Sleep restriction impairs visually and memory-guided force control

Sarah A Brinkerhoff; Gina M Mathew; William M Murrah; Anne-Marie Chang; Jaimie A Roper; Kristina A Neely

doi:10.1371/journal.pone.0274121

. 2022 Sep 2;17(9):e0274121. doi: 10.1371/journal.pone.0274121

Sleep restriction impairs visually and memory-guided force control

Sarah A Brinkerhoff ^1,^*, Gina M Mathew ², William M Murrah ³, Anne-Marie Chang ⁴, Jaimie A Roper ¹, Kristina A Neely ¹

Editor: Kenichi Shibuya⁵

PMCID: PMC9439228 PMID: 36054227

Abstract

Sleep loss is a common phenomenon with consequences to physical and mental health. While the effects of sleep restriction on working memory are well documented, it is unknown how sleep restriction affects continuous force control. The purpose of this study was to determine the effects of sleep restriction on visually and memory-guided force production magnitude and variability. We hypothesized that both visually and memory-guided force production would be impaired after sleep restriction. Fourteen men participated in an eleven-day inpatient sleep study and completed a grip force task after two nights of ten hours’ time in bed (baseline); four nights of five hours’ time in bed (sleep restriction); and one night of ten hours’ time in bed (recovery). The force task entailed four 20-second trials of isometric force production with the thumb and index finger targeting 25% of the participant’s maximum voluntary contraction. During visually guided trials, participants had continuous visual feedback of their force production. During memory-guided trials, visual feedback was removed for the last 12 seconds of each trial. During both conditions, participants were told to maintain the target force production. After sleep restriction, participants decreased the magnitude of visually guided, but not memory-guided, force production, suggesting that visual attention tasks are more affected by sleep loss than memory-guided tasks. Participants who reported feeling more alert after sleep restriction and recovery sleep produced higher force during memory-guided, but not visually guided, force production, suggesting that the perception of decreased alertness may lead to more attention to the task during memory-guided visual tasks.

Introduction

It is recommended that adults receive at least seven hours of sleep each night for optimal mental and metabolic health and performance [1]. However, according to the Centers for Disease Control, 35% of adults in the United States achieve fewer than seven hours of continuous sleep per day, and these individuals are more likely to be obese, physically inactive, and have chronic health conditions such as diabetes, depression, stroke, and coronary heart disease [2]. Therefore, it is important to understand the multi-system effects of restricted sleep schedules on health and functioning.

In addition to the above-mentioned health consequences, sleep restriction negatively affects speed [3–8] and accuracy [7] on tests of sustained attention such as the psychomotor vigilance task (PVT). Performance in more complex domains, such as working memory, is also impaired after sleep restriction [7, 9, 10]. Working memory is the temporary storage and manipulation of information, and is an intersection between memory, perception, and the attentional control of behavior [11], and is necessary for motor skill acquisition [12, 13]. Indeed, the motor system and working memory tasks recruit common neural pathways [14–18]. Considering that sleep restriction negatively affects working memory, it follows that sleep restriction may also affect performance of motor tasks that rely on working memory. However, there is a dearth of research examining such tasks.

Although it is unclear whether sleep restriction affects the performance of continuous motor tasks, prior research demonstrates that sleep restriction impairs performance of discrete button-press tasks. For example, individuals exhibit more variability on the PVT (e.g., standard deviation of reaction time) after total sleep deprivation [19, 20]. However, discrete button-press tasks confound sensory, motor, and cognitive processes into a single dichotomous response. Furthermore, many occupational tasks and activities of daily living require relatively short periods of continuous motor output, such as carrying a cup of hot coffee. Sleep-restricted participants exhibit slower reaction time [21, 22] and variability [22] in driving simulation tasks, but there is a lack of literature on the effects of sleep restriction on tasks requiring both continuous motor output and working memory, such as playing an instrument or participating in an athletic activity.

Therefore, the purpose of this study was to evaluate the effects of sleep restriction on force production during visually and memory-guided force control tasks. We hypothesized that sleep restriction would impact memory-guided but not visually guided force production.

Methods

Ethics statement

The Institutional Review Board at Penn State approved all procedures, which were consistent with the Declaration of Helsinki. After receiving a complete description of the study, each participant provided written informed consent prior to enrollment in screening and study procedures. Participants received monetary compensation for their involvement.

Participants

Healthy young adult male participants, ages 20–35, were recruited for this study through websites (such as Penn State research websites and Craigslist) and flyers posted at the Penn State University Park campus and surrounding areas.

A clinician at the Penn State Clinical Research Center reviewed participants’ medical history and conducted physical exams to assess physical health, and a clinical psychologist conducted a structured clinical interview with each participant to determine psychiatric and psychological suitability. Participants were included if they were deemed healthy and were adherent to screening procedures. Women were excluded due to the effects of the menstrual cycle on sleep and circadian rhythms [23] and adults over 35 were excluded due to the effect of aging on sleep patterns [24]. Exclusion criteria were tobacco or drug use (confirmed by urine toxicology), excessive alcohol consumption, prescription medication use, hearing or vision impairment, neurological disorder, night or shift work within the previous three years, travel across > 2 time zones within the previous 3 months, acute or chronic medical conditions, or history of sleep disorder, which was evaluated through the Sleep Disorders Questionnaire [25]. Exclusion criteria also included metal implants that would be unsafe for magnetic resonance imaging and any of the following sub-clinical metabolic disturbances: body mass index of ≤ 18 kg/m², systolic blood pressure of ≥ 130 mmHg or a diastolic blood pressure of ≥ 85 mmHg, hemoglobin A1C glycosylation level of ≥ 5.7% (pre-diabetic or diabetic per ADA criteria), HDL cholesterol level of < 40 mg/dL, LDL cholesterol level of ≥ 145 mg/dL, plasma triglyceride level of ≥ 150 mg/dL, plasma triglyceride to HDL cholesterol ratio of ≥ 3.5, fasting glucose ≥ 100 mg/dL, and/or waist circumference > 102 cm.

Procedures

As previously described [7, 26], the current work was part of an 11-day inpatient study that followed a within-subjects design. Participants were admitted to the Clinical Research Center of the Pennsylvania State University at approximately 11:00 on admission day to a private, windowless room under constant (artificial) light levels (< 100 lux in the angle of gaze during wake; complete dark at 0 lux during scheduled sleep) and temperature conditions (20°-22°C). The private room contained a single bed, a desk for administration of cognitive batteries, and a bathroom with a shower. Participants were not permitted to nap, sit, or recline in bed during scheduled wake times and were monitored by research assistants to confirm adherence. Light-emitting personal devices such as mobile phones and laptop computers were removed 2 hours before scheduled bedtime and were returned at least 2 hours after scheduled wake time to limit exposure to the alerting effects of blue light near the sleep episode [27, 28].

Each participant underwent all three of the following conditions: baseline, sleep restriction, and recovery sleep (Fig 1). The rested baseline condition consisted of three nights of 10 hours in bed. To ensure that participants were adapted to the in-lab baseline condition, participants were instructed to spend 10 hours in bed each night at home for the week prior to admission into the study. In addition, their sleep schedules were monitored using wrist actigraphy and time-stamped calls at bedtimes and wake times during this week. Following the rested baseline condition, participants were restricted to five hours in bed for five nights during the sleep restriction condition. The final two nights of the study consisted of 10 hours in bed during recovery sleep. Hours spent sleeping was determined via polysomnography (PSG) and wrist actigraphy, however, the current analysis included only actigraphy data due to missingness of PSG for specific nights of interest in two participants. For the full duration of the study, food intake was strictly controlled according to a eucaloric diet consisting of weighed foods with predetermined macronutrient and micronutrient contents [26]. During the study, participants were permitted to engage in activities such as reading, completing puzzles, light stretching, and browsing the internet, provided no study procedures were scheduled at that time.

The motor force task took place once per study condition at 10:15 AM on days three, eight, and 10 of the inpatient stay. These corresponded to the second baseline day (day three of the study), the fourth sleep restriction day (day eight of the study), and the first recovery day (day 10 of the study). In the current analyses, baseline, sleep restriction, and recovery sleep refer to these three days, respectively.

Measurement of sleep

Actigraphy data were downloaded with Philips Actiware software (versions 6.0.4. and 6.0.9.). At least two independent scorers (blinded to each other) determined “day” cut-point times, validity of days, and set sleep intervals using a previously validated procedure [29]. The scorers adjudicated each recording by verifying number of valid days, cut point, number of sleep intervals, and differences greater than 15 minutes in duration and wake after sleep onset for each sleep interval. The measure of interest calculated by actigraphy was total sleep time (TST) in hours for the sleep interval.

Sleepiness scores

Self-reported sleepiness was assessed using the Karolinska Sleepiness Scale (KSS; [30] administered on a secure website (Research Electronic Data Capture, REDCap, versions 6.10.11 through 8.1.16; [31])). This is a 9-point scale with higher numbers indicating greater sleepiness: 1 = extremely alert; 3 = alert; 5 = neither alert nor sleepy; 7 = sleepy but no difficulty remaining awake; and 9 = extremely sleepy, fighting sleep. Participants may select even-number ratings, which have no descriptors, but represent sleepiness ratings in between the odd numbers. The KSS was administered 3–6 times on days three, eight, and 10 of the study, but for the current analysis, only the most proximal ratings to the motor force task were used (~1:30 PM).

Motor force tasks

Precision grip strength of the dominant (right) hand was assessed by obtaining each participant’s maximum voluntary contraction (MVC) using a pinch grip dynamometer (Lafayette Hydraulic Gauge, Lafayette, IN). The average of three 5-second trials determined each participant’s MVC in newtons. Participants completed MVCs and the visually and memory-guided force control tasks on baseline, sleep restriction, and recovery sleep days.

Participants were seated in an upright chair (JedMed Straight Back Chair, St. Louis, MO) 127 cm from a 102 cm Samsung television that had a resolution of 1920 x 1080 and a refresh rate of 120 Hz. With their forearm resting at approximately 100 degrees of elbow flexion, participants used their thumb and index finger to form a pinch grip against two ELFF-B4 model load cells constructed from piezoresistive strain gauges (Measurement Specialties, Hampton, VA) (Fig 2A). Force data were collected by Coulbourn Instruments Type B V72-25B amplifiers at an excitation voltage of 5V. The voltage was transmitted via a 16-bit A/D converter and digitized at 62.5 Hz. The A/D board units were transformed to newtons using a calibration factor derived from known weights. The voltage range was -10 to 10V, and the A/D board was able to detect force levels as low as 0.0016 newtons. The summed output from the load cells was presented to the participant on the television screen in real time. Voltage data acquisition, voltage-to-force transformation, and stimuli presentation were all conducted using customized programs written in LabVIEW (National Instruments, Austin, TX). Stimuli were presented on the television screen.

Fig 2 — A) The precision grip apparatus with load cells under the thumb and index finger; B) the experimental procedure was 130 s in length. Each block of 20 s of force was separated by 10 s of rest; C) the visual display contained two horizontal bars presented against a black background. The target bar (white) was stationary, and the red/green force bar provided real-time visual feedback. In the visually guided task, visual feedback was available for the duration of the trial. In the memory-guided task, the force bar disappeared for the last 12 s of the trial.

