained roughly a week apart and in order to maintain consistency, measures were recorded in the same order and environment across sessions. The post-test was completed within a week after the last training session.

Fig 1

A) A photo of the Operation® Iron Man 2 game (Hasbro, Canada) used in experiments, courtesy of G Pearcey. Objects that were removed are highlighted in green and the tweezers used by participants are highlighted in magenta. B) Miss Smith’s results from 1894. The percent of trials with errors are plotted on the primary y-axis for the untrained (yellow) and trained (black) hand. Lines indicate the number of trials performed per day and are plotted against the secondary y-axis. C) The experimental timeline. D) The group (n = 9) mean (± 95% CI) time to completion (black; primary y-axis) and number of errors committed (gold; secondary y-axis) over the training protocol. Each time point represents a single trial from the weekly training session that was performed in the laboratory. Asterisks indicate significant differences from the pre-test mean.

Bilateral measures of muscular control and strength, muscle activation, and reflex excitability were assessed during the pre- and post-test. Muscular control was quantified by examining errors and time to completion for one round of with each hand during pre- and post-testing. Muscular strength was quantified by measuring handgrip and pinch grip strength. Muscle activation was measured using electromyography (EMG) of the flexor carpi radialis (FCR), extensor carpi radialis (ECR), biceps (BB), and triceps brachii (TB) during muscular strength tests. Reflex excitability was assessed using H-reflex recruitment curves during a low-level FCR contraction (10% MVC).

Unilateral training of muscular control

Participants completed five weeks of unilateral functional training using the game for three sessions/week (twice at home and once in the lab). Each session included five rounds of the game with their dominant hand. The goal of the game was to take all 11 pieces out of Iron Man without touching the metal edges of the holes with the tweezers. An instructed order of piece removal was not specified, and participants could remove in any order of their choosing. If the tweezers touched the metal edge, an alarm sounded and an error was counted. In keeping with the early literature, participants were not explicitly instructed how to hold the tweezers (5.4cm in length with a 7mm aperture), but the most common way was to hold the tweezers between the tips of the thumb and index finger with support from the middle finger. Objects ranged in size with graspable regions ranging from 1-4mm, while the holes ranged in size with the narrowest region being 16mm (see Fig 1A for shapes of objects and holes). This game was selected as the closest commercially available approximation of the original apparatus from 1894, which was an electrified drill board into which a needle was inserted. Contact with the circumference of the drill hole closed the circuit and rang a bell. Each participant used the same game for their pre and post training measures as well as their training in the lab. However, while training at home participants had their own game to practice on but this game was an identical model to the game that they were tested on. Errors, identified by an electrified buzzer, and completion time were only recorded in lab training sessions, however feedback about the number of errors committed was not provided to participants.

Assessing muscular control

The primary outcome measure was completing the game in as little time as possible, while committing as few errors as possible, rather than simply completing a given number of trials without time restriction. Hence, we assessed both time to completion and the number of times the buzzer sounded (i.e. number of errors). During each of the three pre tests, participants played the game one time with each hand, first with the dominant hand (trained) and then with the non-dominant (untrained). Participants did not play the game with their untrained hand again until the post test. Therefore, we did not track the improvements in the untrained hand to minimize any effect from repeated tests. The game of Operation® Iron Man 2 is similar to the electrified pegboard that was initially used by Edward Scripture to examine the cross-education of skill, however the added time constraint probably added another layer of difficulty.

Assessing muscular strength

Maximal voluntary force during handgrip and pinch grip were evaluated using a hand grip dynamometer and MICROFET2 force evaluation device, respectively. Participants completed three attempts of maximal voluntary contractions (MVCs) in each hand for hand and pinch grip tests. MVCs were held for three seconds and separated by two minutes of rest. Strength assessment was recorded in a seated position with the non-tested handed placed in their lap.

Hoffmann (H) reflexes as a proxy of spinal cord plasticity

Spinal cord excitability was estimated by evoking H-reflexes in the FCR. The role of the FCR during gameplay is somewhat unclear, since its primarily involved in wrist flexion and radial deviation, but it does act at the wrist and is almost certainly involved in synergies for wrist stabilization and subtle wrist movements that contribute to task performance. Additionally, other training tasks show "spillover" to other muscles that could be detected with the approach we took. Therefore, due to the methodological convenience and experience of our lab, we evoked H-reflexes of the FCR as a proxy of spinal reflex excitability in distal arm muscles by delivering 1 ms square wave pulses to the median nerve just proximal to the medial epicondyle with bipolar surface electrodes (Thought Technology Ltd., Montreal, QC, Canada) using a Digitimer (Medtel, NSW, Australia) constant current stimulator (model DS7A). Current delivered for each stimulus was measured with a non-contact milliammeter (mA-2000, Bell Technologies, Orlando, FL, USA). H-reflexes were recorded in a seated position with the non-tested hand resting in their lap. Their tested arm was placed in customized brace that restricted movement and maintained joint angles. Each arm was fixed with the: 1) shoulder in 30 and 15 degrees abduction and flexion, respectively, 2) elbow at 110 degrees, and 3) wrist pronated with the fingers open and strapped to a wooden fixture. These measurements were taken with a manual goniometer by the same investigator for consistency. Participants were asked to maintain a low-level wrist flexion contraction (~10% of maximum), while they received visual feedback from a computer screen. Feedback consisted of a 100 ms moving average of the rectified EMG. H-reflexes were induced and recorded following procedures that have been described elsewhere [29-31]. To examine input-output properties of the H-reflex pathway, M-H recruitment curves were measured over a range of intensities with 40 stimuli delivered pseudorandomly between 1 and 3 s. Stimulus intensity was increased and decreased incrementally (ranged from 0.1 to 1 mA per increment) based on inter-individual differences in the excitability of the reflex pathway. Careful attention was taken to ensure that supramaximal M-wave amplitudes were achieved by increasing larger increments once the H-reflex amplitude started to decrease in size. Peak to peak amplitudes of the H and M waves were calculated using custom written software (Matlab, Nantick, MA) and data was then imported into custom written LabView software where it was fit with a sigmoid function [32]. We normalized the stimulation current to that required to evoke 50% of Mmax and amplitudes of M-waves and H-reflexes to Mmax. We then combined the amplitude vs current arrays for all responses in the three pre tests prior to performing a sigmoid fit. The normalized and combined recruitment curve was then compared with the sigmoid fit of the post recruitment curve. Detailed descriptions of all recruitment curve variables can be found in Klimstra and Zehr [32]. Briefly, the current at threshold was the relative stimulation current required to evoke the smallest H-reflex, current at 50% H was the relative stimulation current required to evoke an H-reflex 50% of maximum amplitude, current at H was the relative stimulation current required to evoke the maximum H-reflex. Using relative current from the pre recruitment curves, we derived comparative values at the same relative current at post. These included the H-reflex size at the current required to evoke the smallest H-reflex from pre, H-reflex size at the current required to evoke 50% of the maximal H-reflex from pre, and H-reflex size at the current required to evoke the maximal H-reflex from pre. The sigmoid fit was used to obtain these values with the procedures outlined in Klimstra and Zehr [32]. In a subset of the participants (n = 3), current intensity values were compromised from at least one of the H-reflex recruitment curves throughout the timeline. Therefore, only Hmax/Mmax ratios are reported for the entire study sample (n = 9), whereas recruitment curve variables are reported for a subset of participants that had current intensity values for both limbs and all time points (n = 6).

Electromyography

Prior to placement of surface electrodes, the skin was prepared with isopropyl alcohol swabs. Electrodes were placed bilaterally over the mid-muscle bellies of the FCR, ECR, BB and TB with an inter electrode distance of 2cm and common reference electrodes were placed on the medial epicondyles. Electrode placement was recorded at the initial baseline test to ensure each electrode was placed in the correct orientation in each subsequent test. During H-reflex recordings, FCR amplification was set to ×2000 and filter settings were adjusted to 10-1000Hz, whereas all other muscles were amplified ×5000 and band pass filtered 100–300 Hz (GRASS P511, AstroMed). Outputs were sent to the A/D interface (National Instruments Corp. TX, USA) and converted to a digital signal. EMG was sampled at 1000 Hz using custom built software (LabVIEW, National Instruments, TX, USA).

Statistical analysis

Statistical procedures were performed using GraphPad Prism (GraphPad Software, San Diego, CA). Separate 2-way (Limb × Time) repeated measures (RM) ANOVAs were used to determine whether there were main or interaction effects of time or limb on the dependent variables of time to completion, number of errors, hand grip strength, pinch grip strength, and maximal H-reflex amplitudes. Assumptions of sphericity and normality were confirmed using Mauchly’s and Kolmogorov-Smirnov tests, respectively. If significant effects of time were identified, Bonferroni’s multiple comparisons tests were used. For the trained limb, we used a 1-way RM ANOVA to determine if there was a main effect for time on time to completion or number of errors throughout the sessions in the lab. If significant effects of time were identified, Dunnett’s multiple comparison test was used to determine differences of all time points from pre. For recruitment curve measures, the variables were obtained from a sigmoid fit [32] and separate 2-way (Limb × Time) repeated measures (RM) ANOVAs were used to determine whether there were main or interaction effects of time or limb on each of the variables. In all cases, statistical significance was set at p ≤ 0.05. Results are reported as means ± SD in text (95% CI in figures). In addition to group statistics, the multiple baseline design allowed us to quantify the number of participants who showed significant improvement over the duration of the experiment by creating a 95% confidence interval from the three baseline tests. If a post-test score was below the lower limit of the baseline confidence interval, it was deemed a significant improvement in time to completion or number of errors.

Results

Timeline of training effects for the trained limb

Fig 1D shows that the time to complete and number of errors during one game of Operation® Iron Man 2 reduced quickly. Separate one way RM ANOVAs revealed that there were significant effects of time for both the time to complete (F(25, 200) = 5.596, p < 0.0001, η2 = 0.41) and number of errors (F(25, 200) = 3.481, p < 0.0001, η2 = 0.30) during one game of Operation® Iron Man 2. Times to completion for all games played in the lab, except the first three games of week one, and both the first and second games of week 2, were significantly less than pre (see Fig 1D–black asterisks). Errors were reduced during all games played in the lab sessions of weeks 3, 4 and 5 (see Fig 1D–gold asterisks).

