Reducing task difficulty during practice improves motor learning in older adults

EXG-09438; No of Pages 7 Experimental Gerontology xxx (2014) xxx–xxx

Contents lists available at ScienceDirect

Experimental Gerontology journal homepage: www.elsevier.com/locate/expgero

3Q1

Tanya Onushko, Changki Kim, Evangelos A. Christou ⁎

4

Department of Applied Physiology and Kinesiology, University of Florida, USA

5

a r t i c l e

6 7 8 9 10 11 12

Article history: Received 30 January 2014 Received in revised form 21 May 2014 Accepted 5 June 2014 Available online xxxx

13 14 15 16 17

Keywords: Aging Task difficulty Motor learning Practice

O R O

i n f o

F

2

Reducing task difficulty during practice improves motor learning in older adults

a b s t r a c t

P

Theoretically, greater motor output variability can inhibit motor learning by inhibiting task acquisition during practice. Although the age-associated differences in motor output variability exacerbate with more difficult tasks, it remains unknown whether task difficulty during task acquisition influences motor learning in older adults. The purpose of this study was to determine whether the difficulty of the practice task affects motor learning in older adults. Twenty four older (72.7 ± 7.4 years; 11 women) and 7 young (23.1 ± 4 years; 1 man) adults participated in this study. Participants were divided into four groups: 8 older adults who practiced an easy task (O-Easy), 8 older adults who practiced a harder task (O-Hard), 8 older adults who did not practice (O-None), and 7 young adults who did not practice (Y-None). The level of difficulty depended on the relative timing (i.e. phase) of abduction force generation between the index and little fingers to track a moving target on the monitor. The O-Easy group practiced the task with 0°, whereas the O-Hard group practiced the task with 90° relative phase. Practice occurred within a single session for 80 trials. Motor learning was quantified as the ability to transfer the practiced tasks to 45°, 135° and 180° relative phases 24 and 168 h after acquisition. Only the O-Easy group was able to significantly transfer the practiced task, as it was indicated by significantly lower force variability and error during all transfer tasks compared with the O-None group (P b 0.05). The O-Hard group was not significantly different from the O-None group (P N 0.2). In addition, during the transfer tasks the O-Easy group exhibited performance similar to that of the young adults who did not practice. These findings suggest that practice with easier tasks may be advantageous to practice with more difficult tasks to improve motor learning in older adults. © 2014 Published by Elsevier Inc.

T

E

D

Section Editor: Christiaan Leeuwenburgh

E

C

1

35

R

39 37 36

R

38

1. Introduction

41

Older adults need to adapt to various motor behaviors and consequently learn new motor skills while experiencing numerous changes to their neuromuscular system. A result of neuromuscular system changes in older adults is increased motor output variability (Enoka et al., 2003), which has been theoretically proposed as an antagonist to motor learning (Christou, 2011). To date, no training protocols exist to improve motor learning in older adults. In this study we examined whether motor learning could be improved in older adults by using a strategy that reduces motor output variability during practice. Motor learning is often quantified as the ability to transfer the learned information to new conditions (Schmidt, 1988). Motor learning in older adults, however, is thought to be impaired because they are often unsuccessful during transfer tasks (Bo et al., 2009; Hinder et al., 2011; Osu et al., 2002; Panzer et al., 2011; Seidler, 2006). For example,

46 47 48 49 50 51 52 53 54

U

44 45

N C O

Q4 40

42 43

Q2 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34

⁎ Corresponding author at: Department of Applied Physiology and Kinesiology, University of Florida, Gainesville, FL 32611-8205, USA. E-mail address: [email protected]fl.edu (E.A. Christou).

older adults exhibit an impaired ability to transfer sequential movements with the elbow to the contralateral arm (Bo et al., 2009), and do not transfer ballistic movements from the right to the left index finger as well as young adults (Hinder et al., 2011). When learning a new motor task, extraction and acquisition of task-related information through practice are essential for motor learning (Wolpert et al., 2011). The impaired motor learning in older adults may be related to difficulties extracting task-related information during practice because of increased motor output variability. Traditional training protocols, such as aerobic and resistance exercise are commonly employed to combat cognitive and physical health declines in older adults (Fleg, 2012; Mayer et al., 2011). However, they do not minimize motor output variability nor have they been shown to improve motor learning in older adults (Christou, 2011; Kamen, 2005). It is clear from the literature that greater task difficulty exacerbates motor output variability in older adults. This is evident from dual-tasks and increasing the number of effectors that need to be coordinated to perform the task. For example, when young and older adults perform a motor task concurrently with a cognitive task (dual tasking), force variability was greater in older adults (Voelcker-Rehage and Alberts, 2007). Similar results are demonstrated when subjects

http://dx.doi.org/10.1016/j.exger.2014.06.006 0531-5565/© 2014 Published by Elsevier Inc.