The visually and memory-guided force tasks in the current study have been previously used to study visuomotor control in autism spectrum disorders [18, 32], attention-deficit/hyperactivity disorder [33], Parkinson’s disease [34], and younger and older adults [17]. During the task, participants viewed two horizontal bars: a red/green force bar that moved up with increasing force and down with decreasing force, and a static white bar representing target force. The target white bar was set at 25% of the participant’s MVC. The onset and offset of force production were cued by a color change of the moveable force bar. Green served as the go cue and red as the stop cue. Participants were instructed to produce force as quickly and as accurately as possible at the time of the color change from red to green and to keep the green bar at the target force level for the duration of the 20-second trial, until offset of force was cued. As shown in Fig 2B, each run started and ended with 10 seconds of rest and included four 20-second trials of force with 10 seconds of rest in between each trial. During visually guided trials, the moveable force bar was visible for the duration of the trial, providing real-time visual feedback about performance. As shown in Fig 2C, during the memory-guided trials, the force bar disappeared for the last 12 seconds of the trial. Participants were instructed to continue producing force at the target level until the trial ended. Participants completed one run of four 20-second visually guided and one run of four 20-second memory-guided trials. The task order was counterbalanced across participants. All participants completed a brief practice session to become familiar with the timing and force output requirements of the task. The force time series data were digitally filtered using a fourth-order Butterworth filter with a 10 Hz low-pass cut-off frequency. We examined force during the last 12 seconds of each 20-second trial, which represents the time in which visual feedback was removed in the memory-guided condition. Force data were collected in newtons and were divided by the participant’s MVC measured on the same day, multiplied by 100%. Therefore, the data were analyzed as a percent of MVC.

Statistical analysis

All data were used in the analysis, but time in seconds and trial number were not included as factors. Therefore, the analysis design was repeated measures with two main design factors—day (baseline, sleep restriction, recovery sleep) and vision condition (visually guided and memory-guided)—and four potential covariates (race, age, TST, and KSS score).

We used a mixed effects multilevel approach to analyze the effects of sleep restriction and visual feedback on mean force produced in the last 12 seconds of each trial, normalized by MVC, in addition to modeling how the effects of day and vision condition varied across individuals. Linear mixed effects models employ a partial pooling method of data aggregation [35, 36], which allows all available data to be used (all time points and all trials per day per condition). The method shrinks the estimates toward the mean estimate, which includes but lessens the effect of the extreme outcomes in the data.

A series of models were estimated in a two-level multilevel framework using the lme4 package [36] in R [37] to model the mean force, normalized by MVC, across days and vision condition, nested within participants. All models were initially estimated using restricted maximum likelihood (REML). Models were compared using the Akaike’s Information Criterion (AIC), where a lower AIC indicated a better fit to the data [38]. Within the best-fitting model, analyses of variance with Satterthwaite’s method of determining degrees of freedom were used to determine if the interactions and main effects were significant [36, 39], where a priori significance for fixed effects was set at 0.05.

First, a series of random intercept models was estimated to understand the effects of the fixed design factors of the study on participants’ mean force production during the last 12 s of each trial (S1 Table). These factors were day (baseline, sleep restriction, recovery sleep; baseline was the reference day) and vision condition (visually guided and memory-guided; visually guided was the reference condition). Second, a series of random intercept models with potential covariates was estimated (S2 Table). The covariates included race, age, TST, and KSS score, where age, TST, and KSS score were grand mean centered. Third, interactions were added to the resulting random intercept models with covariates that improved model fit by AIC (S3 Table). As this study was among the first to explore the effects of sleep restriction on continuous motor force production, we included interactions between terms to elucidate if and in what manner these covariates interacted with the factors of interest (day and vision condition). Fourth, we estimated a model allowing individual intercepts to vary across day and vision condition to determine if sleep and vision conditions lead to difference in force variability (S4 Table). The best-fitting model of those estimated above, determined by lowest AIC (Lohse, 2020), would be deemed the final model, and the main effects and interactions of this final model would be evaluated.

Results

Participants

Fourteen healthy young adult males (M ± SD, 22 ± 3 years, 9 white non-Hispanic, 3 Asian, 3 black non-Hispanic) participated in this study. Table 1 includes covariates and characteristics of the participants. Data were missing for KSS score for the baseline day for one participant and force data for the sleep restriction day for one participant.

Table 1. Participant characteristics.

Participant	KSS Baseline	KSS Restriction	KSS Recovery	MVC Baseline (N)	MVC Restriction (N)	MVC Recovery (N)
Participant 1	4	9	5	20.02	22.24	26.68
Participant 2	8	9	7	48.90	44.48	73.40
Participant 3	5	7	5	46.71	26.69	33.36
Participant 4	1	3	1	66.72	66.72	66.72
Participant 5	N/A	5	3	73.40	75.62	77.84
Participant 6	3	5	3	73.40	51.15	71.17
Participant 7	1	8	1	40.03	26.68	37.81
Participant 8	7	9	7	26.69	22.24	26.69
Participant 9	4	5	5	57.83	46.71	37.80
Participant 10	3	4	3	46.71	68.95	60.05
Participant 11	6	6	3	66.72	57.83	60.05
Participant 12	4	9	4	53.38	53.38	64.50
Participant 13	2	4	4	42.26	48.93	48.93
Participant 14	1	5	3	93.41	82.29	104.53
Mean	4	6	4	54.01	49.57	56.40
Standard Deviation	2	2	2	19.63	19.77	22.41

Open in a new tab

Note. KSS, Karolinska Sleepiness Scale; MVC, maximum voluntary contraction; N, newtons.

Grip force

The model including random slopes for the day and vision condition (AIC = 1.06x10⁶) fit the data better than the model assuming no variation due to day and vision across individuals (AIC = 1.08x10⁶). Therefore, random slopes across repeated measures factors were included in the final model.

The best-fitting model, shown in Table 2, included main effects for day, vision condition, KSS score, and the three-way interaction among day, vision condition, and KSS score. The final model also included random slopes for both day and vision condition. Fig 3A shows the mean force produced during the final 12 s for each participant and averaged across participants. Although time was not a factor in the model, Fig 3B displays force production (averaged across participants), by day and vision condition, over the final 12 s (which was the time period analyzed from each trial). Table 3 includes the estimated marginal means of force output by vision condition and by day.

Table 2. Results of the multilevel mixed-effects model estimating mean force (%MVC).

Variable	Estimate (SE)
Intercept (Day = Baseline, Vision = VG, KSS = 4)	24.80 (0.16)^***
Day (Restriction vs. Baseline)	-0.84 (0.38)^*
Day (Recovery vs. Baseline)	0.15 (0.23)
Vision (MG vs. VG)	-0.71 (0.29)^*
KSS	-0.02 (0.06)
Day (Restriction vs. Baseline) x Vision (MG vs. VG)	0.10 (0.03)^**
Day (Recovery vs. Baseline) x Vision (MG vs. VG)	0.03 (0.02)
Day (Restriction vs. Baseline) x KSS	0.12 (0.12)
Day (Recovery vs. Baseline) x KSS	0.02 (0.11)
Vision (MG vs. VG) x KSS	-0.21 (0.01)^***
Day (Restriction vs. Baseline) x Vision (MG vs. VG) x KSS	0.31 (0.01)^***
Day (Recovery vs. Baseline) x Vision (MG vs. VG) x KSS	0.31 (0.01)^***
AIC	1059064.60
BIC	1059303.54
Log-likelihood	-529509.30
Num. obs.	239985
Num. groups: Participant	14
Var: Participant (intercept)	0.33
Var: Participant Day (Restriction)	1.03
Var: Participant Day (Recovery)	0.73
Var: Participant Vision (Memory-Guided)	1.14
Cov: Participant (Intercept) Day (Restriction)	-0.03
Cov: Participant (Intercept) Day (Recovery)	-0.42
Cov: Participant (Intercept) Vision (Memory-Guided)	-0.08
Cov: Participant Day (Restriction) x Day (Recovery)	0.23
Cov: Participant Day (Restriction) x Vision (Memory-Guided)	-0.58
Cov: Participant Day (Recovery) Vision (Memory-Guided)	0.01
Var: Residual	4.82

Open in a new tab

Note. SE = standard error; VG = Visually Guided; MG = Memory-Guided; KSS = Karolinska Sleepiness Scale.

***p < 0.001

** p < 0.01

*p < 0.05.

Fig 3 — The red dashed line indicates the target force (25% MVC). A) Average force production over the last 12 s of each trial. The black bar indicates the visually guided condition, and the grey bar indicates the memory-guided condition. Error bars indicate the standard deviation of the mean. Open circles represent each individual participant’s mean force. ***p < 0.001; ** p < 0.01; *p < 0.05. B) Force production over the last 12 s of each trial. The dark black line indicates the visually guided condition, and the grey line indicates the memory-guided condition. Shaded areas represent the standard error of the mean.

Table 3. Estimated marginal means and estimated marginal trends for force (% MVC).

Estimated Marginal Means
	Baseline	Sleep Restriction	Recovery Sleep
Visually Guided	24.8% (0.2%)	24.0% (0.4%)*	24.9% (0.1%)
Memory-guided	24.0% (0.3%)^†	23.5% (0.3%)	24.3% (0.3%)
Estimated Marginal Trends By KSS Score
	Baseline	Sleep Restriction	Recovery Sleep
Visually Guided	-0.02 (0.06)	0.10 (0.12)	-0.003 (0.07)
Memory-guided	-0.23 (0.06)	0.21 (0.12) *	0.09 (0.07) *

Open in a new tab

Note. Force production was measured in percent of maximum voluntary contraction in newtons. Values in paratheses are standard error of the mean. Asterisks indicate significant difference between a given day and Baseline, and daggers indicate significant differences between Visually and memory-guided conditions.

When the model indicated significant interactions or main effects, Satterthwaite pairwise comparisons were conducted. As shown in Fig 3A, there was an effect of vision condition such that participants produced less force in the memory-guided condition than in the visually guided condition at baseline (Β = -0.711%, p = 0.027). There was an effect of day within the visually guided condition such that when visual feedback was present, participants produced less force during the sleep restriction day than during the baseline day (Β = -0.838%, p = 0.041). There was an interaction between day and vision condition such that participants did not produce different force during the sleep restriction day than during the baseline day during memory-guided force production (Β = 0.96%, p = 0.005) (Fig 3A). There was no effect of recovery sleep on visually guided force production (Β = 0.15%, p = 0.520) or on memory-guided production (Β = 0.03%, p = 0.224) (Fig 3A).

There were significant three-way interactions between day, vision condition, and KSS score, shown in Fig 4. Follow-up general linear models were run to estimate the effect of KSS score in each vision condition on each day. The relationship between KSS score and visually guided force production did not differ from baseline to sleep restriction to recovery sleep. However, the relationship between KSS score and memory-guided force production significantly changed after sleep restriction (p = 0.002) and after recovery sleep (p = 0.031) relative to baseline. After sleep restriction and recovery sleep, higher KSS scores (i.e., greater self-reported sleepiness) were associated with higher force production during the memory-guided condition relative to baseline.