On muscular control

Time to completion

The overall time to complete one game of Operation® Iron Man 2 did not differ between pre tests, but was significantly reduced for both the trained and untrained limb after the 5 week training program (see Fig 2A). The RM ANOVA revealed a significant effect of time (F(3, 48) = 17.56, p < 0.0001, η2 = 0.36), but no effect of limb (F(1, 16) = 2.628, p = 0.125, η2 = 0.02) or an interaction (F(3, 48) = 1.776, p = 0.164, η2 = 0.036). Post test time to completion was significantly less than pre 1 (p < 0.0001), 2 (p < 0.0001) and 3 (p = 0.0014), however there were no significant differences between pre test values, or between the trained and untrained limb. Compared to baseline, a total of 8/9 and 6/9 participants improved their time to completion over the duration of the training program for the trained and untrained limb, respectively.

Fig 2

Individual and group (n = 9) mean (± 95% CI) values are shown with individual points and red bars, respectively, for A) time to completion and B) number of errors, and C) relative improvement for both the trained and untrained limbs. Asterisks indicate significant differences from all pre values for A and B, and a significant difference between the trained and untrained limbs in C.

Number of errors

The total number of errors committed by participants during one game of Operation® Iron Man 2 did not differ between pre tests, but was significantly reduced for both the trained and untrained limb after the 5 week training program (see Fig 2B). The RM ANOVA revealed a significant effect of time (F(3, 48) = 9.35, p < 0.0001, η2 = 0.23), but no effect of limb (F(1, 16) = 0.058, p = 0.813, η2 = 0.001) or an interaction (F(3, 48) = 0.803, p = 0.499, η2 = 0.002). The number of errors in the post test was significantly less than pre 1 (p < 0.0001), 2 (p = 0.0004) and 3 (p = 0.002), however there were no significant differences between pre test values, or between the trained and untrained limb. Compared to baseline, a total of 5/9 and 5/9 participants significantly reduced the number of errors committed over the duration of the training program for the trained and untrained limb, respectively. It is important to note, however, that any participant who committed 0 errors on any of their pre tests or who had a confidence interval that extended into negative values, a significant improvement was not possible. Indeed, this was the case for 2 participants’ trained limb and 3 participants’ untrained limbs. Thus, conclusions based on these individual participant results must be made with caution as they likely underestimate the number of participants with significant improvements in performance.

Relative improvement

The relative improvement of both time to completion and number of errors is plotted in Fig 2C. Although there is a general trend for performance to improve, the mean of all three pre tests was used to calculate the relative pre-post improvement. A 2-way (limb × measure) RM ANOVA revealed a significant effect of limb (F(1, 16) = 7.507, p = 0.0145, η2 = 0.13) but no effect of measure (F(1, 16) = 1.191, p = 0.291, η2 = 0.04) or an interaction effect (F(1, 16) = 0.0148, p = 0.905, η2 = 0.002). The trained limb showed ~15% greater improvement than the untrained limb across measures.

On muscular power

Although hand grip (F(1, 16) = 11.11, p = 0.0011, η2 = 0.889) strength was higher for the trained (pre = 26.1 ± 6.0 kg, post = 25.41 ± 4.94 kg) compared to untrained limb (pre = 23.0 ± 6.35 kg, post = 22.89 ± 5.62 kg), there were no significant effects of time or interaction effects. No significant effects were revealed for pinch grip strength (trained pre = 3.65 ± 1.25 kg, trained post = 3.77 ± 1.07 kg, untrained pre = 3.42 ± 1.13 kg, trained post = 3.26 ± 1.18 kg).

On excitability of H-reflexes

The overall H-reflex excitability was modestly reduced in the trained and untrained limbs after the 5 week training program (see Fig 3). In the subset of participants with reliable stimulation current intensity data (n = 6), the RM ANOVA revealed a significant effect of time, but no effects of limb or interactions for maximal H-reflex amplitude relative to maximal M-wave amplitude (Time main effect: F(1, 10) = 6.18, p < 0.0323, η2 = 0.12), and threshold current required to evoke the H-reflex (Time main effect: F(1, 10) = 6.42, p = 0.029, η2 = 0.25). Across limbs, maximal H-reflex amplitudes decreased by 9.35% Mmax, and threshold current required to evoke the smallest possible H-reflex increased by 18% of relative current. A similar trend was revealed in the Hmax/Mmax ratio for the entire group (n = 9) and can be appreciated from a single participant in Fig 4. Group mean Hmax/Mmax ratio was reduced by 9.5 ± 8% (pre Hmax/Mmax ratio = 41 ± 10.8%, post Hmax/Mmax ratio = 36.6 ± 13.8%, d = 0.51) in the trained limb and 3.2 ± 7% (pre Hmax/Mmax ratio = 51 ± 20.6%, post Hmax/Mmax ratio = 49.4 ± 12.2%, d = 0.23) in the untrained limb from pre- to post-training, however, despite the moderate Cohen’s d values, no statistical differences were observed. Mmax amplitudes ranged from 2.792 to 13.271 mV on the untrained side and from 2.871 to 14.746 mV on the trained side but no significant effects were revealed for limb or time.

Fig 3

Individual and group (n = 6) mean (± 95% CI) values are shown with individual points and red bars.

In all panels, the untrained limb is represented with yellow fill, whereas the trained limb is represented with black fill. A-C: the current required to evoke: H-reflex threshold; 50% Hmax; and, Hmax. D-F: the amplitudes of H-reflexes at the current from pre to evoke: threshold; 50% Hmax; and, Hmax. Detailed descriptions of all variables can be found in Klimstra and Zehr [2008]. Asterisks indicate significant main effects of time.

Fig 4

Individual EMG records for each stimulation pulse (n = 40) are shown for a participant’s pre (left; black) and post (right; grey), and untrained (top) and trained (bottom) FCR M-wave and H-reflex recruitment curves. All traces are overlaid and Hmax/Mmax ratios are indicated for the trial as a percent of Mmax.

Discussion

The most important findings from this study were 1) a ~70% improvement in both time to completion and number of errors committed with the trained limb; 2) a ~55% improvement in the time to completion and number of errors committed with the untrained contralateral limb; and, 3) a modest reduction in bilateral H-reflex excitability from pre- to post-training. Collectively, these findings suggest that training for ‘muscular control’ with one limb causes bilateral improvements in performance and that reductions in spinal reflex excitability may contribute to the improved task performance. Moreover, and most importantly, our work highlights the critical importance for additional studies focusing on female participants both out of need, necessity, and equity, and out of respect for and deference to the foundational contributions to the field of cross-education from Emily Brown and Theodate Smith.

Improvements with the trained limb

It is quite surprising how quickly and the overall extent to which participants improved their performance (see Fig 1D). Improvements were observed in the first week of training for time to completion and within the second week for the number of errors committed. Observed improvements manifest more quickly than those reported for strength training [27], but align well with studies examining the learning of a novel skilled motor task. In such studies, improved performance of the trained limb is noted within a session that can be less than 30 minutes in duration [20]. Interestingly, the relative extent of improvement (i.e. % change) is > 2 times as large as the improvements in strength during similar length training interventions [33]. These findings demonstrate that practice of a task that involves ‘muscular control’ is effective at improving fine motor control, evidenced by improvements in both the time to completion and the number of errors committed.

Improvements with the untrained limb

Training the dominant limb for ‘muscular control’ resulted in performance improvements on the contralateral untrained side as well. We cannot comment on the time course of the cross-education of ‘muscular control’ because our experimental timeline would not allow, however, we can point out the relative improvement in task performance in the contralateral untrained limb compared to the trained side. In particular, it is fascinating to see that both the trained and untrained sides showed immense improvement in both time to completion and number of errors committed (see Fig 2C). Of note however, is that the untrained side relative improvement was greater than 50% on average, which is substantially greater than the improvements that have been shown in the literature for the cross-education of ‘muscular power’. A meta-analysis by Green and Gabriel [33] showed that the group average relative improvement in strength from unilateral training ranges from 15–35% and 12–29% on the trained and untrained sides, respectively. This results in a cross-body transfer (i.e. relative improvement on the untrained side compared to the trained side) that ranges from 48–80%. In our study, the cross-body transfer was on the higher end, but still within this range (72.6% for time to completion and 91.3% for number of errors committed), suggesting that the cross-education of muscular control is at least as effective as the cross-education of ‘muscular power’. This corroborates Scripture and colleagues [1] report that showed similar cross-body transfer for ‘muscular power’ (i.e. 57%) as they did for ‘muscular control’ (i.e. 55%). Together, these findings demonstrate that cross-education of ‘muscular control’ is highly effective, as evidenced by the large relative improvement with the untrained limb compared to the trained limb and the substantial cross-body transfer of the training effect.

As suggested by the “bilateral access” and “cross-activation” hypotheses [5], it is likely that transcallosal pathways are crucial for the cross-education of both muscular control and power. In both cases, it is highly likely that motor plans become available for the contralateral limb for subsequent performance (i.e. bilateral access), and ipsilateral pathways are activated during unilateral practice (i.e. cross-activation). As such, it is no surprise that similar adaptations are often observed along the corticospinal pathway between strength and skill training [34]. Where the modalities differ is in the specific circuits mediating such neural adaptations. When training for improvements in strength, increased activation of the agonist and reduced activation of antagonist muscles is vital to improved force output about a joint, which can be attributed to changes throughout the neuraxis [35, 36]. When training for improvements in fine motor control, precise control and coordination of the joint(s) are most important. Compared to self-paced strength training, unilateral strength training to a metronome and visuomotor (i.e. skill) training result in bilateral adaptations in corticospinal excitability and inhibitory cortical circuits (i.e. short-latency intra-cortical inhibition). Such adaptations in the ipsilateral cortex underlie the mechanisms proposed within the cross-activation hypothesis, suggesting that training for skill facilitates cortical adaptations associated with the cross-activation hypothesis [37]. As such, greater contributions of the cross-activation hypothesis associated with training that involves a skill could explain the greater extent of adaptations in the contralateral limb when training for muscular control (this study), compared with previous reports of the cross-education of strength.

Reduced spinal reflex excitability—An additional potential site for an underlying mechanism?