Please cite this article as: Onushko, T., et al., Reducing task difficulty during practice improves motor learning in older adults, Exp. Gerontol. (2014), http://dx.doi.org/10.1016/j.exger.2014.06.006

55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75

T. Onushko et al. / Experimental Gerontology xxx (2014) xxx–xxx

99 100

Twenty four older (72.7 ± 7.4 years; 11 women) and 7 young (23.1 ± 4 years; 1 man) adults participated in the study. All subjects were right handed according to a standardized survey (Oldfield, 1971) and reported being moderately active and healthy without any orthopedic or neurological disorders. All procedures for this study were approved by the University of Florida's Institutional Review Board, and subjects gave written informed consent before participation in this study.

101

2.2. Experimental protocol and procedures

102 103

Older adults were randomly assigned into one of three groups (8 subjects per group): easy task practice (O-Easy; 72.4 ± 6.2 years), hard task practice (O-Hard; 73.7 ± 8.8 years), or no practice (O-None; 72.3 ± 8.6 years). Young subjects also received no practice (Y-None). Subjects assigned to the O-Easy or O-Hard group participated in three experimental sessions while subjects who did not practice (Y-None and O-None) participated in only two sessions (Fig. 1). During the first session, subjects in the O-Easy or O-Hard group practiced their assigned task for 80 repetitions (16 trials per block × 5 blocks; each trial lasted 24 s), with 10 second rest given between trials and 5 minute rest given between blocks. Subjects then revisited the lab 24 and 168 h after practicing the task for a second and third session, respectively (Fig. 1). During each of these sessions, O-Easy and O-Hard groups performed the same task assigned for practice (retention task) and three similar but new tasks (transfer tasks). The tasks were randomly assigned for sessions 2 and 3. The O-None and Y-None groups participated in experimental sessions 2 and 3 (separated by 168 h) and performed the same 5 tasks as the practice groups. Twenty-five trials were performed for sessions 2 and 3. For all groups, subjects were given 90 s of familiarization to the experimental tasks before beginning

110 111 112 113 114 115 116 117 118 119 120 121

C

E

108 109

R

106 107

R

104 105

2.3.2. Force measurement The isometric force produced by abduction of the index finger and the little finger were recorded with single-axis force transducers (one per finger; Futek LRF400, Futek Advanced Sensor Technology Inc., CA, USA). The axis of the force transducer was aligned perpendicular to the proximal inter-phalangeal (PIP) joint of each finger (Fig. 2). The force signals were sampled at 1 kHz (USB-6210, National Instruments, Austin, TX, USA), low pass filtered at 30 Hz using Matlab (MathWorks™ Inc., Natick, Massachusetts, USA) and stored on a personal computer.

141 142

2.3.3. MVC task Maximal voluntary isometric contractions (MVCs) were measured from each finger individually before the start of the experimental tasks during the first visit (session 1 for the O-Hard and O-Easy groups, and session 2 for the O-None and Y-None groups). Subjects were instructed to exert the maximal abduction force of each finger by first increasing the finger's force from baseline to their maximum over a 2-s period and then maintaining the maximal force for about 4–7 s. These recordings were made until two of the maximal forces produced were within 5% of each other. The MVC force was quantified as the

O

97 98

C

95 96

N

89 90

U

87 88

126

F

94

85 86

2.3.1. Experimental setup and apparatus Subjects were seated comfortably in an upright position and faced a 30-inch monitor (Apple A1083 30-inch Cinema HD LCD Monitor) located 1.2 m away at eye level. The monitor was used to display the force produced by the abduction of the index and little fingers through a custom-written Matlab program (MathWorks™ Inc., Natick, MA, USA). The left arm was placed within a custom-built device securing the arm to a table, with the shoulder abducted 45° and the elbow positioned in ~90° flexion (refer to Fig. 2). The left forearm was pronated ~80° and secured to the custom-built device using a Velcro strap. The thumb, middle and ring fingers of the left hand and forearm were restrained from movement (Fig. 2). Only the index and the little fingers of the left hand were free to move. This arrangement allowed abduction of the index and little fingers about the metacarpophalangeal joint in the horizontal plane.

O

2.1. Participants

83 84

125

R O

93

82

2.3. Experiment arrangement

P

2. Methods

80 81

D

92

78 79

each session. Subjects were allowed additional practice trials until they 122 understood the task. The practice trials were not the same as the exper- 123 imental trials. 124

T

91

must increase the number of fingers to perform a dexterous motor task (Olafsdottir et al., 2007; Shinohara et al., 2003). Greater task difficulty can also increase physiological stress, which has been shown to exacerbate the age-associated differences in motor output variability (Christou, 2005; Christou et al., 2004). Despite the above evidence, an important but unresolved question is: can interventions that incorporate less challenging (easier) tasks, and thus lower variability, improve motor learning in older adults? In this study, we tested whether motor output variability affects motor learning in older adults, as evidenced by motor transfer. Specifically, the purpose of this study was to compare the ability of older adults to transfer a motor task following practice with an easy (low variability) and more difficult task (high variability). We hypothesized that practice with an easier motor task will result in better motor transfer because of lower motor output variability during task acquisition. Part of these data has been presented in abstract form (Onushko et al., 2012).