To describe the heterogeneity of variance across levels of the repeated measures factors, we investigated the model residuals across levels of day and vision condition, as shown in Fig 5 and described in Table 4. Visual inspection showed that there was larger variability in the residuals (as measured through standard deviation) on the sleep restriction day than at baseline and after recovery sleep, and there was larger variability in residuals during memory-guided than during visually guided force production. However, the effect of sleep restriction on model residual variability was considerably larger during visually guided then during memory-guided force production. This effect was only observed when one participant (Participant 12) was removed from the model. Participant 12 was the only participant with considerable variability in the visually guided condition on the sleep restriction day (standard deviation of 6.7% MVC compared to the sample standard deviation (without Participant 12) of 2.4% MVC). The bottom section of Table 4 includes the variability by day and vision condition with and without Participant 12 included.

Table 4. Standard deviation of the residuals of force production (% MVC) for experimental conditions, with and without Participant 12.

All Participants (N = 14)
	Visually Guided	Memory-Guided
Baseline	0.93%	2.37%
Sleep Restriction	2.90%	2.75%
Recovery	0.74%	2.48%
N = 13
Baseline	0.80%	2.37%
Sleep Restriction	2.32%	2.84%
Recovery	0.75%	2.45%

Open in a new tab

Note. One participant of the fourteen had considerable force variability in the visually guided condition during the sleep restriction day compared to other participants. When that participant was removed from the model (bottom portion of the table), there was a noticeably higher residual variability in the memory-guided condition compared to the visually guided condition during sleep restriction, paralleling the results of the baseline and recovery days.

Finally, to further evaluate model fit, we examined the model residual variability described in Table 2 (Var: Residual). Specifically, after accounting for the between-person fixed and random effects, the standard deviation of the residuals was 2.19% MVC, demonstrating substantial remaining variance (relative to the fixed effects coefficients) in force production after accounting for day, vision condition, KSS, and MVC.

Discussion

The purpose of this study was to investigate the effects of sleep restriction and visual feedback on force production in healthy young adult males. This study had two main findings: (1) Visually guided force production was more sensitive to sleep loss than memory-guided force production, and (2) During memory-guided force production, higher self-reported sleepiness was related to higher force after sleep restriction and recovery sleep.

Visually guided force production was more sensitive to sleep loss than memory-guided force production. Specifically, in the visually guided condition, participants produced less force after four nights of sleep restriction compared to baseline. In contrast, in the memory guided condition, participants’ force output did not differ across baseline, sleep restriction, and sleep recovery. The finding that visually guided force production is more sensitive (to sleep restriction) than memory-guided force production is consistent with literature reporting that memory is less affected by sleep loss than simple attention [6]. Indeed, the maintenance of visually guided force output at a specified target level requires sustained attention and feedback-based corrections [40], both of which may be sensitive to sleep loss. Performing a continuous visually guided force task may be more similar to responding during real-world tasks, such as operating a motor vehicle, compared to a discrete button-pressing tasks commonly employed in sleep loss paradigms.

The effect of sleep restriction on force production varies across people. Figs 3 and 4, and Table 4, visualize and report the between-participant variability in the amount of force produced. Between-participant variability in the memory-guided force condition was consistent across days (baseline, sleep restriction, and recovery sleep). However, the between-participant variability in the visually guided condition was nearly three times higher after sleep restriction than at baseline and after recovery sleep. Considering that visually guided force control requires feedback-based motor corrections [40], the results suggest that for some individuals, feedback-based force control may be more sensitive to sleep restriction than memory-guided force control. The current study’s findings of interindividual effects of sleep restriction on performance corroborate previous research using other cognitive tasks [7]. Future studies could explore the effect of sleep restriction as a longitudinal (over the 12 seconds of gripping) factor in multilevel models rather than as a categorical factor, to determine the potentially compounding effect of sleep loss on visually and memory-guided motor output across time.

Although visually guided force production was impaired after sleep restriction, this was not explained by self-reported sleepiness. Self-reported sleepiness did not relate to force produced during the visually guided condition at baseline, after sleep restriction, or after recovery sleep. As shown in Fig 3, sleep restriction reduced visually guided force output in some individuals more than others; however, self-reported sleepiness was not associated with force output. This finding suggests that the KSS may not be a useful measure to predict visually guided motor output. Self-reported sleepiness did, however, predict the amount of force produced in the memory-guided condition after sleep restriction and after recovery sleep. Specifically, after sleep restriction and recovery sleep, sleepiness predicted more force in the memory-guided condition (Fig 4). Sleepier, sleep-restricted young adults may be less reliant on visual feedback and more reliant on proprioceptive feedback during memory-guided force production. Considering that visual tracking and attention are impaired after sleep restriction [6, 41], it is possible that sleepier sleep-restricted young adults may trigger a compensatory feedforward strategy while completing the memory-guided force production task, which is more cognitively taxing than the visually guided task [6, 41]. Higher sleepiness could induce a compensatory strategy to mitigate a larger perceived decrement in their force production when visual feedback is removed due to reliance on proprioceptive feedback [7]. While self-reported sleepiness predicted force output, TST, measured by actigraphy, did not. As expected, the pseudo-R² for day predicting TST was 0.93, and because TST and day were so highly correlated, TST did not improve force estimation beyond that predicted by day.

Limitations

This study had some limitations. First, the force task asked participants to maintain a grip force output of 25% of MVC. Grip force of 20–40% MVC is important for activities of daily living such as writing, unlocking a door, and tying shoes [42], but it may not be directly applicable to other activities requiring higher or lower grip force (such as operating heavy machinery). Future studies should investigate the effect of sleep restriction on various motor force tasks to better understand the wholistic effect of sleep restriction on movement. Also, we modeled force data without including a longitudinal effect (over the 12 s of gripping). As this was the first study exploring the effect of sleep restriction on grip force output, we were interested in the overall effect (or lack thereof) of sleep restriction and recovery sleep on motor output. Future studies may model continuous force production over time to determine if the effects of sleep restriction we found here change with duration of force production.

Conclusion

Sleep restriction impairs visually but not memory-guided continuous force output. Both mean force and between-participant variability in force are more sensitive to sleep restriction when visual feedback is present. Also, after sleep restriction and recovery sleep, higher levels of self-reported sleepiness are associated with higher force production in the memory-guided condition. Taken together, these findings demonstrate that sleep restriction impacts visually guided force output; however, the response to sleep restriction is heterogeneous, in spite of having tested a homogeneous sample of young, healthy males, under well-controlled experimental conditions. Additionally, sleep restriction impairs visually guided force output regardless of perceived sleepiness; however, in the memory-guided condition, force becomes more impaired with increasing levels of perceived alertness.

Supporting information

S1 Table. Random intercept models.

Note. Values given in Estimate (Standard Error); VG = Visually Guided; MG = Memory-Guided. First series of random intercept models estimated to understand the effects of the fixed design factors of the study on percent of a participant’s MVC during the last 12 sec of each trial. The best-fitting model of the series—determined by AIC—was Model 4: Day*Vision interactions. ***p < 0.001; ** p < 0.01; *p < 0.05.

(DOC)

Click here for additional data file.^{(647.8KB, doc)}

S2 Table. Random intercept models with potential covariates.

Note. Values given in Estimate (Standard Error); VG = Visually Guided; MG = Memory-Guided; TST = Total Sleep Time; KSS = Karolinska Sleepiness Scale. Second series of random intercept models with potential covariates. The best-fitting model of the series—determined by AIC—was Model 4: KSS. ***p < 0.001; ** p < 0.01; *p < 0.05.

(DOCX)

Click here for additional data file.^{(21.4KB, docx)}

S3 Table. Random intercept models with study design and covariates.

Note. Values given in Estimate (Standard Error); VG = Visually Guided; MG = Memory-Guided; KSS = Karolinska Sleepiness Scale. Third series of random intercept models with interactions between study design variables and covariates that improved model fit by AIC. The best-fitting model of the series—determined by AIC—was Model 2: Main effects, all interactions. ***p < 0.001; ** p < 0.01; *p < 0.05.

(DOCX)

Click here for additional data file.^{(18.8KB, docx)}

S4 Table. Random slopes models with study design and KSS.

Note. Values given in Estimate (Standard Error); VG = Visually Guided; MG = Memory-Guided; KSS = Karolinska Sleepiness Scale. Fourth series of models, comparing random slopes for day and vision and no random slopes. The best-fitting model of the series—determined by AIC—was Model 4: Random Slopes on Day and Vision. ***p < 0.001; ** p < 0.01; *p < 0.05.

(DOCX)

Click here for additional data file.^{(22.4KB, docx)}

S1 Data. Demographics, day, total sleep time, KSS, MVC, vision condition, trial, time in seconds, and force in newtons.

(CSV)

Click here for additional data file.^{(18.3MB, csv)}

Acknowledgments

We thank the individuals who participated in the study. We also thank the students and staff of the Chang and Buxton laboratories for their assistance with study procedures, particularly Nicole Nahmod for participant recruitment; the staff of the Clinical Research Center; Chloe House, PhD, who conducted the psychological interviews; and study collaborators.

Data Availability

All relevant data are within the manuscript and its Supporting Information files.

Funding Statement

This project was supported by a pilot grant (PI: A-M. Chang) from the Pennsylvania State University Clinical and Translational Sciences Institute (funded by the National Center for Advancing Translational Sciences, National Institutes of Health (NIH), through Grant UL1TR002014) and institutional funds from the College of Health and Human Development of the Pennsylvania State University (to A-M. Chang and O.M. Buxton). The Research Electronic Data Capture (REDCap) survey platform and the Clinical Research Center are supported by the National Center for Advancing Translational Sciences, NIH, through grants UL1 TR002014 and UL1 TR00045. The content is solely the responsibility of the authors and does not necessarily represent the official views of the NIH. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.