Previous work has shown that training for ‘muscular control’ causes reductions in H-reflex excitability [10]. Here, we have provided preliminary evidence for the reduction in bilateral H-reflex excitability of the upper limb as a result of training for ‘muscular control’ using the Operation® Iron Man 2 game. Since a subset of participants was used to examine H-reflex excitability, it is crucial to acknowledge to the limitations of these conclusions. For instance, we cannot conclude that the reduction in H-reflex excitability was functionally relevant and contributed to improved game performance, or whether it was a passive adaptation that occurred as a result of upstream modifications in descending neural drive. For instance, it is possible that corticospinal tracts could increase inhibition of the Ia terminals through increased presynaptic inhibition [38], which would help with fine motor control. Our data are sparse and further work is required to elucidate the neural mechanisms responsible for the cross-education of ‘muscular control’.

Important considerations

It is important to emphasize a few limitations that should be considered before drawing conclusions from this brief research report. For example, although there were no significant changes across baseline measures, visual inspection of our data may suggest a trend towards improvement in these sessions. Future studies may consider multiple post-test measures as a way to evaluate this. Further, we did not directly assess the kinematics and muscle activity during gameplay, thus limiting our ability to assess what control strategy was altered to improve performance. When participants committed an error a buzzer sounded, and there is potential that the buzzer caused some startle, especially early in gameplay, which caused time loss during the startle response. However, with repeated errors throughout training, it is possible that the participants became accustomed to the buzzer sound and therefore lost less time during each error due to this reduced startle. Although no obvious startle responses were observed in initial gameplay by the investigators, we cannot exclude this possibility. This habituation to the startle response make help explain the counterintuitive trend for reduced time to complete the game with the non-dominant hand, compared to the dominant hand (i.e. see Fig 2A).

The small sample size of participants in this study from which reliable H-reflex recruitment curves could be obtained limits the conclusions that can be drawn from the alterations observed in spinal reflex excitability, however, it still suggests that bilateral inhibition of spinal reflexes may play a role in the acquisition of fine motor skills resulting from unilateral training. Related to this, the small sample size limits our ability to assess correlations between alterations in reflex excitability and adaptations in muscular control. As such, it is unknown whether the change in reflex excitability was related to training improvements, or whether it occurred en passé. The role of the muscle we used to assess spinal reflex excitability in the successful completion of the Operation® Iron Man 2 game is not well characterized. Sampling from, for example first dorsal interosseous, thenar muscles, or flexor digitorum superficialis would be directly relevant to the pinch grip tweezer type task and fine movements of the fingers during gameplay. Therefore, any future study should try and explore excitability changes in intrinsic hand or finger flexor/extensor muscles during tasks that train muscular control.

Our sample of subjects was young (22–24) and the application to older individuals may be perceived as limited. It should be noted, however, that cross-education of strength has been observed in our lab well into their 7th decade [11, 12]. Most importantly, however, this work highlights the value of single participant case studies and illustrates some false assumptions underlying what is actually control versus control group in scientific papers. We also note with some irony the incredible contributions two young women (i.e. Ms. Smith and Ms. Brown) made to what became an enormous research enterprise (i.e. cross-education) spawning hundreds of papers but that studies using women as participants have been largely neglected in the century that followed. An exception to this trend has come from the work of Jon Farthing, who examined the application of cross-education of strength in women on multiple occasions [39-41], and two other studies that compared cross-education of strength between young and older women [42, 43]. Investigations of the effects of cross-education of skill in women, however, requires further work.

Conclusions

Although the work of Scripture, Smith and Brown [1] had only one woman for each of the conditions cross-education of ‘muscular control’ and cross-education of ‘muscular power’, their findings established an entire field of research and remain influential more than a century later. Here we corroborate and extend their findings to show, in a cohort of women, that unilateral training of a task that requires precise movements in a timely fashion (i.e. high muscular control) causes bilateral improvements in task performance. We also show that training for such ‘muscular control’ causes bilateral reductions in H-reflex excitability that may or may not contribute to improvements in fine motor control. These findings suggest that, like the cross-education of strength, cross-education of skill should be further examined for efficacy in the rehabilitation of musculoskeletal and neurological impairments. Lastly, we hope that our study focusing exclusively on women helps encourage more studies in female participants. The legacy of Miss Smith and Miss Brown, along with their colleague Edward Scripture, must be continuously acknowledged, preserved, and enhanced.

16 Feb 2021

PONE-D-20-38248

Even when only one limb is a player, both limbs get game: revisiting cross-education of skilled muscular control in women

PLOS ONE

Dear Dr. Zehr,

Thank you for submitting your manuscript to PLOS ONE. After careful consideration, we feel that it has merit but does not fully meet PLOS ONE’s publication criteria as it currently stands. Therefore, we invite you to submit a revised version of the manuscript that addresses the points raised during the review process.

The majority of reviewers were positive on the study while they made major suggestions to improve the clarity and on additional analyses. Please follow all in revising the manuscript.

Please submit your revised manuscript by Mar 29 2021 11:59PM. If you will need more time than this to complete your revisions, please reply to this message or contact the journal office at plosone@plos.org. When you're ready to submit your revision, log on to https://www.editorialmanager.com/pone/ and select the 'Submissions Needing Revision' folder to locate your manuscript file.

Please include the following items when submitting your revised manuscript:

A rebuttal letter that responds to each point raised by the academic editor and reviewer(s). You should upload this letter as a separate file labeled 'Response to Reviewers'.

A marked-up copy of your manuscript that highlights changes made to the original version. You should upload this as a separate file labeled 'Revised Manuscript with Track Changes'.

An unmarked version of your revised paper without tracked changes. You should upload this as a separate file labeled 'Manuscript'.

If you would like to make changes to your financial disclosure, please include your updated statement in your cover letter. Guidelines for resubmitting your figure files are available below the reviewer comments at the end of this letter.

If applicable, we recommend that you deposit your laboratory protocols in protocols.io to enhance the reproducibility of your results. Protocols.io assigns your protocol its own identifier (DOI) so that it can be cited independently in the future. For instructions see: http://journals.plos.org/plosone/s/submission-guidelines#loc-laboratory-protocols

We look forward to receiving your revised manuscript.

Kind regards,

Kei Masani

Academic Editor

PLOS ONE

Journal Requirements:

When submitting your revision, we need you to address these additional requirements.

1. Please ensure that your manuscript meets PLOS ONE's style requirements, including those for file naming. The PLOS ONE style templates can be found at

https://journals.plos.org/plosone/s/file?id=wjVg/PLOSOne_formatting_sample_main_body.pdf and

https://journals.plos.org/plosone/s/file?id=ba62/PLOSOne_formatting_sample_title_authors_affiliations.pdf

2. In your Methods section, please provide a justification for the sample size used in your study, including any relevant power calculations (if applicable).

3. Thank you for stating the following in the Competing Interests section:

"The authors have declared that no competing interests exist."

We note that one or more of the authors are employed by a commercial company: Zanshin Consulting, Inc.

3.1. Please provide an amended Funding Statement declaring this commercial affiliation, as well as a statement regarding the Role of Funders in your study. If the funding organization did not play a role in the study design, data collection and analysis, decision to publish, or preparation of the manuscript and only provided financial support in the form of authors' salaries and/or research materials, please review your statements relating to the author contributions, and ensure you have specifically and accurately indicated the role(s) that these authors had in your study. You can update author roles in the Author Contributions section of the online submission form.

Please also include the following statement within your amended Funding Statement.

“The funder provided support in the form of salaries for authors [insert relevant initials], but did not have any additional role in the study design, data collection and analysis, decision to publish, or preparation of the manuscript. The specific roles of these authors are articulated in the ‘author contributions’ section.”

If your commercial affiliation did play a role in your study, please state and explain this role within your updated Funding Statement.

3.2. Please also provide an updated Competing Interests Statement declaring this commercial affiliation along with any other relevant declarations relating to employment, consultancy, patents, products in development, or marketed products, etc.

Within your Competing Interests Statement, please confirm that this commercial affiliation does not alter your adherence to all PLOS ONE policies on sharing data and materials by including the following statement: "This does not alter our adherence to  PLOS ONE policies on sharing data and materials.” (as detailed online in our guide for authors http://journals.plos.org/plosone/s/competing-interests) . If this adherence statement is not accurate and  there are restrictions on sharing of data and/or materials, please state these. Please note that we cannot proceed with consideration of your article until this information has been declared.

Please include both an updated Funding Statement and Competing Interests Statement in your cover letter. We will change the online submission form on your behalf.

Please know it is PLOS ONE policy for corresponding authors to declare, on behalf of all authors, all potential competing interests for the purposes of transparency. PLOS defines a competing interest as anything that interferes with, or could reasonably be perceived as interfering with, the full and objective presentation, peer review, editorial decision-making, or publication of research or non-research articles submitted to one of the journals. Competing interests can be financial or non-financial, professional, or personal. Competing interests can arise in relationship to an organization or another person. Please follow this link to our website for more details on competing interests: http://journals.plos.org/plosone/s/competing-interests

4. We note that you have indicated that data from this study are available upon request. PLOS only allows data to be available upon request if there are legal or ethical restrictions on sharing data publicly. For more information on unacceptable data access restrictions, please see http://journals.plos.org/plosone/s/data-availability#loc-unacceptable-data-access-restrictions.

In your revised cover letter, please address the following prompts:

a) If there are ethical or legal restrictions on sharing a de-identified data set, please explain them in detail (e.g., data contain potentially sensitive information, data are owned by a third-party organization, etc.) and who has imposed them (e.g., an ethics committee). Please also provide contact information for a data access committee, ethics committee, or other institutional body to which data requests may be sent.

b) If there are no restrictions, please upload the minimal anonymized data set necessary to replicate your study findings as either Supporting Information files or to a stable, public repository and provide us with the relevant URLs, DOIs, or accession numbers. For a list of acceptable repositories, please see http://journals.plos.org/plosone/s/data-availability#loc-recommended-repositories.

We will update your Data Availability statement on your behalf to reflect the information you provide.

5. PLOS requires an ORCID iD for the corresponding author in Editorial Manager on papers submitted after December 6th, 2016. Please ensure that you have an ORCID iD and that it is validated in Editorial Manager. To do this, go to ‘Update my Information’ (in the upper left-hand corner of the main menu), and click on the Fetch/Validate link next to the ORCID field. This will take you to the ORCID site and allow you to create a new iD or authenticate a pre-existing iD in Editorial Manager. Please see the following video for instructions on linking an ORCID iD to your Editorial Manager account: https://www.youtube.com/watch?v=_xcclfuvtxQ

[Note: HTML markup is below. Please do not edit.]