76 77

E

2

Fig. 1. Experimental timeline. Older subjects assigned in a practice group (O-Easy and O-Hard) performed their assigned task for 80 trials during session 1. One group of young and one group of older adults (Y-None and O-None) received no practice. All subjects participated in session 2 and session 3 to perform the transfer and retention tasks.

Fig. 2. Experimental setup. The left index and little fingers were aligned perpendicular to single-axis force transducers. The forearm and middle fingers were secured from movement using Velcro straps.

Please cite this article as: Onushko, T., et al., Reducing task difficulty during practice improves motor learning in older adults, Exp. Gerontol. (2014), http://dx.doi.org/10.1016/j.exger.2014.06.006

127 128 129 130 131 Q5 132 133 134 135 136 137 138 139 140

143 144 Q6 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159

T. Onushko et al. / Experimental Gerontology xxx (2014) xxx–xxx

182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202

2.4.2. Motor learning To investigate whether task difficulty affected motor learning, we examined savings, retention and transfer of force errors, force variability and timing errors for the O-Easy and O-Hard groups. Savings and retention were calculated as the percent change between session 1 and session 2, and between session 1 and session 3. The percent change for savings was calculated as: 100 × (([value from later session] − [session 1 block 1]) / [session 1 block 1]). The percent change for retention was calculated as: 100 × (([value from later session] − [session 1 block 5]) / [session 1 block 5]). Transfer was examined as the absolute values for 45°, 135° and 180° relative phases. The three relative phases were averaged for each subject within each session because we found no statistical significance among phases (P N 0.05) for Total Variance, Total RMSE, and Total Delay.

226 227

2.5. Statistical analysis

240

O

F

204 205

R O

180 181

2.4.1. Force measurements The index and little finger forces were used to characterize force variability, tracking accuracy and timing errors during the practice, retention and transfer tasks. The index and little finger force data were analyzed off-line using custom-written programs in Matlab (MathWorks™ Inc., Natick, MA, USA). Prior to data analysis, the force signal was low-pass filtered with a 4th order (bi-directional) Butterworth filter with a 30 Hz cut-off frequency. For analysis, the first 3 s and last 3 s of each 24-second trial were removed to omit data when subjects were starting or ending the force generation. The force signal from each finger was normalized to 20% of the respective finger's MVC. Errors in tracking accuracy were quantified as the root mean square error (RMSE) between the sinusoidal trace and force for each finger. The RMSE values from the index and little fingers were then summed (Total RMSE). Force variability was quantified as the sum of the variance of each finger's force trace (Total Variance). Before calculating the variance, the force data were high-pass (0.5 Hz) filtered (4th order Butterworth filter) to remove the tracking frequency. Error in timing was quantified as the absolute error between the peak of the subject's force and the peak of the sinusoidal trace. The timing errors were calculated separately for each finger and then summed together (Total Delay).

P

178 179

203

D

176 177

2.4. Data analysis

E

174 175

T

172 173

C

170 171

E

168 169

2.3.4. Coordination task Subjects were instructed to track a moving target on the monitor by coordinating isometric abduction forces of the index and little fingers. A two-dimensional (X–Y) coordinate plane was displayed on the monitor. The X-axis represented abduction force of the index finger (0–20% MVC) and the Y-axis represented abduction force of the little finger (0–20% MVC). The target was represented by a red-filled circle. The target traveled in a predetermined path that was composed of two sinusoidal waveforms (Fig. 3A) generated by the following equations: x(t) = A sin(at + δ) and y(t) = B sin(at), where A and B are 10% MVC of the index and little fingers, respectively, t is the task time, a is the frequency (0.125 Hz), and δ is the phase. Subjects were given force feedback from the index and little fingers (Fig. 3B). The force feedback was displayed as a single cursor (a blue circle) on the monitor using Lissajous feedback. The index finger force moved the cursor's position horizontally (X-axis) and the little finger force moved the cursor's position vertically (Y-axis). A custom-written Matlab program controlled the feedback and the target's trajectory. The level of task difficulty depended on the relative phase (δ) of abduction force between the index and little fingers (Fig. 3A; Haken et al., 1985). The more difficult (harder) task δ = 90°, in which subjects needed to produce force in the index finger such that it followed the little finger force by a quarter cycle. The less difficult task (easier) δ = 0°, in which subjects must produce abduction force from the index and little fingers simultaneously. The target's path for δ = 90° was a circle and the path for δ = 0° was a straight line (Fig. 3B). The transfer tasks consisted of 3 different relative phases: δ = 45°, δ = 135° and δ = 180° (Fig. 3C). In order to correctly track the moving target, subjects had to control the abduction force of each finger in a sinusoidal pattern with amplitude between 5 and 15% MVC of the respective finger. During the first session, all subjects were given instructions about the feedback. Subjects were then asked questions to test their understanding of the task and were given a few (1–3) instructive practice trials (the relative phase used during the instructive practice trials was different from the experimental practice and transfer tests). Subjects could begin the experimental trials once they successfully demonstrated that they understood the task requirements.