References

1.Watson NF, Badr MS, Belenky G, Bliwise DL, Buxton OM, Buysse D, et al. Joint Consensus Statement of the American Academy of Sleep Medicine and Sleep Research Society on the Recommended Amount of Sleep for a Healthy Adult: Methodology and Discussion. Sleep. 2015;38(8):1161–83. doi: 10.5665/sleep.4886 [DOI] [PMC free article] [PubMed] [Google Scholar]
2.Centers for Disease Control and Prevention, National Center for Chronic Disease Prevention and Health Promotion, Division of Population Health. Sleep and Sleep Disorders: Data and Statistics 2017. [Available from: https://www.cdc.gov/sleep/data_statistics.html]. [Google Scholar]
3.Belenky G, Wesensten NJ, Thorne DR, Thomas ML, Sing HC, Redmond DP, et al. Patterns of performance degradation and restoration during sleep restriction and subsequent recovery: a sleep dose‐response study. Journal of Sleep Research. 2003;12:1–12. doi: 10.1046/j.1365-2869.2003.00337.x [DOI] [PubMed] [Google Scholar]
4.Dinges DF, Pack F, Williams K, Gillen KA, Powell JW, Ott GE, et al. Cumulative sleepiness, mood disturbance, and psychomotor vigilance performance decrements during a week of sleep restricted to 4–5 hours per night. Sleep. 1997;4:267–77. [PubMed] [Google Scholar]
5.Lim J, Dinges DF. Sleep deprivation and vigilant attention. Ann N Y Acad Sci. 2008;1129:305–22. doi: 10.1196/annals.1417.002 [DOI] [PubMed] [Google Scholar]
6.Lim J, Dinges DF. A meta-analysis of the impact of short-term sleep deprivation on cognitive variables. Psychol Bull. 2010;136(3):375–89. doi: 10.1037/a0018883 [DOI] [PMC free article] [PubMed] [Google Scholar]
7.Mathew GM, Strayer SM, Bailey DS, Buzzell K, Ness KM, Schade MM, et al. Changes in Subjective Motivation and Effort During Sleep Restriction Moderate Interindividual Differences in Attentional Performance in Healthy Young Men. Nat Sci Sleep. 2021;13:1117–36. doi: 10.2147/NSS.S294409 [DOI] [PMC free article] [PubMed] [Google Scholar]
8.Van Dongen HPA, Maislin G, Mullington JM, Dinges DF. The Cumulative Cost of Additional Wakefulness: Dose-Response Effects on Neurobehavioral Functions and Sleep Physiology From Chronic Sleep Restriction and Total Sleep Deprivation. Sleep. 2003;26(2). doi: 10.1093/sleep/26.2.117 [DOI] [PubMed] [Google Scholar]
9.Drake CL, Roehrs TA, Burduvali E, Bonahoom A, Rosekind M, Roth T. Effects of rapid versus slow accumulation of eight hours of sleep loss. Psychophysiology. 2001;28:979–87. doi: 10.1111/1469-8986.3860979 [DOI] [PubMed] [Google Scholar]
10.Santisteban JA, Brown TG, Ouimet MC, Gruber R. Cumulative mild partial sleep deprivation negatively impacts working memory capacity but not sustained attention, response inhibition, or decision making: a randomized controlled trial. Sleep Health. 2019;5(1):101–8. doi: 10.1016/j.sleh.2018.09.007 [DOI] [PubMed] [Google Scholar]
11.Baddeley A. Working Memory. Science. 1992;255. [DOI] [PubMed] [Google Scholar]
12.Adams JA. A closed-loop theory of motor learning. J Mot Behav. 1971;3(2):111–49. doi: 10.1080/00222895.1971.10734898 [DOI] [PubMed] [Google Scholar]
13.Fitts PM. Perceptual-motor skill learning. Categories of human learning: Academic Press; 1964. p. 243–85. [Google Scholar]
14.Gerver CR, Neely KA, Kurkela KA, Diaz MT, Goodman JT, Blouch S, et al. Shared neural recruitment across working memory and motor control tasks as a function of task difficulty and age. Neuropsychol Dev Cogn B Aging Neuropsychol Cogn. 2020;27(6):864–79. doi: 10.1080/13825585.2019.1700898 [DOI] [PMC free article] [PubMed] [Google Scholar]
15.Marvel CL, Morgan OP, Kronemer SI. How the motor system integrates with working memory. Neurosci Biobehav Rev. 2019;102:184–94. doi: 10.1016/j.neubiorev.2019.04.017 [DOI] [PMC free article] [PubMed] [Google Scholar]
16.Neely KA, Wang P, Chennavasin AP, Samimy S, Tucker J, Merida A, et al. Deficits in inhibitory force control in young adults with ADHD. Neuropsychologia. 2017;99:172–8. doi: 10.1016/j.neuropsychologia.2017.03.012 [DOI] [PMC free article] [PubMed] [Google Scholar]
17.Neely KA, Samimy S, Blouch SL, Wang P, Chennavasin A, Diaz MT, et al. Memory-guided force control in healthy younger and older adults. Exp Brain Res. 2017;235(8):2473–82. doi: 10.1007/s00221-017-4987-3 [DOI] [PMC free article] [PubMed] [Google Scholar]
18.Neely KA, Mohanty S, Schmitt LM, Wang Z, Sweeney JA, Mosconi MW. Motor Memory Deficits Contribute to Motor Impairments in Autism Spectrum Disorder. J Autism Dev Disord. 2019;49(7):2675–84. doi: 10.1007/s10803-016-2806-5 [DOI] [PMC free article] [PubMed] [Google Scholar]
19.Doran SM, Dongen HPA, Dinges DF. Sustained attention performance during sleep deprivation evidence of state instability. Archives Italiennes de Biologie. 2001;139:253–67. [PubMed] [Google Scholar]
20.Jung CM, Ronda JM, Czeisler CA, Wright KP Jr., Comparison of sustained attention assessed by auditory and visual psychomotor vigilance tasks prior to and during sleep deprivation. J Sleep Res. 2011;20(2):348–55. doi: 10.1111/j.1365-2869.2010.00877.x [DOI] [PMC free article] [PubMed] [Google Scholar]
21.Mahajan K, Velaga NR. Effects of Partial Sleep Deprivation on Braking Response of Drivers in Hazard Scenarios. Accid Anal Prev. 2020;142:105545. doi: 10.1016/j.aap.2020.105545 [DOI] [PubMed] [Google Scholar]
22.Mahajan K, Velaga NR. Sleep-deprived car-following: Indicators of rear-end crash potential. Accid Anal Prev. 2021;156:106123. doi: 10.1016/j.aap.2021.106123 [DOI] [PubMed] [Google Scholar]
23.Baker FC, Driver HS. Circadian rhythms, sleep, and the menstrual cycle. Sleep Med. 2007;8(6):613–22. [DOI] [PubMed] [Google Scholar]
24.Mander BA, Winer JR, Walker MP. Sleep and Human Aging. Neuron. 2017;94(1):19–36. doi: 10.1016/j.neuron.2017.02.004 [DOI] [PMC free article] [PubMed] [Google Scholar]
25.Douglass AB, Bornstein R, German N-M, Keenan S, Laughton M, Zarcone VP, et al. The Sleep Disorders Questionnaire I: Creation and Multivariate Structure of SDQ. Sleep. 1994;17(2):160–7. [DOI] [PubMed] [Google Scholar]
26.Ness KM, Strayer SM, Nahmod NG, Chang AM, Buxton OM, Shearer GC. Two nights of recovery sleep restores the dynamic lipemic response, but not the reduction of insulin sensitivity, induced by five nights of sleep restriction. Am J Physiol Regul Integr Comp Physiol. 2019;316(6):R697–R703. doi: 10.1152/ajpregu.00336.2018 [DOI] [PMC free article] [PubMed] [Google Scholar]
27.Figueiro MG, Bierman A, Plitnick B, Rea MS. Preliminary evidence that both blue and red light can induce alertness at night. BMC Neurosci. 2009;10:105. doi: 10.1186/1471-2202-10-105 [DOI] [PMC free article] [PubMed] [Google Scholar]
28.Chang AM, Aeschbach D, Duffy JF, Czeisler CA. Evening use of light-emitting eReaders negatively affects sleep, circadian timing, and next-morning alertness. Proc Natl Acad Sci U S A. 2015;112(4):1232–7. doi: 10.1073/pnas.1418490112 [DOI] [PMC free article] [PubMed] [Google Scholar]
29.Marino M, Li Y, Rueschman MN, Winkelman JW, Ellenbogen JM, Solet JM, et al. Measuring sleep: accuracy, sensitivity, and specificity of wrist actigraphy compared to polysomnography. Sleep. 2013;36(11):1747–55. doi: 10.5665/sleep.3142 [DOI] [PMC free article] [PubMed] [Google Scholar]
30.Åkerstedt T, Gillberg M. Subjective and Objective Sleepiness in the Active Individual. International Journal of Neuroscience. 1990;52:29–37. doi: 10.3109/00207459008994241 [DOI] [PubMed] [Google Scholar]
31.Harris PA, Taylor R, Thielke R, Payne J, Gonzalez N, Conde JG. Research electronic data capture (REDCap)—a metadata-driven methodology and workflow process for providing translational research informatics support. J Biomed Inform. 2009;42(2):377–81. doi: 10.1016/j.jbi.2008.08.010 [DOI] [PMC free article] [PubMed] [Google Scholar]
32.Wang Z, Magnon GC, White SP, Greene RK, Vaillancourt DE, Mosconi MW. Individuals with autism spectrum disorder show abnormalities during initial and subsequent phases of precision gripping. J Neurophysiol. 2015;113(7):1989–2001. doi: 10.1152/jn.00661.2014 [DOI] [PMC free article] [PubMed] [Google Scholar]
33.Neely KA, Chennavasin AP, Yoder A, Williams GK, Loken E, Huang-Pollock CL. Memory-guided force output is associated with self-reported ADHD symptoms in young adults. Exp Brain Res. 2016;234(11):3203–12. doi: 10.1007/s00221-016-4718-1 [DOI] [PMC free article] [PubMed] [Google Scholar]
34.Vaillancourt DE, Slifkin AB, Newell KM. Visual control of isometric force in Parkinson’s disease. Neuropsychologia. 2001;39:1410–8. doi: 10.1016/s0028-3932(01)00061-6 [DOI] [PubMed] [Google Scholar]
35.Gelman A. Multilevel (Hierarchical) Modeling: What It Can and Cannot Do. Technometrics. 2006;48(3):432–5. [Google Scholar]
36.Bates D. Fitting linear mixed-effects models using lme4. Journal of Statistical Software. 2015;67(1):1–48. [Google Scholar]
37.R Core team. R: A language and environment for statistical computing Vienna, Austria: R Foundation for Statistical Computing; 2019. [Available from: https://www.R-project.org/. [Google Scholar]
38.Lohse KR, Shen J, Kozlowski AJ. Modeling Longitudinal Outcomes: A Contrast of Two Methods. Journal of Motor Learning and Development. 2020;8(1):145–65. [Google Scholar]
39.Satterthwaite FE. An Approximate Distribution of Estimates of Variance Components. Biometrics Bulletin. 1946;2(6):110–4. [PubMed] [Google Scholar]
40.Neely KA, Planetta PJ, Prodoehl J, Corcos DM, Comella CL, Goetz CG, et al. Force control deficits in individuals with Parkinson’s disease, multiple systems atrophy, and progressive supranuclear palsy. PLoS One. 2013;8(3):e58403. doi: 10.1371/journal.pone.0058403 [DOI] [PMC free article] [PubMed] [Google Scholar]
41.Heaton KJ, Maule AL, Maruta J, Kryskow EM, Ghajar J. Attention and visual tracking degradation during acute sleep deprivation in a military sample. Aviat Space Environ Med. 2014;85(5):497–503. doi: 10.3357/asem.3882.2014 [DOI] [PubMed] [Google Scholar]
42.Marshall Matthew M., Armstrong TJ. Observational Assessment of Forceful Exertion and the Perceived Force Demands of Daily Activities. Journal of Occupational Rehabilitation. 2004;14(4). doi: 10.1023/b:joor.0000047430.22740.57 [DOI] [PubMed] [Google Scholar]

PLoS One. doi: 10.1371/journal.pone.0274121.r001

Decision Letter 0

Kenichi Shibuya

2 May 2022

PONE-D-22-02851Sleep Restriction Impairs Visually and Memory-Guided Force ControlPLOS ONE

Dear Dr. Brinkerhoff,

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.

Please submit your revised manuscript by Jun 16 2022 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: https://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,

Kenichi Shibuya, Ph.D.

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

Additional Editor Comments (if provided):

I have completed my evaluation of your manuscript. The reviewers recommend reconsideration of your manuscript following major revision. I invite you to resubmit your manuscript after addressing the comments below. When revising your manuscript, please consider all issues mentioned in the reviewers' comments carefully: please outline every change made in response to their comments and provide suitable rebuttals for any comments not addressed. Please note that your revised submission may need to be re-reviewed.