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #1: No

Reviewer #2: Yes

Reviewer #3: Yes

**********

2. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: No

Reviewer #2: Yes

Reviewer #3: Yes

**********

3. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #1: No

Reviewer #2: Yes

Reviewer #3: No

**********

4. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #1: Yes

Reviewer #2: Yes

Reviewer #3: Yes

**********

5. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #1: This manuscript by Pearcey et al. examines the cross-education of muscular control in terms of behavioral and neural parameters. The authors claimed both trained and untrained arms showed immense improvements in behavioral parameters, and a reduction in H-reflex.

The aim of the manuscript is interesting, but I felt the methodology was inappropriate.

Main concerns

Perez et al. (ref. 26) showed "the H-max/M-max ratio were depressed after repetition of the visuo-motor skill task and returned to baseline after 10 min". Therefore, the timing of H-reflex measurement is important for your research. It should be clearly stated in the manuscript. (I think it's better to measure the H-max/M-max ratio before and just after the game in Pre and Post -tests.)

If H-reflex was measured 10 min or more after the game, the reduction in H-reflex should be nonspecific change (i.e. not related to specific task). This is quite surprising. Why a small number of repetition of the game changed the H-reflex despite the fact that we perform skilled movements in our daily life (e.g., typing, writing, cooking, video game, playing music instruments, ...).

H-reflex was measured from the FCR. The function of the FCR is wrist flexion, wrist abduction, and forearm pronation. I guess the wrist joint was almost fixed during the game. That is, a precise activation control for the FCR are not needed during the game. I suppose the muscles whose activation timing and intensities need to be controlled precisely during the game are extensors and flexors of fingers, elbow, and shoulder. I don't think the H-reflex measurement from the FCR is appropriate for the purpose of this study.

H-max/M-max ratio was not statistically different between pre and post for the entire group (n=9). Discussion must be made on this result. The results obtained from a small number of subjects can easily change.

A "general trend for performance to improve" during three pre-tests must not be ignored. The differences in both time to completion and # of errors between Pre3 and Post were small especially for the untrained arm (Fig. 2A and 2B). If Pre4 were performed for the untrained arm, the results of Pre4 may be similar to the results actually obtained in Post judging from "a general trend for performance to improve". That is, it is not clear whether shorter time to completion and smaller number of errors obtained in Post were resulted from the cross-education effect or repetition of the test. It is a fatal problem that the present study did not provide evidence to deny this expectation.

Generally, brain activities during unimanual action using dominant hand and nondominant hand are asymmetric. Thus it is important to analyze the results from right arm training group and those from left arm training group separately.

I don't think Operation® Iron Man 2 is sufficient tool to evaluate skill for the following two reasons.

1. The buzzer sound startle the subjects, which may result in movement delay. After hearing the buzzer sound repeatedly, they will not be startled by the buzzer sound. That is, familiarity to the buzzer sound would contribute to shorten the time to completion.

2. Whether a certain amount of movement error during pulling out a plastic object buzzes or not depends on the hole because all small holes have different size and shape. The extent of movement error during every pulling out motion should be measured. Counting the buzzer sound is not enough to evaluate movement error.

Minor concerns

Abstract

L34

"the subjects" should be used instead of "women".

L35

Please explain the movement while playing the game instead of the name of the game.

L35-37

No H-max/M-max ratio reduction was found for nine subjects. （I guess）no correlation was found between H-max/M-max ratio reduction and behavioral parameters.

L39

No results gave evidence to support the claim "highlight the need to focus on female participants".

Introduction

A "generalized motor program" presented by Lashley (1942) , and the following studies are worth considering in your research.

L80

Perez et al. (ref. 26) showed "the H-max/M-max ratio were depressed after repetition of the visuo-motor skill task and returned to baseline after 10 min".

L106

If you are interested in gender effect, you should include both female and male subjects.

L122

Why did you include one left-handed subject?

L132-133

Judging from this sentence, H-reflex was measured three times (Pre1, Pre2, and Pre3). Why did you show the results of movement and neural parameters differently?（Fig. 2 and Fig. 3）

L137

No results of ECR, BB, and TB were shown.

L152

Did you use a fixed order for training session at home and in the lab? (an example of a fixed order: {lab, home, home},{lab, home, home}, ...; an example of a non-fixed order: {lab, home, home}, {home, lab, home}, ...)

L152-153

Total number of pulling out movement (55 = 11*5) should be shown clearly. It is difficult to understand the meaning of # of Errors in Fig. 2B.

L168

The effect of test on untrained arm cannot be ignored.

L254

I don't agree this interpretation of the result. A statistical test using only Pre1, Pre2, and Pre3 would show whether behavioral parameters were constant or not among three pre-tests.

L298-299

This isn't in consistent with Fig. 3F.

L330-332

No results support this statement because only female subjects participated in this study.

It is useless to compare % of skill improvements and % of strength improvements in previous studies.

The discussion is too short for the main topic of the study.

The kinematics and muscle activity during gameplay, and enough sample size are essential to the purpose of this study.

Reviewer #2: This study has investigated the inter-manual transfer of motor skills using a real gaming task. The authors demonstrated that the skill learned by one hand was substantially transferred to the skill of another hand. Furthermore, the skill acquisition was likely to be accompanied by the reduction of H-reflex induced in the FCR muscle. This is a significant study showing the presence of cross-education of motor skills in a realistic situation, but I would like to raise several points that the authors need to consider for the improvement.

Major points

1) The information about the task was insufficient. The authors need to describe the details of the game and task more clearly. Which muscles and fingers did the subjects mainly use? What were the size of the objects, the holes, and the tweezers?

2) The reduction of H-reflex induced by the practice is an interesting observation. As the authors already mentioned in the manuscript, there were several limitations, like the small number of subjects in which the reflex could be measured. However, I believe that further analysis would help examine if the reduction of H-reflex was actually related to the improvement of the performance. It would be interesting to see the correlation between the amount of improvement of performance and the amount of H-reflex reduction. I am very interested in the amount of reduction of H-reflex was correlated between both hands.

Minor points

Figure 2: There must be a trade-off (speed-accuracy trade-off) between the “time to completion” and the errors. The scatter plot (e.g., x-axis is time to completion and y-axis is errors) could help the readers recognize how the practice changed the performance.

Figure 4: The current figure is not informative. Adding the recruitment curves would be helpful.

Reviewer #3: The paper by Pearcey and colleagues explores an interesting concept and is uniquely framed with reference to findings reported over 125 years ago. The phenomenon of cross-education has potential relevance to neurorehabilitation and musculoskeletal rehabilitation and the inclusion of a female-only sample addresses gaps in the current literature. The study demonstrates that unilateral training of a task with the dominant limb leads to bilateral improvement in task performance as measured by time to completion and the number of errors committed. Reductions in spinal excitability are also observed. The paper is generally well-written, but I would ask the authors to consider the following points:

1. One of the study hypotheses focuses on spinal reflex excitability, but as the Authors highlight in the introduction, transcallosal processes are likely at play. Indeed, the recent Delphi Consensus statement on cross education (Manca et al. 2021) supports the view that interhemispheric inhibition is the mechanism most likely to be mediating the phenomenon. The Authors double-down on the interhemispheric focus by introducing concepts related to stroke recovery in the present paper, where interhemispheric disinhibition is a likely contributor to motor impairment. Given the that the likely mechanisms at play are cortical in origin, it isn’t clear as to why the Authors chose to focus on measures of spinal excitability rather than cortical excitability to probe the neurophysiological underpinnings related to cross-education in muscular control. The experimental approach and hypothesis don’t quite align with putative mechanisms for the cross-education effect described by the Authors.

In lines 76-80, the use of spinal excitability assessment in the present study is justified by questioning whether there is equivalence between modulation of spinal circuitry after training for muscle control as there appears to be in muscle power. But if the primary mechanism for the phenomenon in the context of muscle power is cortically mediated, why not focus on probing whether muscular control is mediated by the same mechanism? Perhaps more justification is required in the introduction to align with the spinal reflex excitability hypothesis as postulated by the Authors.

2. By having testing and training take place using the same device, there is no evidence as to whether there is a transfer effect to other tasks. While it could be argued that ‘transfer’ is what is being examined in terms of skill development from one limb to it’s untrained contralateral homologue, the functional relevance in a clinical context would require that therapeutic benefits be transferred to other tasks. While I agree that there is potential clinical utility to this approach, I wonder whether the Authors have considered this form of transfer? Perhaps the statements related to clinical relevance should be tempered or have this point added as stipulation or limitation?

3. Were participants screened for occupational, musical or sport training that could contribute to facilitation of the learning/practice effects reported in the study?

4. Line 190 – as stated, it isn’t clear how a ‘low-level wrist flexion contraction (10% of maximum)’ was measured/monitored? Was there a load cell built into the apparatus to measure maximum and steady-state wrist flexion torque? Or was this based on EMG amplitude? Please clarify.

5. The resolution of Figure 1-3 is low. It is difficult to see what is being depicted here. Some of the axis labels and in-figure labels are illegible. Please fix this.

6. Errors and time to completion were measured during training (as stated on line 161). Were the participants aware that training performance was being monitored and were they given feedback as to their performance outcome? If so, how frequent was the feedback? To what extent could the effects observed in the present study be attributable to augmented feedback due to knowledge of results during training rather than from the cross-education effect? This consideration may warrant comment in the Discussion.

7. Figure 2A - Visual inspection of the data presented in this panel seems to indicate that time to completion for the untrained limb at Pre1, Pre2, and Pre3 is less variable and faster (generally speaking) than for the trained limb. For example, the slowest time to completion for the untrained limb at Pre1 looks to be approximately 100 seconds faster than the slowest time to completion for the trained limb. Pre2 and Pre3 seem to illustrate a similar pattern. The authors state that there is no statistically significant main effect of limb or limb x time interaction (limb x time) nor were there “significant differences between pre-test values, or between the trained and untrained limb” (Line 259). However, intuitively, one would expect that at baseline, time to completion should take longer with the untrained (non-dominant) limb. I suppose a rebuttal to this could have something to do with a speed/accuracy trade-off where participants could favour speed over accuracy. This position is somewhat supported by the Error data in panel B for Pre1; however, Pre2 and Pre3 don’t really follow. Can the authors provide an explanation for this seemingly counter-intuitive distribution of trained vs untrained outcomes at Baseline timepoints in Panels A and B?