R

166 167

R

164 165

N C O

162 163

average force over 3–6 s (constant part) of the trial with the greatest force produced. This procedure allows for the identification of a more conservative MVC that reflects the capability of the person to perform constant isometric contractions. At the completion of the session, all subjects performed an additional MVC of the index and little fingers to test for fatigue.

206 207 208 209 210 211 212 213 214 215 216 217 Q7 218 219 220 221 222 223 224 225

228 229 230 231 232 Q8 233 234 Q9 235 236 237 238 239

A mixed two-way ANOVA (2 practice conditions [O-Easy vs. 241 O-Hard] × 5 blocks) with repeated measures on blocks was used to 242

U

160 161

3

Fig. 3. Task difficulty and visual feedback. (A) Task difficulty was dependent on the relative phase between force generation of the index finger (solid line) and little finger (dotted line). (B) A Lissajous plot was provided to subjects for visual feedback. The target's path (dashed line) was a line for the 0° relative phase and a circle for the 90° relative phase. The filled-circle represents the target subjects were asked to follow. The open circle represents the subject's force feedback. (C) Lissajous plots for the transfer tasks.

Please cite this article as: Onushko, T., et al., Reducing task difficulty during practice improves motor learning in older adults, Exp. Gerontol. (2014), http://dx.doi.org/10.1016/j.exger.2014.06.006

252 253 254 255 256 257 258 259 260

3. Results

262

3.1. Strength and fatigue

263

To test whether our protocol induced muscle fatigue to the subjects, we compared the MVC before and immediately after practice. Subjects in the O-Easy and O-Hard groups exhibited similar MVC before (O-Easy Index: 21.8 ± 10.7 N; O-Easy Little: 10.1 ± 4.3 N; O-Hard Index: 21.2 ± 7.6 N; O-Hard Little: 11.3 ± 5.0 N) and after (O-Easy Index: 19.8 ± 12.3 N; O-Easy Little: 11.4 ± 5.7 N; O-Hard Index: 22.7 ± 10.1 N; O-Hard Little: 11.6 ± 5.2 N; O-Easy Index: t(7) = 1.756, P = 0.122, d = 0.62; O-Easy Little: t(7) = − 0.82, P = 0.439, d = 0.29; O-Hard Index: t(7) = − 1.349, P = 0.219, d = 0.48; O-Hard Little: t(7) = − 0.691, P = 0.512, d = 0.24). Neither protocol induced fatigue to the subjects. Additionally, there were no significant differences in strength between the O-Easy and O-Hard groups (Index:

C

E R R

274

O

272 273

C

270 271

N

268 269

U

266 267

278

3.3. Effects of task difficulty on retention

298

We tested retention (repetition of practice task) and savings (change from block 1) of the practiced task 24 and 168 h after practice. There were no significant differences in retention and savings for older adults who practice with the easy and hard tasks (Mann–Whitney Test, P N 0.05).

299 300

3.4. Effects of task difficulty on transfer

304

We examined the effects of task difficulty on motor learning by comparing force variability, force error and timing errors during the transfer tasks performed 24 h after task acquisition. The main effect of practice group was significant for Total Variance (F3,25 = 4.26, P = 0.015, η2 = 0.34), Total RMSE (F3,26 = 5.18, P = 0.006, η2 = 0.37) and Total Delay (F3,26 = 3.07, P = 0.045, η2 = 0.26). The post-hoc tests revealed that the older adults who practiced with the easier task exhibited significantly lower force variability (P = 0.039; Fig. 6A), force error (P =0.039; Fig. 6B) and timing error (P = 0.070; Fig. 6C) compared with older adults who did not train (O-None). Additionally, these subjects also exhibited significantly lower force variability compared with subjects who practiced with the harder task (P = 0.028; Fig. 6A). Interestingly, subjects who practiced with the easier task performed at a similar level as the young controls (Total Variance, P = 0.509; Total RMSE, P = 0.130; Total Delay, P = 0.380; Fig. 6). In contrast, older adults who practiced with the harder task (O-Hard) and those who received no practice (O-None) demonstrated significantly greater force variability, force error and timing errors compared with the young control group (P b 0.05; Fig. 6). We also examined the effects of task difficulty on motor transfer 168 h after task acquisition. A univariate ANOVA detected a practice condition effect for Total Variability (F3,25 = 4.75, P = 0.009, η2 = 0.35), Total RMSE (F3,26 = 6.068, P = 0.003, η2 = 0.41) and Total Delay (F3,25 = 4.719, P = 0.01, η2 = 0.36). The post-hoc tests showed that the O-Easy group maintained lower force variability compared with the O-Hard group (P = 0.039) and no difference between the Y-None group (P = 0.195; Fig. 7). However, both practice protocols resulted in similar force errors (P = 0.292) and timing errors (P = 0.989) for O-Hard and O-Easy groups.