[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: Partly

Reviewer #2: No

**********

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

Reviewer #1: I Don't Know

Reviewer #2: I Don't Know

**********

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: Yes

Reviewer #2: Yes

**********

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

**********

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: I reviewed the manuscript “Sleep restriction impairs visually and memory-guided force control” by Brinkerhoff and Colleagues. The authors report results from a study in which male participants produced isometric force between the index finger and thumb at various times as they underwent a sleep restriction paradigm. These participants’ ability to produce a steady level of force, and to maintain it under visual feedback, decreased when they were sleep-restricted. Overall I think the experiment described is technically sound; my main concern is that the unconventional statistical methods make the results difficult to understand.

Main:

1. The treatment of time-series data is confusing, although I think it is alluded to around Line 191. Does this mean that the input data for the model fitting are all timepoints for each trial (e.g. 62.5x12 = 750 samples)? This seems to be the case given the number of observations in Table 2, but I am concerned that each datapoint is treated as independent to be fit by the model, when there is time series dependence between certain sets of data points, e.g. if the participant is producing 24% MVC at time t, they are more likely to produce 24.3% than 5% at t+1. I think the authors should more clearly lay out how time-series data are treated and explain why it is alright not to include a “time in trial” or similar covariate in the model. Overall I would strongly suggest the authors use more summarized data rather than the current approach.

2. The statistics are also confusing because the statistical analysis is not carried out on the actual interpreted outcome variables. For example, the variability of participants’ force data is referred to in terms of magnitude of standard deviation changing in the results, but the actual tests are on the size of model residuals. While these are related, it seems like testing the standard deviations themselves would be clearer, especially because it’s unclear how standard deviation is calculated in light of the handling of time series data: what is “N” the denominator of SD? SD across participants and trials?

2a. I think the visual appearance that more sleepiness results in “better” force production -- that is: closer to the target level (Figure 4B) -- is another counterintuitive outcome of this analysis design. Similarly force production is not averaged but the data in Fig 3A still are means and error bars are calculated based on trial-by-trial or participant-by-participant SEM, although these measures are not what the statistics tested exactly.

3. The justification for use of a “mixed effects approach” (line 186) is confusing because it makes it seem like there is a direct association between variability of motor output and model variance, although the association is not so direct. Further, the approach that ends up being used here still has the same requirements, and homoscedasticity is never assessed. Overall, I think that more conventional statistics (perhaps using data transformation if necessary) would make the results much easier to interpret.

Minor:

1. My impression is that the association between working memory and memory-guided force production is usually related to how much force production changes after visual feedback is removed (e.g. Vaillancourt & Russell 2002, EBR). Is there a reason this measure was not assessed? This seems potentially important if some of the significance of the study is related to working memory.

Reviewer #2: The authors showed that sleep restriction negatively affected accuracy and variability of force production. One night of recovery sleep was effective to return to the baseline performance only for the visual-guided condition. Although the study results were interesting, this reviewer has concerns, including the details of statistical analysis and how those were reported, and the study rationale.

Specific comments:

Introduction

- # 13: The author mentioned that sleep restriction negatively affects working memory. However, there was a lack of explanations about what working memory is and why investigating the effects of sleep restriction on working memory is essential. Please provide more explanation and supporting evidence that supports your study rationale.

- # 17: The participants in this study were young, but the major reference that explains the study rationale comes from a study that examined older adults.

- # 25 – 31: Many daily living tasks especially involving manual dexterity (e.g., hands) require continuous modulation of force output, but the experimental task used in this study was not the case. Please justify how the experimental task used in this study represents daily motor tasks.

Methods

- #207: The author stated that analyses of variance with Satterthwaite’s method was used. More detail is needed to explain the aim of using the method, which could benefit readers who are not familiar with the method.

- #215: It seemed that there were three main factors, but the statistical section would be improved by having more details about what final models the author used and how the results will be interpreted in the Results section.

Results

- Readers could better understand when more direct explanations for the statistical results were included. The authors utilized a mixed model to examine the main and interaction effects with coefficient values in Table 2. However, it would be challenging to interpret the entire table and then match the results in the table with what the authors described their findings in the results section (also for the Discussion). For example, the authors may first describe the best model and the results used to describe the mean differences with statistical results (e.g. p values) in the Results (e.g., Figure 3).

- #250: Please provide numbers (“these effects were fairly small in magnitude”).

- Table 2: It was challenging to understand this table. As mentioned for the statistical section, there were three main factors (please correct me if I am wrong), and I expected some information about significant interaction between Day (three levels; Baseline, Restriction, and Recovery) and Vision (two levels; Visually-Guided and Memory-Guided). The author should provide more details about how Table 2 can be interpreted..

- When an interaction effect is significant, pair-wise comparisons are usually the next step, as shown in previous studies using ANOVA, but those statistical results with p values were not included. Did not the mixed model provide that information? A similar question was raised in Figure 3 also. Statistical results from within-group comparisons only seemed to be shown, so please provide statistical results for the other paired comparisons (e.g., Visually vs. Memory in the baseline).

- # 285: The authors interpreted the main effect of vision condition to describe their finding, but there was no further information on it. Adding a supporting statistical result can strengthen the sentence.

- Figure 5: Most results were reported by force relative to MVC (% MVC), but in Figure 5 the data were in newtons. Please clarify what values (e.g., N or % MVC) were used in that Figure. Could the normalization affect the findings?

Discussion

- The study findings were from an isometric pinching force production at a specific constant force output (25% MVC). The authors needed to address how the findings can be generalizable in other force production tasks that include different force production types (e.g., concentric), different amounts of force levels (less or higher than 25% MVC), and force production trajectories (e.g., continuous control of force level). Adding that information in a separate study limitation section can address this issue.

**********

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

[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.

PLoS One. 2022 Sep 2;17(9):e0274121. doi: 10.1371/journal.pone.0274121.r002

Author response to Decision Letter 0

24 Jun 2022

Editor Comments (if provided):

REVIEWER #1:

I reviewed the manuscript “Sleep restriction impairs visually and memory-guided force control” by Brinkerhoff and Colleagues. The authors report results from a study in which male participants produced isometric force between the index finger and thumb at various times as they underwent a sleep restriction paradigm. These participants’ ability to produce a steady level of force, and to maintain it under visual feedback, decreased when they were sleep-restricted. Overall I think the experiment described is technically sound; my main concern is that the unconventional statistical methods make the results difficult to understand.

Author Response:

We thank the reviewer for their helpful comments and suggestions to enhance our manuscript. We have addressed each individual comment below. However, before doing so, we would like to mention two major revisions in response to reviewer feedback. First, we normalized force output to each participant’s maximum voluntary contraction (MVC) before entering force output into the statistical model. As a result, MVC has been removed from the model. We believe this change helps to clarify the Results. This change in our analysis affected the Results, such that the relationship between sleepiness and visual condition depended on the day; higher self-reported sleepiness after sleep restriction and recovery sleep led to smaller memory-guided force.

Main comments:

1. The treatment of time-series data is confusing, although I think it is alluded to around Line 191. Does this mean that the input data for the model fitting are all timepoints for each trial (e.g. 62.5x12 = 750 samples)? This seems to be the case given the number of observations in Table 2, but I am concerned that each datapoint is treated as independent to be fit by the model, when there is time series dependence between certain sets of data points, e.g. if the participant is producing 24% MVC at time t, they are more likely to produce 24.3% than 5% at t+1. I think the authors should more clearly lay out how time-series data are treated and explain why it is alright not to include a “time in trial” or similar covariate in the model. Overall, I would strongly suggest the authors use more summarized data rather than the current approach.

Author Response:

We appreciate the reviewer indicating the importance of considering time within our models. Yes, we used all data points within the models, and then estimated the effects of two design factors: day and vision condition. In this case, as the reviewer suggests, each data point was treated as independent from the point(s) before and after it in time. We agree that the time-series nature of the force data means that data samples close in time to one another are correlated with one other. One option to handle the data, as the reviewer suggests, is to average the data over the 12 seconds and across trials, and each participant would have a single data point per day, per vision condition (complete pooling). However, this method of pooling the data gives equal weight to extreme values in the outcome. The alternative method, used in our paper, is to allow the mixed models to aggregate the data across time, but not including time as a factor in the models. Accordingly, the linear mixed models also use averages, but they calculate those averages in the modeling process using partial pooling. This approach has three advantages: It 1) uses all available information in the data (all time points and trials within a day and condition), 2) weights the averages by the amount of information available (this is not a big benefit in this study as people have almost exactly the same number of time points in each trial), and 3) shrinks the estimates toward the mean estimate, which still includes, but lessens the effect of, extreme outcomes observed in the data (Bates, 2015; Gelman & Hill, 2008). Partial pooling assumes that each datapoint is some deviation from the average (i.e., regression to the mean) and applies less weight to datapoints that are not removeable outliers by design and should therefore be included in the dataset, which would otherwise pull the average in a direction that is not indicative of the participants’ true average force produced (Gelman & Hill, 2008). Therefore, the mixed models approach allows us to use all datapoints while more accurately summarizing the participant’s true force output (per day, per vision condition) than we would get if we manually averaged across design factors. While complete pooling and partial pooling both offer ways to summarize the data across time and trials, we believe partial pooling is more ecologically valid in this case.

This benefit is now emphasized in the statistical methods section of the manuscript:

Line 183-191: “All data were used in the analysis, but time in seconds and trial number were not included as factors. Therefore, the analysis design was repeated measures with two main design factors: day and vision condition. We used a mixed effects multilevel approach to analyze the mean effects of sleep restriction and visual feedback on motor force production, in addition to modeling how the effects of day and vision condition varied across individuals. Linear mixed effects models employ a partial pooling method of data aggregation (Gelman, 2006; Bates, 2015), which allows all available data to be used (all time points and all trials per day per condition). The method shrinks the estimates toward the mean estimate, which includes but lessens the effect of the extreme outcomes in the data.”

Additionally, we did not include a variable for time-in-trial because the question posed by this study was not longitudinal (over the 12 seconds of gripping) in nature; rather, we were interested in the overall effect of sleep restriction and recovery sleep on continuous force production.

Author Response:

We have revised the manuscript wording to make clearer the outcomes we examined. We examined the variation across participants estimated by the residuals, not the variation within participants.

The Results section now reads:

Line 308-321: “To describe the heterogeneity of variance across levels of the repeated measures factors, we investigated the model residuals across levels of day and vision condition, as shown in Figure 5 and described in Table 4. Visual inspection showed that there was larger variability in the residuals (as measured through standard deviation) on the sleep restriction day than at baseline and after recovery sleep, and there was larger variability in residuals during memory-guided than during visually guided force production. However, the effect of sleep restriction on model residual variability was considerably larger during visually guided then during memory-guided force production. This effect was only observed when one participant (Participant 12) was removed from the model. Participant 12 was the only participant with considerable variability in the visually guided condition on the sleep restriction day (standard deviation of 6.7% MVC compared to the sample standard deviation (without Participant 12) of 2.4% MVC). The bottom section of Table 4 includes the variability by day and vision condition with and without Participant 12 included.”

Line 336-341: “Finally, to further evaluate model fit, we examined the model residual variability described in Table 2 (Var: Residual). Specifically, after accounting for the between-person fixed and random effects, the standard deviation of the residuals was 2.19% MVC, demonstrating substantial remaining variance (relative to the fixed effects coefficients) in force production after accounting for day, vision condition, KSS, and MVC.”