8. The sections in the Discussion interpreting the ‘Improvements with the trained limb’ and ‘Improvements with the untrained limb’ highlight differences in the rate and extent of improvement in both limbs in the present study in comparison to what has been reported in muscle power studies (Line 337, 340, 352-356). I think the Authors could elaborate on potential mechanisms for this as the muscle control element is the apparent novel contribution here. I appreciate that the section on spinal excitability (Line 366) is meant to describe a potential mechanism, but this doesn’t really differentiate between mechanisms for muscle power vs mechanisms for muscle control.

Line 478 – Information for Reference 23 is incomplete.

Line 502 – Reference 32 – is Volume/issue information available?

**********

6. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: No

Reviewer #2: No

Reviewer #3: No

[NOTE: If reviewer comments were submitted as an attachment file, they will be attached to this email and accessible via the submission site. Please log into your account, locate the manuscript record, and check for the action link "View Attachments". If this link does not appear, there are no attachment files.]

While revising your submission, please upload your figure files to the Preflight Analysis and Conversion Engine (PACE) digital diagnostic tool, https://pacev2.apexcovantage.com/. PACE helps ensure that figures meet PLOS requirements. To use PACE, you must first register as a user. Registration is free. Then, login and navigate to the UPLOAD tab, where you will find detailed instructions on how to use the tool. If you encounter any issues or have any questions when using PACE, please email PLOS at figures@plos.org. Please note that Supporting Information files do not need this step.

12 Jul 2021

1. Please ensure that your manuscript meets PLOS ONE's style requirements, including those for file naming. The PLOS ONE style templates can be found at

https://journals.plos.org/plosone/s/file?id=wjVg/PLOSOne_formatting_sample_main_body.pdf and

https://journals.plos.org/plosone/s/file?id=ba62/PLOSOne_formatting_sample_title_authors_affiliations.pdf

Authors’ response: the document has been updated per the guidelines in the link above

2. In your Methods section, please provide a justification for the sample size used in your study, including any relevant power calculations (if applicable).

Authors’ response: We have added a justification of our sample size, which was based on previous investigations and reports in our lab.

3. Thank you for stating the following in the Competing Interests section:

"The authors have declared that no competing interests exist."

We note that one or more of the authors are employed by a commercial company: Zanshin Consulting, Inc.

Authors’ response: Zanshin Consulting Inc is the company of E. Paul Zehr who is President and sole proprietor. This company purchased the games used in the research. In order to simplify we have deleted an affiliation and mention this instead in the acknowledgements.

3.1. Please provide an amended Funding Statement declaring this commercial affiliation, as well as a statement regarding the Role of Funders in your study. If the funding organization did not play a role in the study design, data collection and analysis, decision to publish, or preparation of the manuscript and only provided financial support in the form of authors' salaries and/or research materials, please review your statements relating to the author contributions, and ensure you have specifically and accurately indicated the role(s) that these authors had in your study. You can update author roles in the Author Contributions section of the online submission form.

Please also include the following statement within your amended Funding Statement.

“The funder provided support in the form of salaries for authors [insert relevant initials], but did not have any additional role in the study design, data collection and analysis, decision to publish, or preparation of the manuscript. The specific roles of these authors are articulated in the ‘author contributions’ section.”

If your commercial affiliation did play a role in your study, please state and explain this role within your updated Funding Statement.

Authors’ response: “The funder provided support in the form of salaries for authors [insert relevant initials], but did not have any additional role in the study design, data collection and analysis, decision to publish, or preparation of the manuscript. The specific roles of these authors are articulated in the ‘author contributions’ section.”

3.2. Please also provide an updated Competing Interests Statement declaring this commercial affiliation along with any other relevant declarations relating to employment, consultancy, patents, products in development, or marketed products, etc.

Within your Competing Interests Statement, please confirm that this commercial affiliation does not alter your adherence to all PLOS ONE policies on sharing data and materials by including the following statement: "This does not alter our adherence to PLOS ONE policies on sharing data and materials.” (as detailed online in our guide for authors http://journals.plos.org/plosone/s/competing-interests) . If this adherence statement is not accurate and there are restrictions on sharing of data and/or materials, please state these. Please note that we cannot proceed with consideration of your article until this information has been declared.

Authors’ response: This is no longer applicable. Please see above.

Please include both an updated Funding Statement and Competing Interests Statement in your cover letter. We will change the online submission form on your behalf.

Please know it is PLOS ONE policy for corresponding authors to declare, on behalf of all authors, all potential competing interests for the purposes of transparency. PLOS defines a competing interest as anything that interferes with, or could reasonably be perceived as interfering with, the full and objective presentation, peer review, editorial decision-making, or publication of research or non-research articles submitted to one of the journals. Competing interests can be financial or non-financial, professional, or personal. Competing interests can arise in relationship to an organization or another person. Please follow this link to our website for more details on competing interests: http://journals.plos.org/plosone/s/competing-interests

Authors’ response: This is no longer applicable. Please see above.

4. We note that you have indicated that data from this study are available upon request. PLOS only allows data to be available upon request if there are legal or ethical restrictions on sharing data publicly. For more information on unacceptable data access restrictions, please see http://journals.plos.org/plosone/s/data-availability#loc-unacceptable-data-access-restrictions.

In your revised cover letter, please address the following prompts:

a) If there are ethical or legal restrictions on sharing a de-identified data set, please explain them in detail (e.g., data contain potentially sensitive information, data are owned by a third-party organization, etc.) and who has imposed them (e.g., an ethics committee). Please also provide contact information for a data access committee, ethics committee, or other institutional body to which data requests may be sent.

b) If there are no restrictions, please upload the minimal anonymized data set necessary to replicate your study findings as either Supporting Information files or to a stable, public repository and provide us with the relevant URLs, DOIs, or accession numbers. For a list of acceptable repositories, please see http://journals.plos.org/plosone/s/data-availability#loc-recommended-repositories.

Authors’ response: We have added our data to OSF (DOI 10.17605/OSF.IO/ZFYEX – https://osf.io/zfyex/)

We will update your Data Availability statement on your behalf to reflect the information you provide.

5. PLOS requires an ORCID iD for the corresponding author in Editorial Manager on papers submitted after December 6th, 2016. Please ensure that you have an ORCID iD and that it is validated in Editorial Manager. To do this, go to ‘Update my Information’ (in the upper left-hand corner of the main menu), and click on the Fetch/Validate link next to the ORCID field. This will take you to the ORCID site and allow you to create a new iD or authenticate a pre-existing iD in Editorial Manager. Please see the following video for instructions on linking an ORCID iD to your Editorial Manager account: https://www.youtube.com/watch?v=_xcclfuvtxQ

Authors’ response: https://orcid.org/0000-0002-1749-2924

Comments to the Author

Reviewer #1:

This manuscript by Pearcey et al. examines the cross-education of muscular control in terms of behavioral and neural parameters. The authors claimed both trained and untrained arms showed immense improvements in behavioral parameters, and a reduction in H-reflex.

The aim of the manuscript is interesting, but I felt the methodology was inappropriate.

Main concerns

Perez et al. (ref. 26) showed "the H-max/M-max ratio were depressed after repetition of the visuo-motor skill task and returned to baseline after 10 min". Therefore, the timing of H-reflex measurement is important for your research. It should be clearly stated in the manuscript. (I think it's better to measure the H-max/M-max ratio before and just after the game in Pre and Post -tests.)

If H-reflex was measured 10 min or more after the game, the reduction in H-reflex should be nonspecific change (i.e. not related to specific task). This is quite surprising. Why a small number of repetition of the game changed the H-reflex despite the fact that we perform skilled movements in our daily life (e.g., typing, writing, cooking, video game, playing music instruments, ...).

Authors’ response: We appreciate the concern of the reviewer, but the work of Perez et al. was conducted to show short-term plasticity in spinal reflexes in response to a single session of visuomotor training. Indeed, in the case of Perez et al. the plasticity did not remain beyond 10 minutes, but one would not expect a single bout of training to cause long-lasting plasticity. On the contrary, in the current experiment, participants performed training multiple times per session, multiple times per week for multiple weeks. This repetitive training is what results in neuroplasticity that can persist, as is the case with operant conditioning, locomotor training, strength training, and other forms of training. In this case fine motor control was trained, and reduced spinal reflex excitability has been shown in folks who are trained for fine motor control (see Nielsen J, Crone C, Hultborn H. H-reflexes are smaller in dancers from The Royal Danish Ballet than in well-trained athletes. Eur J Appl Physiol Occup Physiol. 1993;66(2):116-21. doi: 10.1007/BF01427051. PMID: 8472692.).

H-reflex was measured from the FCR. The function of the FCR is wrist flexion, wrist abduction, and forearm pronation. I guess the wrist joint was almost fixed during the game. That is, a precise activation control for the FCR are not needed during the game. I suppose the muscles whose activation timing and intensities need to be controlled precisely during the game are extensors and flexors of fingers, elbow, and shoulder. I don't think the H-reflex measurement from the FCR is appropriate for the purpose of this study.

Authors’ response: We thank the reviewer for their concerns, but methodological constraints limit our ability to elicit reliable H-reflexes from the distal hand muscles or muscles acting across the shoulder/elbow, and since the FCR is still involved in fine control/movement of the wrist during the game, we chose to include it as a probe for overall spinal reflex excitability. It remains for future researchers to further explore other muscles and other measures.

H-max/M-max ratio was not statistically different between pre and post for the entire group (n=9). Discussion must be made on this result. The results obtained from a small number of subjects can easily change.

Authors’ response:

We feel the issues with H-reflex measures, including our explicit statement that effects are modest, is clearly articulated in the results and discussion. We would like to reply more effectively to the reviewer based on the last sentence but don’t understand. “Can easily change” what?