305 306

T

261

264 265

We examined adaptations in force variability, force error, and timing error between subjects who practiced either with the easier or with the harder two-finger coordination task. Fig. 4 illustrates representative data from one subject who practiced with the easier task (top row) and one subject who practiced with the harder task (bottom row) during an early (block 1) and later (block 5) trial. Across all trials, the O-Easy group performed the practice task with significantly lower variability compared with the O-Hard group (P = 0.036; Fig. 5B). Although errors in force were exacerbated in the O-Hard group compared with the OEasy group, there was no statistically significant difference between groups for the Total RMSE (O-Easy: 38.5 ± 15.4 N; O-Hard: 48.4 ± 14.2 N; practice condition main effect: F1,14 = 1.80, P = 0.201, η2 = 0.114) or the Total Delay (O-Easy: 2.3 ± 3.0 s; O-Hard: 4.9 ± 3.6 s; practice condition main effect: F1,14 = 2.13, P = 0.166, η2 = 0.132). However, there was a block main effect for Total RMSE (block 1 N blocks 2, 4; F4,56 = 5.65, P = 0.001, η2 = 0.29) and Total Variance (block 1 N block 5; F4,56 = 6.98, P = 0.004, η2 = 0.33; Fig. 5A). This result demonstrates that both groups decreased force error and force variability with practice. There was no significant difference in timing errors among the blocks (F4,56 = 1.412, P = 0.259, η2 = 0.092).

F

250 251

277

R O

249

3.2. Effects of task difficulty on performance during practice

P

247 248

t(14) = 1.54, P = 0.88, d = 0.15; Little: t(14) = − 0.485, P = 275 0.635, d = 0.48). 276

D

245 246

compare force errors (Total RMSE), force variability (Total Variance), and timing errors (Total Delay) during task acquisition. A univariate ANOVA (4 groups [O-Easy, O-Hard, O-None, and Y-None]) was used to compare force accuracy, force variability and timing errors for the transfer tasks during the transfer tasks. Separate ANOVAs were performed on the transfer tasks for the 24 and 168 hour tests. Post-hoc comparisons were performed using the Least Significant Difference. The retention and savings quantification violated the assumption of a normal distribution. Hence, we used a non-parametric test (Mann–Whitney Test) to compare force accuracy, force variability and time errors. Statistical analyses were performed using IBM Statistics 21.0 software (SPSS, Inc., Chicago, IL). The alpha level for all statistical tests was 0.05. Data are reported as means ± SD within the text and as means ± SE in the figures. Only significant practice main effects and interactions are presented, unless otherwise noted. Outliers were removed from statistical analyses if the data exceeded the mean ± 2 SDs. Less than 6% of data (1–2 points) were removed from the Total RMSE, Total Delay or Total Variance data from the transfer tasks.

E

243 244

T. Onushko et al. / Experimental Gerontology xxx (2014) xxx–xxx

O

4

Fig. 4. Representative force data. A single trial for one subject in the easy task practice (O-Easy; top row) and the hard task practice (O-Hard; bottom row) group. Older adults who practiced with the easier task exhibited lower force variability and errors.

Please cite this article as: Onushko, T., et al., Reducing task difficulty during practice improves motor learning in older adults, Exp. Gerontol. (2014), http://dx.doi.org/10.1016/j.exger.2014.06.006

279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 Q10 295 296 297

301 302 303

307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 Q11 329 330 331 332 333 Q12

5

O

F

T. Onushko et al. / Experimental Gerontology xxx (2014) xxx–xxx

345

4.1. Motor output variability and motor learning

346 347

During task acquisition, effective gathering of task-relevant information is critical for learning a new motor skill (Wolpert et al., 2011). However, the ability of older adults to gather task-relevant information during acquisition may be limited because of greater motor output variability (Baweja et al., 2012; Kennedy and Christou, 2011). This heightened motor output variability may distort the feedback that they receive during practice of a motor task and reduce the task-relevant information needed to learn the task. In the current study, older adults who practiced with the easier task exhibited less variability during task acquisition compared with older adults who practiced with the difficult task (see Fig. 5B). While we did not find any differences in the retention or savings between practice groups, we did find that older adults in the easy group were able to perform the transfer tasks better than the older adults who practiced with the difficult/more variable task (see Fig. 6). Thus, practicing the task with lower variability could have strengthened their use of incoming sensory information and consequently improved their ability to learn the task.