Author Response:

We agree that this finding is interesting and initially counterintuitive. The revised Discussion now includes more emphasis on this finding:

Line 386-396: “Specifically, after sleep restriction and recovery sleep, sleepiness predicted more force in the memory-guided condition (Figure 4). Sleepier, sleep-restricted young adults may be less reliant on visual feedback and more reliant on proprioceptive feedback during memory-guided force production. Considering that visual tracking and attention are impaired after sleep restriction (Lim & Dinges, 2010; Heaton et al, 2014), it is possible that sleepier sleep-restricted young adults may trigger a compensatory feedforward strategy while completing the memory-guided force production task, which is more cognitively taxing than the visually guided task (Lim & Dinges, 2010; Heaton et al, 2014). Higher sleepiness could induce a compensatory strategy to mitigate a larger perceived decrement in their force production when visual feedback is removed due to reliance on proprioceptive feedback (Mathew et al, 2021).”

The reviewer is correct that we included all datapoints in the model (please see detailed explanation in response to point #1). Since the model aggregates the data over any not-included factors (trial number and time in seconds), the visualization of the force magnitudes in Figure 3A are reflective of the way the model handles the data: by creating one summarized datapoint per person per day per vision condition.

Author Response:

We thank the reviewer for raising these concerns. We have rewritten the statistical methods to improve clarity.

The first paragraph of the statistical methods section now reads:

Line 184-188: “Therefore, the analysis design was repeated measures with two main design factors: day and vision condition. We used a mixed effects multilevel approach to analyze the mean effects of sleep restriction and visual feedback on motor force production, in addition to modeling how the effects of day and vision condition varied across individuals.”

The second paragraph in the results section now reads:

Line 234-237: “The model including random slopes for the day and vision condition fit the data better than the model assuming no variation due to day and vision across individuals. Therefore, random slopes across repeated measures factors were included in the final model.”

Finally, we did not formally test for homogeneity of variance. However, the residuals of the mixed models allow us to discuss the variance across levels of repeated measures factors:

Minor:

Author Response:

Indeed, the paradigm investigated here is that of Vaillancourt and Russell (2002, EBR), in which a finding of “force decay” was revealed when visual feedback was removed. In previous work, we evaluated the presence (or absence) of decay during this “no vision” period (i.e., Neely et al., 2019; Neely, Samimy, et al., 2017; Neely, Wang, et al., 2017). The focus of those four papers (Vaillancourt & Russell, 2022, Neely et al., 2019; Neely, Samimy, et al., 2017; Neely, Wang, et al., 2017) was a between-group analysis to examine how force decay differs in clinical populations.

In the current work, we utilized the Vaillancourt and Russell (2002) paradigm to evaluate the between- and within- subjects effects of sleep restriction on both visually and memory guided force control. Our examination of force output in these conditions revealed large between-subject variability. In other words, we observed interindividual differences in the response to sleep restriction in both vision conditions. Therefore, the goal of the manuscript was not to examine force decay.

Reviewer 2 suggested we add to the limitations section how the findings “can be generalizable in other force production tasks that include…force production trajectories (e.g., continuous control of force level).” Therefore, the limitations section now includes the following:

Line 407-413: “Also, we modeled force data without including a longitudinal effect (over the 12 s of gripping). As this was the first study exploring the effect of sleep restriction on grip force output, we were interested in the overall effect (or lack thereof) of sleep restriction and recovery sleep on motor output. Future studies may model continuous force production over time to determine if the effects of sleep restriction we found here change with duration of force production.”

REVIEWER #2:

The authors showed that sleep restriction negatively affected accuracy and variability of force production. One night of recovery sleep was effective to return to the baseline performance only for the visual-guided condition. Although the study results were interesting, this reviewer has concerns, including the details of statistical analysis and how those were reported, and the study rationale.

Author Response:

We thank the reviewer for this suggestion. In response, we provide additional background literature as well as clear rationale for studying the effect of sleep restriction on memory-based motor performance.

Line 12-20: “Performance in more complex domains, such as working memory, is also impaired after sleep restriction (Mathew et al, 2021; Drake et al, 2001; Santisteban et al, 2019). Working memory is the temporary storage and manipulation of information, and is an intersection between memory, perception, and the attentional control of behavior (Baddeley, 1992), and is necessary for motor skill acquisition (Adams, 1971; Fitts, 1964). Indeed, the motor system and working memory tasks recruit common neural pathways (Gerver et al., 2020; Marvel, Morgan, & Kronemer, 2019; Neely et al., 2019; Neely, Samimy, et al., 2017; Neely, Wang, et al., 2017). Considering that sleep restriction negatively affects working memory, it follows that sleep restriction may also affect performance of motor tasks that rely on working memory. However, there is a dearth of research examining such tasks.”

- # 17: The participants in this study were young, but the major reference that explains the study rationale comes from a study that examined older adults.

Author Response:

Indeed, one of the seminal papers on memory-guided force output examined older adults with and without Parkinson’s disease (Vaillancourt et al, 2001). More recently, our team has published work examining memory-guided force in individuals ranging from age 7 to 85 (Neely & Chennavasin et al, 2016; Neely & Mohanty et al, 2016; Neely et al, 2017). Additionally, impairments in working memory after sleep restriction have been reported in younger adults (Drake et al., 2001; Mathew et al., 2021; Santisteban, Brown, Ouimet, & Gruber, 2019).

We have added a review reference that encompasses the broad range of literature supporting the overlap between motor system pathways and working memory:

Line 16-17:“Indeed, the motor system and working memory tasks recruit common neural pathways (Gerver et al., 2020; Marvel, Morgan, & Kronemer, 2019; Neely et al., 2019; Neely, Samimy, et al., 2017; Neely, Wang, et al., 2017)."

Author Response:

Thank you for indicating the importance of examining how our task translates to real-world tasks. We revised the real-life examples to be more reflective of the task used in this study, which required participants to moderate continuous grip force for 12 seconds. The sentence now reads:

Line 28: “Furthermore, many occupational tasks and activities of daily living require relatively short periods of continuous motor output, such as carrying a cup of hot coffee.”

Author Response:

This Satterthwaite method of estimating degrees of freedom for a two-sample t-test is used when only the estimates of the variance of distributions are known, as in mixed models. We updated the sentence to include the citation for the package used in R that estimates the degrees of freedom for ANOVAs within mixed models using Satterthwaite’s method. We also cited Satterthwaite’s original paper on this method. The sentence now reads:

Line 196-199: “Within the best-fitting model, analyses of variance with Satterthwaite's method of determining degrees of freedom were used to determine if the interactions and main effects were significant (Bates, 2015; Satterthwaite, 1946), where a priori significance for fixed effects was set at 0.05.”

Author Response:

We added statements to the methods and results sections to better clarify which model was chosen as the final model that was subsequently evaluated for interactions and main effects of the three main factors, as the reviewer stated.

In the statistical methods section, we added the statement:

Line 215-217: “The best-fitting model of those estimated above, determined by lowest AIC (Lohse, 2020), would be deemed the final model, and the main effects and interactions of this final model would be evaluated.”

In the results section, we added the statement:

Results

Author Response:

Thank you for the suggestion to improve clarity. We have revised the results section to include statistical results so that readers can follow the statistics without referring to the table. As an example:

Line 277-279: “As shown in Figure 3A, there was an effect of vision condition such that participants produced less force in the memory-guided condition than in the visually guided condition at baseline (Β = -0.711%, p = 0.027).”

- #250: Please provide numbers (“these effects were fairly small in magnitude”).

Author Response:

During manuscript revision, this sentence was removed.

Author Response:

We thank the reviewer for this suggestion. The interactions between Day and Vision condition are presented in the 6th and 7th lines of Table 2.

In addition to the revisions to the Results section noted above, we revised both the table and the footnote of table 2:

The footnote now reads:

Line 263-264: “Note. SE = standard error; VG = Visually Guided; MG = Memory-Guided; KSS = Karolinska Sleepiness Scale. ***p < 0.001; ** p < 0.01; *p < 0.05.”

Author Response:

The pairwise comparisons are now shown in Table 4 of the estimated marginal means and trends, and indicated in Figures 3 and 4.

Author Response:

We have updated the results section to include statistics when results are mentioned. As an example:

Author Response:

The analysis has been updated so that the outcome measure is force as a percent of MVC instead of force in newtons. As a result, MVC was removed as an interaction term from the model. Therefore, all figures and analyses (including the figure indicated in this comment figure 5) are relative to force as a percent of MVC.

Discussion

Author Response:

The reviewer brings up a good point. We have added a limitations section before the conclusions paragraph, copied below:

Line 400-413: “This study had some limitations. First, the force task asked participants to maintain a grip force output of 25% of MVC. While this task is relatable to some activities of daily living, such as holding hot beverages, it may not be directly applicable to other activities requiring changes in grip force (such as operating heavy machinery) or gripping at higher or lower levels of force (such as holding a pencil or a large child). Future studies should investigate the effect of sleep restriction on various motor force tasks to better understand the wholistic effect of sleep restriction on movement. Also, we modeled force data without including a longitudinal effect (over the 12 s of gripping). As this was the first study exploring the effect of sleep restriction on grip force output, we were interested in the overall effect (or lack thereof) of sleep restriction and recovery sleep on motor output. Future studies may model continuous force production over time to determine if the effects of sleep restriction we found here change with duration of force production.”

References

Gelman, A., and Hill, J. (2008). Data Analysis Using Regression and Multilevel/Hierarchical Models. Cambridge, MA: Cambridge University Press.

Bates D. Fitting linear mixed-effects models using lme4. Journal of Statistical Software. 2015;67(1):1-48.

Neely, K. A., Chennavasin, A. P., Yoder, A., Williams, G. K., Loken, E., & Huang-Pollock, C. L. (2016). Memory-guided force output is associated with self-reported ADHD symptoms in young adults. Exp Brain Res, 234(11), 3203-3212. doi:10.1007/s00221-016-4718-1

Neely, K. A., Mohanty, S., Schmitt, L. M., Wang, Z., Sweeney, J. A., & Mosconi, M. W. (2016). Motor Memory Deficits Contribute to Motor Impairments in Autism Spectrum Disorder. J Autism Dev Disord. doi:10.1007/s10803-016-2806-5

Neely, K. A., Samimy, S., Blouch, S. L., Wang, P., Chennavasin, A., Diaz, M. T., & Dennis, N. A. (2017). Memory-guided force control in healthy younger and older adults. Experimental brain research, 235(8), 2473-2482.

Marvel, C. L., Morgan, O. P., & Kronemer, S. I. (2019). How the motor system integrates with working memory. Neurosci Biobehav Rev, 102, 184-194. doi:10.1016/j.neubiorev.2019.04.017

Lohse KR, Shen J, Kozlowski AJ. Modeling Longitudinal Outcomes: A Contrast of Two Methods. Journal of Motor Learning and Development. 2020;8(1):145-65.

Attachment

Submitted filename: response to reviewers_final.docx

Click here for additional data file.^{(42KB, docx)}

PLoS One. doi: 10.1371/journal.pone.0274121.r003

Decision Letter 1

Kenichi Shibuya

28 Jul 2022

PONE-D-22-02851R1Sleep Restriction Impairs Visually and Memory-Guided Force ControlPLOS ONE

Dear Dr. Brinkerhoff,

Please submit your revised manuscript by Sep 11 2022 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'.