A "general trend for performance to improve" during three pre-tests must not be ignored. The differences in both time to completion and # of errors between Pre3 and Post were small especially for the untrained arm (Fig. 2A and 2B). If Pre4 were performed for the untrained arm, the results of Pre4 may be similar to the results actually obtained in Post judging from "a general trend for performance to improve". That is, it is not clear whether shorter time to completion and smaller number of errors obtained in Post were resulted from the cross-education effect or repetition of the test. It is a fatal problem that the present study did not provide evidence to deny this expectation.

Authors’ response:

We are not completely certain we follow the reviewers main point? This is a training intervention where repeated exposure may lead to plastic adaptation and enhancement of skill. As such, it does remain possible that pre assessments were producing some skill learning. If so, then our evidence of changes due to 5 weeks of training is actually underrepresenting the true change. That is, any changes to pre make changes in post harder to find. So, we suggest it’s actually not a problem and instead makes our outcomes more conservative.

Generally, brain activities during unimanual action using dominant hand and nondominant hand are asymmetric. Thus it is important to analyze the results from right arm training group and those from left arm training group separately.

Authors’ response: We do not think it is necessary to analyze subjects based on handedness since 1) there is only one left-hand dominant subject, 2) our group showed improvement overall, and 3) all subjects followed the same trend of improvement.

I don't think Operation® Iron Man 2 is sufficient tool to evaluate skill for the following two reasons.

1. The buzzer sound startle the subjects, which may result in movement delay. After hearing the buzzer sound repeatedly, they will not be startled by the buzzer sound. That is, familiarity to the buzzer sound would contribute to shorten the time to completion.

2. Whether a certain amount of movement error during pulling out a plastic object buzzes or not depends on the hole because all small holes have different size and shape. The extent of movement error during every pulling out motion should be measured. Counting the buzzer sound is not enough to evaluate movement error.

Authors’ response: Overall, we disagree and believe that the game is sufficient to provide a proxy for motor skill learning. People learn to complete the game faster and with less error, which means it by definition is a way to evaluate skill. Also, it is a faithful reproduction of the original 1894 study where a buzzer sounded when errors were made. Nevertheless, we have added a statement to acknowledge that a reduction in startle may have contributed to decreased time to completion.

Minor concerns

Abstract

L34 "the subjects" should be used instead of "women".

Authors’ response: We disagree. The subjects are women, and therefore “women” can be used here. This is especially relevant since we are highlighting the need to study women.

L35 Please explain the movement while playing the game instead of the name of the game.

Authors’ response: Our preference is to keep abstract details brief. We have added details about the game in the main text of the methods.

L35-37 No H-max/M-max ratio reduction was found for nine subjects. （I guess）no correlation was found between H-max/M-max ratio reduction and behavioral parameters.

Authors’ response: Hmax/Mmax ratios are a gross test of overall H-reflex excitability that is not typically sensitive to slight changes due to training. Hence, the recruitment curves are more sensitive to small changes in smaller populations of motor units within the reflex pathway.

L39 No results gave evidence to support the claim "highlight the need to focus on female participants".

Authors’ response: We disagree. The results show that cross-education occurs in a women only sample. Women-only samples are rarely studied in the cross education field, and therefore the statement stands. We are actually rather stunned to read that the reviewer does not realize how understudied women have been in this area.

Introduction

A "generalized motor program" presented by Lashley (1942) , and the following studies are worth considering in your research.

Authors’ response: Thanks for the suggestion, but we believe that there are many caveats with the notion of a generalized motor program that make it irrelevant in this case. In particular, there are issues with storage and how the CNS deals with feedback during movement that are not considered within the generalized motor program framework (see central contributions to motor control in Motor Control and Learning, eds Schmidt, Lee, Winstein, Wulf and Zelaznik).

L80

Perez et al. (ref. 26) showed "the H-max/M-max ratio were depressed after repetition of the visuo-motor skill task and returned to baseline after 10 min".

Authors’ response: We agree that this was found which is why we use the words “during acquisition of a novel fine motor skill” indicating that there is a brief period during/immediately after training where H-reflex is reduced. We have added some context.

L106

If you are interested in gender effect, you should include both female and male subjects.

Authors’ response: We were not interested in gender effects but rather the cross-education of skill in women, which we have shown. Again, the reviewer seems to not understand the base issue we are trying to address. To help with this we have added clearer language in the abstract especially.

L122

Why did you include one left-handed subject?

Authors’ response: we did not have exclusion criteria based on handedness.

L132-133

Judging from this sentence, H-reflex was measured three times (Pre1, Pre2, and Pre3). Why did you show the results of movement and neural parameters differently?（Fig. 2 and Fig. 3）

Authors’ response: We have clarified that the pre-test recruitment curve variables were normalized and combined across pre-tests (i.e. 40x3 = 120 stimuli) and compared vs post.

L137

No results of ECR, BB, and TB were shown.

Authors’ response: It is standard practice in our lab to monitor and record activity of heteronymous muscles while recording H-reflexes to ensure remote activity does not influence our measures (please see Zehr 2002 – Considerations for use of the Hoffmann reflex in exercise studies)

L152

Did you use a fixed order for training session at home and in the lab? (an example of a fixed order: {lab, home, home},{lab, home, home}, ...; an example of a non-fixed order: {lab, home, home}, {home, lab, home}, ...)

Authors’ response: A fixed order was used, but this order differed between subjects in order to accommodate participants in the lab.

L152-153

Total number of pulling out movement (55 = 11*5) should be shown clearly. It is difficult to understand the meaning of # of Errors in Fig. 2B.

Authors’ response: We disagree. We have quantified errors based on buzzer sounds.

L168

The effect of test on untrained arm cannot be ignored.

Authors’ response: The effects of three tests performed 5 weeks prior to the post test is unlikely to be a major contributor, rather there is likely training induced cross-education of skill.

L254

I don't agree this interpretation of the result. A statistical test using only Pre1, Pre2, and Pre3 would show whether behavioral parameters were constant or not among three pre-tests.

Authors’ response: We will have to agree to disagree with the reviewer here. This is based on our comments above and our prior (and ongoing) extensive use of multiple baseline control procedures in intervention studies.

L298-299

This isn't in consistent with Fig. 3F.

Authors’ response: We apologize, but we don’t understand what is intended by this comment?

L330-332

No results support this statement because only female subjects participated in this study.

Authors’ response: The results support the inclusion of women-only samples since we have shown that cross-education of skill occurs in a women-only sample. We encourage others to study women-only samples as well.

It is useless to compare % of skill improvements and % of strength improvements in previous studies.

Authors’ response: We believe it is a reasonable comparison in the context it is used. In fact, relative performance assessment as a ratio or percentage is one of the most commonly used procedures to compare across studies and participants. How else would one compare strength to skill adaptation if not a relative change?

The discussion is too short for the main topic of the study.

Authors’ response: We do not agree but if the reviewer has specific recommendations on what to discuss in more detail, we are keen to listen. However, we don’t understand the suggestion to make the discussion longer because it is too short?

The kinematics and muscle activity during gameplay, and enough sample size are essential to the purpose of this study.

Authors’ response: As stated in our section on “important considerations,” we did not measure kinematics or muscle activity during gameplay, so we cannot provide this. In addition, we discuss the limitations associated with our small sample in this section, yet we have now given a rationale of our sample size, which has been added in text.

“The sample size was determined based on prior related work in the laboratory using similar designs and outcome measures, which were sufficient to achieve significant cross-education effects of strength training with moderate effect sizes (Barss et al 2018 doi: 10.1152/japplphysiol.00390.2017, Sun et al 2018 doi: 10.1007/s00221-018-5275-6).”

Reviewer #2:

This study has investigated the inter-manual transfer of motor skills using a real gaming task. The authors demonstrated that the skill learned by one hand was substantially transferred to the skill of another hand. Furthermore, the skill acquisition was likely to be accompanied by the reduction of H-reflex induced in the FCR muscle. This is a significant study showing the presence of cross-education of motor skills in a realistic situation, but I would like to raise several points that the authors need to consider for the improvement.

Authors’ response: Thanks for the constructive feedback, which we believe helped improve the quality of the manuscript.

Major points

1) The information about the task was insufficient. The authors need to describe the details of the game and task more clearly. Which muscles and fingers did the subjects mainly use? What were the size of the objects, the holes, and the tweezers?

Authors’ response: We have added some more details about the game. In particular, we briefly described that participants were not instructed how to hold the tweezers, but the most common way was to hold the tweezers between the tips of the thumb and index finger with support from the middle finger. Objects ranged in size with graspable regions ranging from 1-4mm. The tweezers were 5.4cmm in length with a 7mm aperture. The holes ranged in size, but the narrowest region was 16mm.

2) The reduction of H-reflex induced by the practice is an interesting observation. As the authors already mentioned in the manuscript, there were several limitations, like the small number of subjects in which the reflex could be measured. However, I believe that further analysis would help examine if the reduction of H-reflex was actually related to the improvement of the performance. It would be interesting to see the correlation between the amount of improvement of performance and the amount of H-reflex reduction. I am very interested in the amount of reduction of H-reflex was correlated between both hands.

Authors’ response: Although these aspects are interesting to explore, the small sample size severely limits our ability to run correlational/regression analyses. In addition, we have no metric of overall improvement of performance, which complicates an answer these questions (i.e. improvement in speed vs errors). We think that the change in the reflex excitability is a minor side story and have tried to downplay this finding, particularly in response to another reviewers concerns.

Minor points

Figure 2: There must be a trade-off (speed-accuracy trade-off) between the “time to completion” and the errors. The scatter plot (e.g., x-axis is time to completion and y-axis is errors) could help the readers recognize how the practice changed the performance.

Authors’ response: We are unsure what the reviewer thinks the scatter below adds to the story. Is this what is suggested?

Figure 4: The current figure is not informative. Adding the recruitment curves would be helpful.

Authors’ response: We do not agree with the reviewer and prefer Figure 4 as it is. It currently shows the relative amplitudes and shapes of waveforms (i.e. M-wave and H-reflexes) along with the latencies of the responses. True raw data is beneficial to ensure replicability in the future.