350 351 352 353 354 355 356 357 358 359 360 361 362

C

E

348 349

R

342 343

R

340 341

N C O

338 339

U

336 337

P

344

The purpose of this study was to compare the ability of older adults to transfer a motor task following practice with either an easy or a more difficult task. We found that practice with an easy task enhanced motor learning in older adults. Older adults in the easy task practice group performed the transfer tasks better than the older control group but similar to the young control group. In contrast, older adults in the difficult task practice group performed the transfer tasks similar to the older control group but poorer compared with the young control group. These results suggest that reducing motor output variability during practice can improve motor learning in older adults.

D

335

Traditionally, retention tasks are used to assess relatively permanent changes following practice of a new skill (Schmidt, 1988). In the current study, we found no differences between practice groups during the retention tasks. This may be explained in two ways: 1) Retention tasks can be unreliable measures of motor learning especially when comparing two groups that practice under different conditions. In the current study, the O-Easy and O-Hard groups start at different levels of performance during acquisition, and, therefore, it is unclear how much each group is truly learning. 2) The group practicing under the 0° relative phase coordination may have demonstrated a “ceiling” effect. The 0° relative phase task may have been intuitive and thus, the O-Easy group may have reached a limit where little learning occurred. In contrast, the 90° relative phase task may have been too difficult for older adults, and possibly required more practice time for them to show improvements in performance. Regardless of the reason, the main finding is still interesting and important to the motor learning literature. When older adults practice a novel motor task, even with a single session of practice, their ability to transfer the practiced task is better when they practice with an easier task than with a more difficult task. This result supports the theoretical framework by Wolpert et al. (2011).

E

4. Discussion

T

334

R O

Fig. 5. Task acquisition performance. (A) Total Variance from the easy task practice (O-Easy) and hard task practice (O-Hard) groups during Acquisition and Transfer. Both O-Easy and O-Hard groups lowered variability following the first block of practice trials (P b 0.05). (B) Overall, the Total Variance for O-Easy was significantly (P = 0.036) lower than that for the O-Hard group during acquisition.

363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382

4.2. Why does the easy task allow older adults to practice with lower 383 variability? 384 One possibility for observing greater variability in older adults during the more difficult task may be related to the coordination timing of the task itself. Based on theories from dynamical models of movement coordination, in-phase (0°) and antiphase (180°) relative phases constitute stable, natural coordination tendencies (Beek et al., 1995). Subjects can adapt to these relative phases with low-energy costs (Sparrow et al., 2005). This is in line with the 0° practice data for the older subjects where we observed lower variability. However, the 90°

Fig. 6. Transfer (24 h). Group comparisons for the Total Variance (A), Total RMSE (B) and Total Delay (C) during the transfer tasks for the young no practice (Y-None), older easy task practice (O-Easy), older hard task practice (O-Hard) and older no practice (O-None) groups. Asterisks indicate significant differences (P b 0.05).

Please cite this article as: Onushko, T., et al., Reducing task difficulty during practice improves motor learning in older adults, Exp. Gerontol. (2014), http://dx.doi.org/10.1016/j.exger.2014.06.006

385 386 387 388 389 390 391 392

T. Onushko et al. / Experimental Gerontology xxx (2014) xxx–xxx

could transfer to more difficult tasks (45°, 135° and 180° relative phases). In summary, our findings suggest that practice with easier tasks may be advantageous to practice with more difficult tasks to improve motor learning in older adults. The underlying mechanism of improvements in motor learning with practice using an easy task may be lower motor output variability during task acquisition. Practicing with low variability may improve motor learning in older adults by reducing cognitive resources and stress, and allow extraction of important task-relevant information. The findings from this study may have significant implications to the design of rehabilitation programs in older adults.

434 435

Conflict of interest

445

F

6

R O

Uncited references Chen et al., 2012 Ranganathan et al., 2001

Fig. 7. Transfer (168 h). Older adults who practiced with the easier task retained lower force variability during the transfer tasks over the 2 transfer sessions. Asterisks represent significant differences (P b 0.05).