We look forward to receiving your revised manuscript.

Kind regards,

Kenichi Shibuya, Ph.D.

Academic Editor

PLOS ONE

Additional Editor Comments :

I just received the comments of the reviewers. Two reviewers suggest to me "Accept". But Reviewer 2 suggests "Major Revision".

Reviewer 2 suggests the important issues for improving your manuscript. Please respond to the comments from Reviewer 2.

[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: All comments have been addressed

Reviewer #2: (No Response)

Reviewer #3: All comments have been addressed

**********

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

Reviewer #1: Yes

Reviewer #2: Partly

Reviewer #3: Yes

**********

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

Reviewer #1: I Don't Know

Reviewer #2: I Don't Know

Reviewer #3: Yes

**********

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

Reviewer #1: Yes

Reviewer #2: Yes

Reviewer #3: Yes

**********

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

Reviewer #1: Yes

Reviewer #2: Yes

Reviewer #3: (No Response)

**********

6. Review Comments to the Author

Reviewer #1: Overall I believe the authors have satisfactorily addressed my concerns. I am still unsure of the applicability of the statistical methodology (in particular treating individual timepoints as samples) and hope a domain expert can give further feedback here. I don't expect results will change if statistics were carried out differently.

Reviewer #2: The authors responded to most of the comments raised by the two reviewers well, but it still needs details and clarification about how they performed the statistical analysis and reported the results. The authors should provide more details about how the force data were analyzed to perform the statistical analysis.

Methods

#179-181: The authors did not fully explain the normalization procedure using the MVC values and how the force data were analyzed.

#183: Please provide readers with details for the levels of each factor sooner than later (e.g., two levels, no visual feedback, and visual feedback). Readers need to wait until seeing that information in lines 203~204.

#187: Please provide readers with details for what variables were used as dependent variables in the analysis (e.g., How was the performance of motor force production evaluated? mean of force data normalized by MVC values for each day?)

Results

#234: Please provide a quantitative information that explains the model selection (e.g., AIC values).

# Table 2. The authors stated that there are two factors in the Method, but there seemed to be another factor (KSS; e.g., #282). It is confusing which one is right. Could the authors clarify this issue?

Discussion

#402-403: Please provide references to support that the force level used in the current study matches the force levels required for performing some activities of daily living.

Reviewer #3: The statistical analysis of the paper is well done and described. I have no comments for the authors and I suggest accepting the paper in its current version.

**********

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

**********

Attachment

Submitted filename: comments_R1.docx

Click here for additional data file.^{(13.1KB, docx)}

PLoS One. 2022 Sep 2;17(9):e0274121. doi: 10.1371/journal.pone.0274121.r004

Author response to Decision Letter 1

15 Aug 2022

REVIEWER #1:

The authors responded to most of the comments raised by the two reviewers well, but it still needs details and clarification about how they performed the statistical analysis and reported the results. The authors should provide more details about how the force data were analyzed to perform the statistical analysis.

Methods

#179-181: The authors did not fully explain the normalization procedure using the MVC values and how the force data were analyzed.

Author Response:

We have updated the end of the “Motor Force task” section to include information about how the force data were normalized to MVC.

-Line 181-183: “Force data were collected in newtons and were divided by the participant’s MVC measured on the same day, multiplied by 100%. Therefore, the data were analyzed as a percent of MVC.”

Author Response:

We named the factor levels earlier in the statistical analysis section, where the reviewer indicated. We also named the explored covariates in the same sentence.

-Line 186-189: “Therefore, the analysis design was repeated measures with two main design factors — day (baseline, sleep restriction, recovery sleep) and vision condition (visually guided and memory-guided) — and four potential covariates (race, age, TST, and KSS score).”

Author Response:

We provided detailed on the exact dependent measure used in the statistical analyses.

-Line 190-193: “We used a mixed effects multilevel approach to analyze the effects of sleep restriction and visual feedback on mean force produced in the last 12 seconds of each trial, normalized by MVC, in addition to modeling how the effects of day and vision condition varied across individuals.”

-Line 198-200: “A series of models were estimated in a two-level multilevel framework using the lme4 package [36] in R [37] to model the mean force, normalized by MVC, across days and vision condition, nested within participants.”

Results

#234: Please provide a quantitative information that explains the model selection (e.g., AIC values).

Author Response:

We added AIC fits for the final model (allowing variation due to levels of repeated measures) and for the model that assumed no variation due to repeated measures factors.

-Line 186-189: “The model including random slopes for the day and vision condition (AIC = 1.06x10^6) fit the data better than the model assuming no variation due to day and vision across individuals (AIC = 1.08x10^6).

In total, 16 models were built-up and consecutively compared for fit to the data. The authors feel that listing all model fits would take space and detract from the results; therefore, all other AIC values (and other model fit values) are provided in supplementary tables.

Author Response:

While there were two design factors, we also included any covariate that significantly improved the model — KSS score was the only covariate that improved model fit to the data. This has been clarified in manuscript in the following lines.

-Line 244-246: “The best-fitting model, shown in Table 2, included main effects for day, vision condition, KSS score, and the three-way interaction among day, vision condition, and KSS score.”

Discussion

#402-403: Please provide references to support that the force level used in the current study matches the force levels required for performing some activities of daily living.

Author Response:

This sentence has been revised to be more general, and citations have been added.

-Line 402-406: “Grip force of 20-40% MVC is important for activities of daily living such as writing, unlocking a door, and tying shoes (Marshall & Armstrong, 2004), but it may not be directly applicable to other activities requiring higher or lower grip force (such as operating heavy machinery).”

References

Marshall, M. M., & Armstrong, T. J. (2004). Observational Assessment of Forceful Exertion and the Perceived Force Demands of Daily Activities. Journal of Occupational Rehabilitation, 14(4), 281–294. https://doi.org/10.1023/B:JOOR.0000047430.22740.57

Attachment

Submitted filename: response to reviewers_revisions2_v2.docx

Click here for additional data file.^{(26.4KB, docx)}

PLoS One. doi: 10.1371/journal.pone.0274121.r005

Decision Letter 2

Kenichi Shibuya

23 Aug 2022

Sleep Restriction Impairs Visually and Memory-Guided Force Control

PONE-D-22-02851R2

Dear Dr. Brinkerhoff,

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,

Kenichi Shibuya, Ph.D.

Academic Editor

PLOS ONE

Additional Editor Comments (optional):

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

Reviewer #2: All comments have been addressed

**********

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

Reviewer #2: Yes

**********

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

Reviewer #2: Yes

**********

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

Reviewer #2: Yes

**********

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

Reviewer #2: Yes

**********

6. Review Comments to the Author

Reviewer #2: The authors responded to the comments well. Although I am not an expert in statistics, this reviewer thinks that the details for statistical analysis that the author provided could enable readers to replicate their work, which may result in other follow-ups and exciting research on that topic.

**********

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 #2: No

**********

PLoS One. doi: 10.1371/journal.pone.0274121.r006

Acceptance letter

Kenichi Shibuya

25 Aug 2022

PONE-D-22-02851R2

Sleep Restriction Impairs Visually and Memory-Guided Force Control

Dear Dr. Brinkerhoff:

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. Kenichi Shibuya

Academic Editor

PLOS ONE

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

S1 Table. Random intercept models.

(DOC)

Click here for additional data file.^{(647.8KB, doc)}

S2 Table. Random intercept models with potential covariates.

(DOCX)

Click here for additional data file.^{(21.4KB, docx)}

S3 Table. Random intercept models with study design and covariates.

(DOCX)

Click here for additional data file.^{(18.8KB, docx)}

S4 Table. Random slopes models with study design and KSS.

(DOCX)

Click here for additional data file.^{(22.4KB, docx)}

S1 Data. Demographics, day, total sleep time, KSS, MVC, vision condition, trial, time in seconds, and force in newtons.

(CSV)

Click here for additional data file.^{(18.3MB, csv)}

Attachment

Submitted filename: response to reviewers_final.docx

Click here for additional data file.^{(42KB, docx)}

Attachment

Submitted filename: comments_R1.docx

Click here for additional data file.^{(13.1KB, docx)}

Attachment

Submitted filename: response to reviewers_revisions2_v2.docx

Click here for additional data file.^{(26.4KB, docx)}

Data Availability Statement

All relevant data are within the manuscript and its Supporting Information files.

[pone.0274121.ref001] 1.Watson NF, Badr MS, Belenky G, Bliwise DL, Buxton OM, Buysse D, et al. Joint Consensus Statement of the American Academy of Sleep Medicine and Sleep Research Society on the Recommended Amount of Sleep for a Healthy Adult: Methodology and Discussion. Sleep. 2015;38(8):1161–83. doi: 10.5665/sleep.4886 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0274121.ref002] 2.Centers for Disease Control and Prevention, National Center for Chronic Disease Prevention and Health Promotion, Division of Population Health. Sleep and Sleep Disorders: Data and Statistics 2017. [Available from: https://www.cdc.gov/sleep/data_statistics.html]. [Google Scholar]

[pone.0274121.ref003] 3.Belenky G, Wesensten NJ, Thorne DR, Thomas ML, Sing HC, Redmond DP, et al. Patterns of performance degradation and restoration during sleep restriction and subsequent recovery: a sleep dose‐response study. Journal of Sleep Research. 2003;12:1–12. doi: 10.1046/j.1365-2869.2003.00337.x [DOI] [PubMed] [Google Scholar]

[pone.0274121.ref004] 4.Dinges DF, Pack F, Williams K, Gillen KA, Powell JW, Ott GE, et al. Cumulative sleepiness, mood disturbance, and psychomotor vigilance performance decrements during a week of sleep restricted to 4–5 hours per night. Sleep. 1997;4:267–77. [PubMed] [Google Scholar]

[pone.0274121.ref005] 5.Lim J, Dinges DF. Sleep deprivation and vigilant attention. Ann N Y Acad Sci. 2008;1129:305–22. doi: 10.1196/annals.1417.002 [DOI] [PubMed] [Google Scholar]

[pone.0274121.ref006] 6.Lim J, Dinges DF. A meta-analysis of the impact of short-term sleep deprivation on cognitive variables. Psychol Bull. 2010;136(3):375–89. doi: 10.1037/a0018883 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0274121.ref007] 7.Mathew GM, Strayer SM, Bailey DS, Buzzell K, Ness KM, Schade MM, et al. Changes in Subjective Motivation and Effort During Sleep Restriction Moderate Interindividual Differences in Attentional Performance in Healthy Young Men. Nat Sci Sleep. 2021;13:1117–36. doi: 10.2147/NSS.S294409 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0274121.ref008] 8.Van Dongen HPA, Maislin G, Mullington JM, Dinges DF. The Cumulative Cost of Additional Wakefulness: Dose-Response Effects on Neurobehavioral Functions and Sleep Physiology From Chronic Sleep Restriction and Total Sleep Deprivation. Sleep. 2003;26(2). doi: 10.1093/sleep/26.2.117 [DOI] [PubMed] [Google Scholar]