Reviewer #3:

The paper by Pearcey and colleagues explores an interesting concept and is uniquely framed with reference to findings reported over 125 years ago. The phenomenon of cross-education has potential relevance to neurorehabilitation and musculoskeletal rehabilitation and the inclusion of a female-only sample addresses gaps in the current literature. The study demonstrates that unilateral training of a task with the dominant limb leads to bilateral improvement in task performance as measured by time to completion and the number of errors committed. Reductions in spinal excitability are also observed. The paper is generally well-written, but I would ask the authors to consider the following points:

Authors’ response: Thanks for the helpful suggestions, which we think help improved the quality of the manuscript.

1. One of the study hypotheses focuses on spinal reflex excitability, but as the Authors highlight in the introduction, transcallosal processes are likely at play. Indeed, the recent Delphi Consensus statement on cross education (Manca et al. 2021) supports the view that interhemispheric inhibition is the mechanism most likely to be mediating the phenomenon. The Authors double-down on the interhemispheric focus by introducing concepts related to stroke recovery in the present paper, where interhemispheric disinhibition is a likely contributor to motor impairment. Given the that the likely mechanisms at play are cortical in origin, it isn’t clear as to why the Authors chose to focus on measures of spinal excitability rather than cortical excitability to probe the neurophysiological underpinnings related to cross-education in muscular control. The experimental approach and hypothesis don’t quite align with putative mechanisms for the cross-education effect described by the Authors.

Authors’ response: We certainly agree that cortical mechanisms are major factors mediating cross-education. However, cortical output must arrive at and pass through the spinal cord to be expressed as motor activity. Thus, spinal adaptations are also relevant to study. We also suggest that our study was not intended as a major mechanistic investigation.

In lines 76-80, the use of spinal excitability assessment in the present study is justified by questioning whether there is equivalence between modulation of spinal circuitry after training for muscle control as there appears to be in muscle power. But if the primary mechanism for the phenomenon in the context of muscle power is cortically mediated, why not focus on probing whether muscular control is mediated by the same mechanism? Perhaps more justification is required in the introduction to align with the spinal reflex excitability hypothesis as postulated by the Authors.

Authors’ response: The measurement of spinal excitability was meant to be an additional probe to examine whether this type of training influences neural circuits within the spinal cord. The main message remains that cross-education of skill occurs in women. A change in spinal excitability should not take away from this message, but rather provide some additional insights about what might contribute to this finding. We have made an attempt to de-emphasize the importance of the reflex findings.

2. By having testing and training take place using the same device, there is no evidence as to whether there is a transfer effect to other tasks. While it could be argued that ‘transfer’ is what is being examined in terms of skill development from one limb to it’s untrained contralateral homologue, the functional relevance in a clinical context would require that therapeutic benefits be transferred to other tasks. While I agree that there is potential clinical utility to this approach, I wonder whether the Authors have considered this form of transfer? Perhaps the statements related to clinical relevance should be tempered or have this point added as stipulation or limitation?

Authors’ response: This is an excellent point that we did consider addressing with a separate game. Unfortunately, only a small subset (n = 2) of subjects completed the “transfer” game at both time points (pre/post), which was too small of a sample to draw conclusions about the transfer. We have made an attempt to temper our conclusions about the clinical relevance of the cross-education of skill based on this game.

3. Were participants screened for occupational, musical or sport training that could contribute to facilitation of the learning/practice effects reported in the study?

Authors’ response: we used a convenience sample from the undergraduate population of the University of Victoria and participants were not screened based on occupational, musical or sport training.

4. Line 190 – as stated, it isn’t clear how a ‘low-level wrist flexion contraction (10% of maximum)’ was measured/monitored? Was there a load cell built into the apparatus to measure maximum and steady-state wrist flexion torque? Or was this based on EMG amplitude? Please clarify.

Authors’ response: Thanks for pointing out this omission of details. We have added that there was EMG feedback given to participants in the following way: “Feedback consisted of a 100 ms moving average of the rectified EMG.”

5. The resolution of Figure 1-3 is low. It is difficult to see what is being depicted here. Some of the axis labels and in-figure labels are illegible. Please fix this.

Authors’ response: We are not sure how this occurred, because on our end, the figures are high-resolution and very clear. We hope that the figures are clear on this second submission.

6. Errors and time to completion were measured during training (as stated on line 161). Were the participants aware that training performance was being monitored and were they given feedback as to their performance outcome? If so, how frequent was the feedback? To what extent could the effects observed in the present study be attributable to augmented feedback due to knowledge of results during training rather than from the cross-education effect? This consideration may warrant comment in the Discussion.

Authors’ response: As stated “Errors and completion time were only recorded in lab training sessions.” No feedback was provided to participants about the number of errors that they committed. In this way, participants could only understand the number of errors if they counted them in their own heads, which was not apparent based on the observations of our team of investigators. We have clarified this in the methods, but we do not believe that we should add anything on this point into the discussion.

7. Figure 2A - Visual inspection of the data presented in this panel seems to indicate that time to completion for the untrained limb at Pre1, Pre2, and Pre3 is less variable and faster (generally speaking) than for the trained limb. For example, the slowest time to completion for the untrained limb at Pre1 looks to be approximately 100 seconds faster than the slowest time to completion for the trained limb. Pre2 and Pre3 seem to illustrate a similar pattern. The authors state that there is no statistically significant main effect of limb or limb x time interaction (limb x time) nor were there “significant differences between pre-test values, or between the trained and untrained limb” (Line 259). However, intuitively, one would expect that at baseline, time to completion should take longer with the untrained (non-dominant) limb. I suppose a rebuttal to this could have something to do with a speed/accuracy trade-off where participants could favour speed over accuracy. This position is somewhat supported by the Error data in panel B for Pre1; however, Pre2 and Pre3 don’t really follow. Can the authors provide an explanation for this seemingly counter-intuitive distribution of trained vs untrained outcomes at Baseline timepoints in Panels A and B?

Authors’ response: This is a great observation, and at first glance is quite puzzling. Indeed, it would be strange if the non-dominant limb was better at gameplay than the dominant limb. To understand this seemingly strange observation, one must consider the order of task completion. In all cases, the participants played the game with the dominant (trained) limb before playing with the non-dominant (untrained) limb. We believe that this order of gameplay most likely explains the visual trend in the data for the non-dominant limb to be faster (albeit with more errors in Pre1) than the dominant limb.

8. The sections in the Discussion interpreting the ‘Improvements with the trained limb’ and ‘Improvements with the untrained limb’ highlight differences in the rate and extent of improvement in both limbs in the present study in comparison to what has been reported in muscle power studies (Line 337, 340, 352-356). I think the Authors could elaborate on potential mechanisms for this as the muscle control element is the apparent novel contribution here. I appreciate that the section on spinal excitability (Line 366) is meant to describe a potential mechanism, but this doesn’t really differentiate between mechanisms for muscle power vs mechanisms for muscle control.

Authors’ response: It is our belief that strength is a skill, too. However training for strength and training for muscular control differ in the skill that is being performed. With strength, plasticity occurs to increase the force generating capacity of a particular group of muscles in a particular direction. This can involve facilitation of agonist and synergist muscles, while inhibiting antagonists. On the other hand, training for muscular control essentially involves training inhibitory circuits to ensure fine control and smooth movements, while enhancing the specific motor plan for the skilled movement that is trained. We have expanded our discussion to point out these subtleties.

Line 478 – Information for Reference 23 is incomplete.

Authors’ response: Our apologies – this must have been an issue with our reference manager. The reference should be:

Manca A, Hortobágyi T, Carroll TJ, Enoka RM, Farthing JP, Gandevia SC, Kidgell DJ, Taylor JL, Deriu F. Contralateral Effects of Unilateral Strength and Skill Training: Modified Delphi Consensus to Establish Key Aspects of Cross-Education. Sports Med. 2021 Jan;51(1):11-20. doi: 10.1007/s40279-020-01377-7. PMID: 33175329; PMCID: PMC7806569.

Line 502 – Reference 32 – is Volume/issue information available?

Authors’ response: Our apologies, again – this must have been another issue with our reference manager. The reference should be:

Sun Y, Ledwell NMH, Boyd LA, Zehr EP. Unilateral wrist extension training after stroke improves strength and neural plasticity in both arms. Exp Brain Res. 2018 Jul;236(7):2009-2021. doi: 10.1007/s00221-018-5275-6. Epub 2018 May 5. PMID: 29730752.

Submitted filename: Response to reviewers.docx

Click here for additional data file.

8 Sep 2021

PONE-D-20-38248R1

1894 revisited: Cross-education of skilled muscular control in women and the importance of representation

PLOS ONE

Dear Dr. Zehr,

Thank you for submitting your manuscript to PLOS ONE. After careful consideration, we feel that it has merit but does not fully meet PLOS ONE’s publication criteria as it currently stands. Therefore, we invite you to submit a revised version of the manuscript that addresses the points raised during the review process.

Reviewer 1 believes that FCR possibly does not play a role in game task, and requests evidence showing the role of FCR in the task provided to the participants. Furthermore, reviewer 1 has also critiqued the overall experimental design, which the Academic Editor does not support too. Please provide clear justifications and/or describe them as the study limitations in the manuscript as well as in the response letter to the reviewer. Also, please respond to the minor comments of reviewer 3

Please submit your revised manuscript by Oct 21 2021 11:59PM. If you will need more time than this to complete your revisions, please reply to this message or contact the journal office at plosone@plos.org. When you're ready to submit your revision, log on to https://www.editorialmanager.com/pone/ and select the 'Submissions Needing Revision' folder to locate your manuscript file.

Please include the following items when submitting your revised manuscript:

A rebuttal letter that responds to each point raised by the academic editor and reviewer(s). You should upload this letter as a separate file labeled 'Response to Reviewers'.

A marked-up copy of your manuscript that highlights changes made to the original version. You should upload this as a separate file labeled 'Revised Manuscript with Track Changes'.

An unmarked version of your revised paper without tracked changes. You should upload this as a separate file labeled 'Manuscript'.

If you would like to make changes to your financial disclosure, please include your updated statement in your cover letter. Guidelines for resubmitting your figure files are available below the reviewer comments at the end of this letter.

If applicable, we recommend that you deposit your laboratory protocols in protocols.io to enhance the reproducibility of your results. Protocols.io assigns your protocol its own identifier (DOI) so that it can be cited independently in the future. For instructions see:  http://journals.plos.org/plosone/s/submission-guidelines#loc-laboratory-protocols . Additionally, PLOS ONE offers an option for publishing peer-reviewed Lab Protocol articles, which describe protocols hosted on protocols.io. Read more information on sharing protocols at  https://plos.org/protocols?utm_medium=editorial-email&utm_source=authorletters&utm_campaign=protocols .