C

E

R

R

O

C

N

U

424

4.3. Consideration

These findings are based on a single session of practice and they may be different if subjects had more time to practice. It has been shown that 427 young adults performing a similar task (bimanual coordination task 428 with 90° relative phase) only required 5 min of practice to perform 429 the task well (Kovacs et al., 2009). Older adults may require longer pe430 riods of practice with a more difficult task to improve motor learning 431 Q15 due to confounding factors, such as stress-induced effects or lower cog432 nitive resources. Nonetheless, it is remarkable that within 80 practice 433 trials older adults who practiced with an easy task (0° relative phase) 425 426

438 439 440 441 442 443 444

446 Q16 447 Q3 448 449 450

P

Acknowledgements

D

This work was supported by National Institute on Aging Grant R01 451 AG-031769 to E. A. Christou. 452 References

453

Baweja, H.S., Kwon, M., Christou, E.A., 2012. Magnified visual feedback exacerbates positional variability in older adults due to altered modulation of the primary agonist muscle. Exp. Brain Res. 222, 355–364. Beek, P.J., Peper, C.E., Stegeman, D.F., 1995. Dynamical models of movement coordination. Hum. Mov. Sci. 14, 573–608. Bernard-Demanze, L., Vuillerme, N., Ferry, M., Berger, L., 2009. Can tactile plantar stimulation improve postural control of persons with superficial plantar sensory deficit? Aging Clin. Exp. Res. 21, 62–68. Bo, J., Borza, V., Seidler, R.D., 2009. Age-related declines in visuospatial working memory correlate with deficits in explicit motor sequence learning. J. Neurophysiol. 102, 2744–2754. Chen, Y.T., Pinto Neto, O., de Miranda Marzullo, A.C., Kennedy, D.M., Fox, E.J., Christou, E.A., 2012. Age-associated impairement in endpoint accuracy of goal-directed contractions performed with two fingers is due to altered activation of the synergistic muscles. Exp. Gerontol. 47, 519–526. Christou, E.A., 2005. Visual feedback attenuates force fluctuations induced by a stressor. Med. Sci. Sports Exerc. 37, 2126–2133. Christou, E.A., 2011. Aging and variability of voluntary contractions. Exerc. Sport Sci. Rev. 39, 77–84. Christou, E.A., Jakobi, J.M., Critchlow, A., Fleshner, M., Enoka, R.M., 2004. The 1- to 2-Hz oscillations in muscle force are exacerbated by stress, especially in older adults. J. Appl. Physiol. 97, 225–235. Crossley, M., Hiscock, M., 1992. Age-related differences in concurrent-task performance of normal adults: evidence for a decline in processing resources. Psychol. Aging 7, 499–506. Enoka, R.M., Christou, E.A., Hunter, S.K., Kornatz, K.W., Semmler, J.G., Taylor, A.M., Tracy, B.L., 2003. Mechanisms that contribute to differences in motor performance between young and old adults. J. Electromyogr. Kinesiol. 13, 1–12. Fleg, J.L., 2012. Aerobic exercise in the elderly: a key to successful aging. Discov. Med. 13, 223–228. Greenwood, P.M., 2000. The frontal aging hypothesis evaluated. J. Int. Neuropsychol. Soc. 6, 705–726. Haken, H., Kelso, J.A., Bunz, H., 1985. A theoretical model of phase transitions in human hand movements. Biol. Cybern. 51, 347–356. Heuninckx, S., Wenderoth, N., Debaere, F., Peeters, R., Swinnen, S.P., 2005. Neural basis of aging: the penetration of cognition into action control. J. Neurosci. 25, 6787–6796. Hinder, M.R., Schmidt, M.W., Garry, M.I., Carroll, T.J., Summers, J.J., 2011. Absence of crosslimb transfer of performance gains following ballistic motor practice in older adults. J. Appl. Physiol. 110, 166–175. Kamen, G., 2005. Aging, resistance training, and motor unit discharge behavior. Can. J. Appl. Physiol. 30, 341–351. Kennedy, D.M., Christou, E.A., 2011. Greater amount of visual information exacerbates force control in older adults during constant isometric contractions. Exp. Brain Res. 213, 351–361. Kovacs, A.J., Buchanan, J.J., Shea, C.H., 2009. Bimanual 1:1 with 90° continuous relative phase: difficult or easy! Exp. Brain Res. 193, 129–136. Malone, L.A., Vasudevan, E.V., Bastian, A.J., 2011. Motor adaptation training for faster relearning. J. Neurosci. 31, 15136–15143.

454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 499 500 501

T

relative phase has been shown to be inherently difficult to coordinate between two effectors (Haken et al., 1985), and has been associated 395 with higher metabolic demands (Sparrow et al., 2005). Thus, increased 396 Q13 energy costs associated with the 90° task may have led to the increased 397 variability in older adults. Another possibility is that the greater difficul398 ty associated with the 90° task may have induced greater stress in older 399 adults. There is evidence that age-associated differences in motor out400 Q14 put variability exacerbate under stressful conditions (Christou et al. 401 2005). 402 Greater motor output variability in older adults who practiced with 403 the more difficult coordination task may also be related to decreased 404 cognitive resources (Greenwood, 2000; Miller, 2000). Age-related de405 crease in cognitive resources (e.g. executive control and attentional re406 sources) has been shown to exacerbate age-related motor deficits 407 (Bernard-Demanze et al., 2009; Crossley and Hiscock, 1992; Heuninckx 408 et al., 2005; Teasdale et al., 1993; Voelcker-Rehage et al., 2006). For 409 instance, simultaneously performing a cognitive and motor task (dual410 task) induces higher attention demands in older adults, and increasing 411 task complexity further worsens their force control (Voelcker-Rehage 412 et al., 2006). In the current study, we observed poorer motor perfor413 mance in the older adults who practiced with the more difficult task. 414 The difficult task may have required greater demand on general cogni415 tive processing resources and induced dual-task-like effects on the con416 trol of force in older adults. Similarly, increasing cognitive resources can 417 slow motor adaptations as well (Malone et al., 2011). Although older 418 adults in the difficult task practice group performed better after the ini419 tial block of trials, their force variability and errors remained higher 420 throughout the practice session compared with the easy task practice 421 group. Perhaps if older adults in the hard practice task group were 422 given more practice trials, they may have decreased force variability to 423 similar levels as the easy task practice group.