[pone.0274121.ref009] 9.Drake CL, Roehrs TA, Burduvali E, Bonahoom A, Rosekind M, Roth T. Effects of rapid versus slow accumulation of eight hours of sleep loss. Psychophysiology. 2001;28:979–87. doi: 10.1111/1469-8986.3860979 [DOI] [PubMed] [Google Scholar]

[pone.0274121.ref010] 10.Santisteban JA, Brown TG, Ouimet MC, Gruber R. Cumulative mild partial sleep deprivation negatively impacts working memory capacity but not sustained attention, response inhibition, or decision making: a randomized controlled trial. Sleep Health. 2019;5(1):101–8. doi: 10.1016/j.sleh.2018.09.007 [DOI] [PubMed] [Google Scholar]

[pone.0274121.ref011] 11.Baddeley A. Working Memory. Science. 1992;255. [DOI] [PubMed] [Google Scholar]

[pone.0274121.ref012] 12.Adams JA. A closed-loop theory of motor learning. J Mot Behav. 1971;3(2):111–49. doi: 10.1080/00222895.1971.10734898 [DOI] [PubMed] [Google Scholar]

[pone.0274121.ref013] 13.Fitts PM. Perceptual-motor skill learning. Categories of human learning: Academic Press; 1964. p. 243–85. [Google Scholar]

[pone.0274121.ref014] 14.Gerver CR, Neely KA, Kurkela KA, Diaz MT, Goodman JT, Blouch S, et al. Shared neural recruitment across working memory and motor control tasks as a function of task difficulty and age. Neuropsychol Dev Cogn B Aging Neuropsychol Cogn. 2020;27(6):864–79. doi: 10.1080/13825585.2019.1700898 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0274121.ref015] 15.Marvel CL, Morgan OP, Kronemer SI. How the motor system integrates with working memory. Neurosci Biobehav Rev. 2019;102:184–94. doi: 10.1016/j.neubiorev.2019.04.017 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0274121.ref016] 16.Neely KA, Wang P, Chennavasin AP, Samimy S, Tucker J, Merida A, et al. Deficits in inhibitory force control in young adults with ADHD. Neuropsychologia. 2017;99:172–8. doi: 10.1016/j.neuropsychologia.2017.03.012 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0274121.ref017] 17.Neely KA, Samimy S, Blouch SL, Wang P, Chennavasin A, Diaz MT, et al. Memory-guided force control in healthy younger and older adults. Exp Brain Res. 2017;235(8):2473–82. doi: 10.1007/s00221-017-4987-3 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0274121.ref018] 18.Neely KA, Mohanty S, Schmitt LM, Wang Z, Sweeney JA, Mosconi MW. Motor Memory Deficits Contribute to Motor Impairments in Autism Spectrum Disorder. J Autism Dev Disord. 2019;49(7):2675–84. doi: 10.1007/s10803-016-2806-5 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0274121.ref019] 19.Doran SM, Dongen HPA, Dinges DF. Sustained attention performance during sleep deprivation evidence of state instability. Archives Italiennes de Biologie. 2001;139:253–67. [PubMed] [Google Scholar]

[pone.0274121.ref020] 20.Jung CM, Ronda JM, Czeisler CA, Wright KP Jr., Comparison of sustained attention assessed by auditory and visual psychomotor vigilance tasks prior to and during sleep deprivation. J Sleep Res. 2011;20(2):348–55. doi: 10.1111/j.1365-2869.2010.00877.x [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0274121.ref021] 21.Mahajan K, Velaga NR. Effects of Partial Sleep Deprivation on Braking Response of Drivers in Hazard Scenarios. Accid Anal Prev. 2020;142:105545. doi: 10.1016/j.aap.2020.105545 [DOI] [PubMed] [Google Scholar]

[pone.0274121.ref022] 22.Mahajan K, Velaga NR. Sleep-deprived car-following: Indicators of rear-end crash potential. Accid Anal Prev. 2021;156:106123. doi: 10.1016/j.aap.2021.106123 [DOI] [PubMed] [Google Scholar]

[pone.0274121.ref023] 23.Baker FC, Driver HS. Circadian rhythms, sleep, and the menstrual cycle. Sleep Med. 2007;8(6):613–22. [DOI] [PubMed] [Google Scholar]

[pone.0274121.ref024] 24.Mander BA, Winer JR, Walker MP. Sleep and Human Aging. Neuron. 2017;94(1):19–36. doi: 10.1016/j.neuron.2017.02.004 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0274121.ref025] 25.Douglass AB, Bornstein R, German N-M, Keenan S, Laughton M, Zarcone VP, et al. The Sleep Disorders Questionnaire I: Creation and Multivariate Structure of SDQ. Sleep. 1994;17(2):160–7. [DOI] [PubMed] [Google Scholar]

[pone.0274121.ref026] 26.Ness KM, Strayer SM, Nahmod NG, Chang AM, Buxton OM, Shearer GC. Two nights of recovery sleep restores the dynamic lipemic response, but not the reduction of insulin sensitivity, induced by five nights of sleep restriction. Am J Physiol Regul Integr Comp Physiol. 2019;316(6):R697–R703. doi: 10.1152/ajpregu.00336.2018 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0274121.ref027] 27.Figueiro MG, Bierman A, Plitnick B, Rea MS. Preliminary evidence that both blue and red light can induce alertness at night. BMC Neurosci. 2009;10:105. doi: 10.1186/1471-2202-10-105 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0274121.ref028] 28.Chang AM, Aeschbach D, Duffy JF, Czeisler CA. Evening use of light-emitting eReaders negatively affects sleep, circadian timing, and next-morning alertness. Proc Natl Acad Sci U S A. 2015;112(4):1232–7. doi: 10.1073/pnas.1418490112 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0274121.ref029] 29.Marino M, Li Y, Rueschman MN, Winkelman JW, Ellenbogen JM, Solet JM, et al. Measuring sleep: accuracy, sensitivity, and specificity of wrist actigraphy compared to polysomnography. Sleep. 2013;36(11):1747–55. doi: 10.5665/sleep.3142 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0274121.ref030] 30.Åkerstedt T, Gillberg M. Subjective and Objective Sleepiness in the Active Individual. International Journal of Neuroscience. 1990;52:29–37. doi: 10.3109/00207459008994241 [DOI] [PubMed] [Google Scholar]

[pone.0274121.ref031] 31.Harris PA, Taylor R, Thielke R, Payne J, Gonzalez N, Conde JG. Research electronic data capture (REDCap)—a metadata-driven methodology and workflow process for providing translational research informatics support. J Biomed Inform. 2009;42(2):377–81. doi: 10.1016/j.jbi.2008.08.010 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0274121.ref032] 32.Wang Z, Magnon GC, White SP, Greene RK, Vaillancourt DE, Mosconi MW. Individuals with autism spectrum disorder show abnormalities during initial and subsequent phases of precision gripping. J Neurophysiol. 2015;113(7):1989–2001. doi: 10.1152/jn.00661.2014 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0274121.ref033] 33.Neely KA, Chennavasin AP, Yoder A, Williams GK, Loken E, Huang-Pollock CL. Memory-guided force output is associated with self-reported ADHD symptoms in young adults. Exp Brain Res. 2016;234(11):3203–12. doi: 10.1007/s00221-016-4718-1 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0274121.ref034] 34.Vaillancourt DE, Slifkin AB, Newell KM. Visual control of isometric force in Parkinson’s disease. Neuropsychologia. 2001;39:1410–8. doi: 10.1016/s0028-3932(01)00061-6 [DOI] [PubMed] [Google Scholar]

[pone.0274121.ref035] 35.Gelman A. Multilevel (Hierarchical) Modeling: What It Can and Cannot Do. Technometrics. 2006;48(3):432–5. [Google Scholar]

[pone.0274121.ref036] 36.Bates D. Fitting linear mixed-effects models using lme4. Journal of Statistical Software. 2015;67(1):1–48. [Google Scholar]

[pone.0274121.ref037] 37.R Core team. R: A language and environment for statistical computing Vienna, Austria: R Foundation for Statistical Computing; 2019. [Available from: https://www.R-project.org/. [Google Scholar]

[pone.0274121.ref038] 38.Lohse KR, Shen J, Kozlowski AJ. Modeling Longitudinal Outcomes: A Contrast of Two Methods. Journal of Motor Learning and Development. 2020;8(1):145–65. [Google Scholar]

[pone.0274121.ref039] 39.Satterthwaite FE. An Approximate Distribution of Estimates of Variance Components. Biometrics Bulletin. 1946;2(6):110–4. [PubMed] [Google Scholar]

[pone.0274121.ref040] 40.Neely KA, Planetta PJ, Prodoehl J, Corcos DM, Comella CL, Goetz CG, et al. Force control deficits in individuals with Parkinson’s disease, multiple systems atrophy, and progressive supranuclear palsy. PLoS One. 2013;8(3):e58403. doi: 10.1371/journal.pone.0058403 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0274121.ref041] 41.Heaton KJ, Maule AL, Maruta J, Kryskow EM, Ghajar J. Attention and visual tracking degradation during acute sleep deprivation in a military sample. Aviat Space Environ Med. 2014;85(5):497–503. doi: 10.3357/asem.3882.2014 [DOI] [PubMed] [Google Scholar]

[pone.0274121.ref042] 42.Marshall Matthew M., Armstrong TJ. Observational Assessment of Forceful Exertion and the Perceived Force Demands of Daily Activities. Journal of Occupational Rehabilitation. 2004;14(4). doi: 10.1023/b:joor.0000047430.22740.57 [DOI] [PubMed] [Google Scholar]

PERMALINK

Sleep restriction impairs visually and memory-guided force control

Sarah A Brinkerhoff

Gina M Mathew

William M Murrah

Anne-Marie Chang

Jaimie A Roper

Kristina A Neely

Roles

Abstract

Introduction

Methods

Ethics statement

Participants

Procedures

Fig 1. Experimental timeline adapted with permission from Ness et. al., 2019.

Measurement of sleep

Sleepiness scores

Motor force tasks

Fig 2. Visually guided and memory-guided force paradigm.

Statistical analysis

Results

Participants

Table 1. Participant characteristics.

Grip force

Table 2. Results of the multilevel mixed-effects model estimating mean force (%MVC).

Fig 3. Force production as a percentage of maximum voluntary contraction (MVC) plotted by day and vision condition.

Table 3. Estimated marginal means and estimated marginal trends for force (% MVC).

Fig 4. Force production by day and vision condition as a percentage of maximum voluntary contraction (MVC), as related to Karolinska Sleepiness Scale (KSS) score.

Fig 5. Residuals of the final model across day and vision condition, demonstrating that residual variability in force production was higher in the memory-guided than the visually guided condition after sleep restriction and recovery sleep across participants.

Table 4. Standard deviation of the residuals of force production (% MVC) for experimental conditions, with and without Participant 12.

Discussion

Limitations

Conclusion

Supporting information

Acknowledgments

Data Availability

Funding Statement

References

Decision Letter 0

Kenichi Shibuya

Roles

Author response to Decision Letter 0

Decision Letter 1

Kenichi Shibuya

Roles

Author response to Decision Letter 1

Decision Letter 2

Kenichi Shibuya

Roles

Acceptance letter

Kenichi Shibuya

Roles

Associated Data

Supplementary Materials

Data Availability Statement

ACTIONS

PERMALINK

RESOURCES

Similar articles

Cited by other articles

Links to NCBI Databases