We look forward to receiving your revised manuscript.

Kind regards,

Lucinda Shen

Staff Editor

On behalf of

Kei Masani

Academic Editor

PLOS ONE

Journal Requirements:

Please review your reference list to ensure that it is complete and correct. If you have cited papers that have been retracted, please include the rationale for doing so in the manuscript text, or remove these references and replace them with relevant current references. Any changes to the reference list should be mentioned in the rebuttal letter that accompanies your revised manuscript. If you need to cite a retracted article, indicate the article’s retracted status in the References list and also include a citation and full reference for the retraction notice.

[Note: HTML markup is below. Please do not edit.]

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. If the authors have adequately addressed your comments raised in a previous round of review and you feel that this manuscript is now acceptable for publication, you may indicate that here to bypass the “Comments to the Author” section, enter your conflict of interest statement in the “Confidential to Editor” section, and submit your "Accept" recommendation.

Reviewer #1: (No Response)

Reviewer #2: All comments have been addressed

Reviewer #3: (No Response)

**********

2. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #1: No

Reviewer #2: Yes

Reviewer #3: Yes

**********

3. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: Yes

Reviewer #2: Yes

Reviewer #3: Yes

**********

4. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #1: Yes

Reviewer #2: Yes

Reviewer #3: Yes

**********

5. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #1: Yes

Reviewer #2: Yes

Reviewer #3: Yes

**********

6. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #1: The authors failed to provide a rationale for associating the results of H-reflex in FCR to behavioral results. The authors did not provide any information about the role FCR plays in the gameplay. My expectation is that in the gameplay the wrist movement is small and rather fixed. That is, the wrist muscles including FCR would not contribute much to skill acquisition/ fine motor control.

Both the time to completion and the number of errors continued to decrease (improve) in the three pre assessments (Fig. 2AB). In other words, repeated assessment improves behavioral performance. The fourth assessment was conducted after the training period, but it is possible that the same results would have been obtained if it had been conducted before the training period. It is a fatal problem that the three pre assessments did not provide reliable baseline.

Reviewer #2: The authors have addressed my concerns.

I also thank the authors for making the scatter plot showing the relationship between the completion time and the number of errors. The authors do not need to include this plot because I understood the information is not so different from Figure 3.

Reviewer #3: The Authors have generally addressed my concerns, but I do make the following comments for consideration:

1. Regarding my previous comment about the counter-intuitive observations of the data presented in Figure 2A, it should be clarified in the Methods that participants always performed the muscular control assessments with the Trained hand first. As an aside, I wonder whether the Authors’ response and acknowledgement of Reviewer 1’s comment about startle in the Limitations section may also contribute to my point regarding Figure 2A. Performing the task with the Trained hand first and experiencing errors (i.e. buzzers sounds) may increase time to completion initially, but may then lead to an attenuation of the effect by the time participants performed the task with the Untrained hand.

2. I note that References 32 and 33 are duplicates. Please correct.

3. The resolution of the images is still poor, but I this could just be on my end.

**********

7. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: No

Reviewer #2: No

Reviewer #3: No

[NOTE: If reviewer comments were submitted as an attachment file, they will be attached to this email and accessible via the submission site. Please log into your account, locate the manuscript record, and check for the action link "View Attachments". If this link does not appear, there are no attachment files.]

While revising your submission, please upload your figure files to the Preflight Analysis and Conversion Engine (PACE) digital diagnostic tool, https://pacev2.apexcovantage.com/. PACE helps ensure that figures meet PLOS requirements. To use PACE, you must first register as a user. Registration is free. Then, login and navigate to the UPLOAD tab, where you will find detailed instructions on how to use the tool. If you encounter any issues or have any questions when using PACE, please email PLOS at figures@plos.org. Please note that Supporting Information files do not need this step.

10 Dec 2021

Decision point #1: Reviewer 1 believes that FCR possibly does not play a role in game task, and requests evidence showing the role of FCR in the task provided to the participants.

Comment from reviewer 1: The authors failed to provide a rationale for associating the results of H-reflex in FCR to behavioral results. The authors did not provide any information about the role FCR plays in the gameplay. My expectation is that in the gameplay the wrist movement is small and rather fixed. That is, the wrist muscles including FCR would not contribute much to skill acquisition/ fine motor control.

Authors’ response: Although the direct role of FCR in gameplay may be limited, its role in stabilization and control of wrist movements is relevant to improving the number of errors. For instance, the FCR works in synergy with the ECR to produce radial deviation, and the FCR acts to stabilize and flex the wrist. To better understand the role of FCR in gameplay, one should consider gain control of the limb during a task being trained. If gain is high, small adjustments in neural drive would result in large adjustments in kinematics, potentially resulting in errors. On the contrary, if training results in reducing the gain of muscles acting at the wrist (e.g. the FCR muscle), adjustments in synaptic input (descending neural drive or sensory input as examples) would simply result in smaller deviations in the kinematics and, thus, reduced errors. Therefore, although not the primary muscle used during gameplay to make fine motor adjustments, the FCR is an appropriate muscle to act as a proxy for gain to arm muscles distal to the elbow. If the reviewer disagrees, we would be happy to revise the manuscript with their suggestion in mind.

In any case we have added some rationale for the choice of this muscle (lines 180-186) and emphasized this as a limitation to our study, which can be found in the “important considerations” section of the Discussion (lines 441-447).

Decision point #2: Reviewer 1 has also critiqued the overall experimental design, which the Academic Editor does not support too. Please provide clear justifications and/or describe them as the study limitations in the manuscript as well as in the response letter to the reviewer.

Comment from reviewer 1: Both the time to completion and the number of errors continued to decrease (improve) in the three pre assessments (Fig. 2AB). In other words, repeated assessment improves behavioral performance. The fourth assessment was conducted after the training period, but it is possible that the same results would have been obtained if it had been conducted before the training period. It is a fatal problem that the three pre assessments did not provide reliable baseline.

Authors’ response: This is an interesting point but we are also slightly confused. The reviewer seems to be arguing that the training exposures didn’t actually train the participants and that without the training a 4th assessment would have yielded a training effect on its own. Is this correct?

Also, please note that when using statistical tests things are either significantly different or they are not. There were not significant changes across baseline tests according to statistical evaluations. Therefore the comment that measures “continued to decrease (improve)” is disingenuous. Even if this were true, and statistical testing supported the reviewer’s suggestion of perceived differences, this actually means we have underestimated the power of the training intervention and are actually being more conservative in our assessments. If the reviewer is asking us to say that our non-significant differences across baseline are “significant”, and we extend this logic, would we then be asked to do the same for any of our experimental measures?

In any case, we think we understand what the reviewer is getting at and have added “Although there were no significant changes across baseline measures, visual inspection of our data may suggest a trend towards improvement in these sessions. Future studies may consider multiple post-test measures as a way to evaluate this.” to the “important considerations” section (lines 422-425).

Decision point #3: Please respond to the minor comments of reviewer 3

1. Regarding my previous comment about the counter-intuitive observations of the data presented in Figure 2A, it should be clarified in the Methods that participants always performed the muscular control assessments with the Trained hand first. As an aside, I wonder whether the Authors’ response and acknowledgement of Reviewer 1’s comment about startle in the Limitations section may also contribute to my point regarding Figure 2A. Performing the task with the Trained hand first and experiencing errors (i.e. buzzers sounds) may increase time to completion initially, but may then lead to an attenuation of the effect by the time participants performed the task with the Untrained hand.

Authors’ response: We agree with both points, and as such have added a point in the methods (line 167) about the test order, and in the further considerations (lines 432-434) about how the habituation of the startle resulting from the buzzer may have contributed to the counter-intuitive results of the non-dominant hand completing the game at a faster rate.

2. I note that References 32 and 33 are duplicates. Please correct.

Authors’ response: We have revised by removing 33.

3. The resolution of the images is still poor, but I this could just be on my end.

Authors’ response: We apologize for this. The uploaded figures are clear on our end. We will ensure they are clear if the manuscript is accepted. We have uploaded .tif files exported at 300DPI from Adobe Illustrator, for the record.

Submitted filename: Responses to editor and reviewers.docx

Click here for additional data file.

16 Feb 2022

1894 revisited: Cross-education of skilled muscular control in women and the importance of representation

PONE-D-20-38248R2

Dear Dr. Zehr,

We’re pleased to inform you that your manuscript has been judged scientifically suitable for publication and will be formally accepted for publication once it meets all outstanding technical requirements.

Within one week, you’ll receive an e-mail detailing the required amendments. When these have been addressed, you’ll receive a formal acceptance letter and your manuscript will be scheduled for publication.

An invoice for payment will follow shortly after the formal acceptance. To ensure an efficient process, please log into Editorial Manager at http://www.editorialmanager.com/pone/, click the 'Update My Information' link at the top of the page, and double check that your user information is up-to-date. If you have any billing related questions, please contact our Author Billing department directly at authorbilling@plos.org.

If your institution or institutions have a press office, please notify them about your upcoming paper to help maximize its impact. If they’ll be preparing press materials, please inform our press team as soon as possible -- no later than 48 hours after receiving the formal acceptance. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information, please contact onepress@plos.org.

Kind regards,

Kei Masani

Academic Editor

PLOS ONE

Additional Editor Comments (optional):

Reviewers' comments:

8 Mar 2022

PONE-D-20-38248R2

1894 revisited: Cross-education of skilled muscular control in women and the importance of representation

Dear Dr. Zehr:

I'm pleased to inform you that your manuscript has been deemed suitable for publication in PLOS ONE. Congratulations! Your manuscript is now with our production department.

If your institution or institutions have a press office, please let them know about your upcoming paper now to help maximize its impact. If they'll be preparing press materials, please inform our press team within the next 48 hours. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information please contact onepress@plos.org.

If we can help with anything else, please email us at plosone@plos.org.

Thank you for submitting your work to PLOS ONE and supporting open access.

Kind regards,

PLOS ONE Editorial Office Staff

on behalf of

Dr. Kei Masani

Academic Editor

PLOS ONE