E

393 394

O

The authors have no conflicts of interests.

436 437

Please cite this article as: Onushko, T., et al., Reducing task difficulty during practice improves motor learning in older adults, Exp. Gerontol. (2014), http://dx.doi.org/10.1016/j.exger.2014.06.006

T. Onushko et al. / Experimental Gerontology xxx (2014) xxx–xxx Mayer, F., Scharhag-Rosenberger, F., Carlsohn, A., Cassel, M., Muller, S., Scharhag, J., 2011. The intensity and effects of strength training in the elderly. Dtsch. Arztebl. Int. 108, 359–364. Miller, E.K., 2000. The prefrontal cortex and cognitive control. Nat. Rev. Neurosci. 1, 59–65. Olafsdottir, H., Zhang, W., Zatsiorsky, V.M., Latash, M.L., 2007. Age-related changes in multifinger synergies in accurate moment of force production tasks. J. Appl. Physiol. 102, 1490–1501. Oldfield, R.C., 1971. The assessment and analysis of handedness: the Edinburgh inventory. Neuropsychologia 9, 97–113. Onushko, T., Kim, C., Christou, E.A., 2012. Training with an Easy Motor Task Enhances Motor Learning in Older Adults. Society for Neuroscience, Washington D.C., Osu, R., Franklin, D.W., Kato, H., Gomi, H., Domen, K., Yoshioka, T., Kawato, M., 2002. Short- and long-term changes in joint co-contraction associated with motor learning as revealed from surface EMG. J. Neurophysiol. 88, 991–1004. Panzer, S., Gruetzmacher, N., Fries, U., Krueger, M., Shea, C.H., 2011. Age-related effects in interlimb practice on coding complex movement sequences. Hum. Mov. Sci. 30, 459–474. Ranganathan, V.K., Siemionow, V., Sahgal, V., Liu, J.Z., Yue, G.H., 2001. Skilled finger movement exercise improves hand function. J. Gerontol. A Biol. Sci. Med. Sci. 56, M518–M522.

Schmidt, R.A. (Ed.), 1988. Motor Control and Learning. Human Kinetics. Seidler, R.D., 2006. Differential effects of age on sequence learning and sensorimotor adaptation. Brain Res. Bull. 70, 337–346. Shinohara, M., Li, S., Kang, N., Zatsiorsky, V.M., Latash, M.L., 2003. Effects of age and gender on finger coordination in MVC and submaximal force-matching tasks. J. Appl. Physiol. 94, 259–270. Sparrow, W.A., Parker, S., Lay, B., Wengier, M., 2005. Aging effects on the metabolic and cognitive energy cost of interlimb coordination. J. Gerontol. A Biol. Sci. Med. Sci. 60, 312–319. Teasdale, N., Bard, C., LaRue, J., Fleury, M., 1993. On the cognitive penetrability of posture control. Exp. Aging Res. 19, 1–13. Voelcker-Rehage, C., Alberts, J.L., 2007. Effect of motor practice on dual-task performance in older adults. J. Gerontol. B Psychol. Sci. Soc. Sci. 62, 141–148. Voelcker-Rehage, C., Stronge, A.J., Alberts, J.L., 2006. Age-related differences in working memory and force control under dual-task conditions. Neuropsychol. Dev. Cogn. B Aging Neuropsychol. Cogn. 13, 366–384. Wolpert, D.M., Diedrichsen, J., Flanagan, J.R., 2011. Principles of sensorimotor learning. Nat. Rev. Neurosci. 12, 739–751.

F

502 503 504 505 506 507 508 509 510 511 512 513 514 515 516 517 518 519 520 521

7

U

N C O

R

R

E

C

T

E

D

P

R O

O

540

Please cite this article as: Onushko, T., et al., Reducing task difficulty during practice improves motor learning in older adults, Exp. Gerontol. (2014), http://dx.doi.org/10.1016/j.exger.2014.06.006

522 523 524 525 526 527 528 529 530 531 532 533 534 535 536 537 538 539