Older adults with history of falls are unable to perform walking and prehension movements simultaneously

NSC 16794

No. of Pages 12

29 December 2015
Please cite this article in press as: Rinaldi NM, Moraes R. Older adults with history of falls are unable to perform walking and prehension movements
simultaneously. Neuroscience (2015), />1

Neuroscience xxx (2015) xxx–xxx

4

OLDER ADULTS WITH HISTORY OF FALLS ARE UNABLE
TO PERFORM WALKING AND PREHENSION MOVEMENTS
SIMULTANEOUSLY

5

N. M. RINALDI a,b* AND R. MORAES b,c

6

a

7
8

b

2
3

9
10

11

et al., 1993; Perracini and Ramos, 2002). Consequently,
fall-related injuries are associated with a poorer quality
of life due to restricted mobility and functional decline
(Tideiksaar, 1996). In addition, one of the major intrinsic
risk factors for falls in older adults is deficit in static and
dynamic postural control (Verghese et al., 2007). Importantly, more than 50% of falls occur during locomotion
(Barak et al., 2006).
Older adults with a history of falls (FOA) present some
gait impairments (Hausdorff et al., 2001), such as a
decrease in stride length and velocity, and an increase in
gait variability and double support time (Kirkwood et al.,
2011; Toebes et al., 2012). These changes in the walking
pattern are even more evident when two motor tasks are
combined (Nordin et al., 2010). FOA have a slower swing
time and step velocity than older adults without a history of
falls in a dual task paradigm (Springer et al., 2006). These
results suggest that FOA may have problems switching
their attention between two motor tasks due to neuromuscular problems (Hawkes et al., 2012). The changes in the
walking behavior during a dual task paradigm can predict
falls in older adults (Beauchet et al., 2009). Moreover,
the level of difficulty of the secondary task can also influence how dual-task-related changes are associated with
a history of falls (Chu et al., 2013). Nordin et al. (2010)
investigated, in FOA, gait changes during dual task conditions at different levels of difficulty. They found that FOA
increased their step width in the two most difficult tasks

(task 1: carry a saucer with a coffee cup in one hand; task
2: perform serial subtractions by three starting from 50).
These results indicated the usage of sensory-motor
resources in a flexible manner to decrease the risk of falls
(i.e., a protective strategy). Hall et al. (2011) investigated
the impact of cognitive task level of difficulty on walking
of FOA. FOA reduced gait speed when cognitive task
demand increased, suggesting that the more difficult the
secondary task is, the greater the impact on gait performance. Furthermore, FOA performed the alphabet and
alternate letters tasks more slowly in walking than in the
seated condition. With an increase in task difficulty, older
adults must allocate more attentional resources to walking
to compensate for the reduction in sensory-motor control
(Stelmach et al., 1990).
The combined task of walking and prehension (i.e.,
reach-to-grasp) is widely performed during activities of
daily life. Older adults exhibit smaller peak wrist velocity
and greater movement times than young adults when
reaching for an object (Roy et al., 1996). Furthermore,

Ribeirao Preto Medical School, University of Sao Paulo, Brazil

Research Support Center on Chronic-Degenerative
Diseases, University of Sao Paulo, Brazil
c

Biomechanics and Motor Control Lab, School of Physical
Education and Sport of Ribeirao Preto, University of Sao Paulo, Brazil

Abstract—Older adults have a greater incidence of falls, and

risk of falls will increase when combining two motor tasks.
Thus, it is interesting to investigate the effect of fall history
on motor performance in older adults when combining walking with another task such as grasping an object. The aim of
this study was to investigate the combined task of walking
and prehension with different levels of manual task difficulty
in older adults with and without a history of falls. Thirty
older adults participated in this study; groups were designated as fallers (n = 15) and non-fallers (n = 15). Participants were asked to reach-to-grasp a dowel during quiet
standing and during walking. Level of manual task difficulty
was manipulated by the type of dowel support and obstacles
located at different distances to the sides of the dowel. Fall
history influenced the performance of this combined task
for the most difficult manual conditions. Fallers were able
to be identified due to differences in the grasping strategies
used while walking compared to non-fallers. In addition,
walking and grasping were mutually modulated due to the
level of difficulty of the manual task. Ó 2015 Published by
Elsevier Ltd. on behalf of IBRO.

Key words: aging, falls, locomotion, prehension, dual task.
12

13

INTRODUCTION

14

It is estimated that one-third of community-dwelling
people aged 65 and older fall every year (O’loughlin

15

*Correspondence to: N. M. Rinaldi, Faculdade de Medicina de
Ribeira˜o Preto, Programa de Po´s-Graduac¸a˜o em Reabilitac¸a˜o e
Desempenho Funcional, Universidade de Sa˜o Paulo, Avenida dos
Bandeirantes, 3900 Ribeira˜o Preto, SP 14049-900, Brazil. Tel: +5516-3315-0359; fax: +55-16-3315-0551.
E-mail address: (N. M. Rinaldi).
Abbreviations: ANOVAs, analysis of variances; AP, anterior–posterior;
COM, center of mass; FOA, Older adults with a history of falls; HCs,
heel contacts; MDS, margin of dynamic stability; ML, medial–lateral;
MMSE, Mini Mental State Examination; OA, older adults with no history
of falls; SB, stable base; SLD, stable base with obstacles at long; SSD,
stable base with obstacles at short; UB, unstable base; WT, walking
baseline.
/>0306-4522/Ó 2015 Published by Elsevier Ltd. on behalf of IBRO.
1

16
17
18
19
20
21
22
23
24
25
26
27
28

29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58

59
60
61
62
63


NSC 16794

No. of Pages 12

29 December 2015

2
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79

80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109

110
111
112
113
114
115
116
117
118
119
120
121
122
123
124

N. M. Rinaldi, R. Moraes / Neuroscience xxx (2015) xxx–xxx

older adults have reduced tactile sensitivity and, consequently, increase the grip force as a compensatory strategy (Gorniak et al., 2011). During walking aging has been
found to affect prehension. Diermayr et al. (2011) investigated the aging effects on grasp control when walking and
transporting an object. They found an increase in grip
force while walking, indicating a decline in manual dexterity while performing functional tasks.
Interestingly, Delbaere et al. (2004) found that walking
and reaching are the most avoided tasks in older adults
with fear of falling. When reaching for an object in an
upright position older adults adopted a hip strategy to perform the task, which is different than that of young adults
who preferred an ankle strategy (Delbaere et al., 2004).
Additionally, Huang and Brown (2015) found that older
adults showed a larger center of pressure excursions
compared to young individuals when combining upright

stance with reach-to-grasp. These different strategies
are likely to compensate for constraints in balance-related
functions. Thus, it becomes interesting to combine these
two tasks because they have the potential to challenge
dynamic stability due to mechanical constraints and, at
the same time, increase cognitive load because this combined task is also a dual task.
Many studies have investigated the interference of
motor/cognitive tasks on walking and the relationship to
fall risk in older adults (Menant et al., 2014). However,
most of these studies involving dual task paradigms and
FOA investigated primarily the main task (i.e., walking)
(Beauchet et al., 2009). Recently, we found modifications
in walking and prehension when combining these two
tasks in young adults (Rinaldi and Moraes, 2015). We
suggested that prehension was superimposed on gait,
although the adaptations in motor behavior were global
because both motor patterns were modified to guarantee
the execution of prehension with different levels of difficulty while walking without stopping. Then, in this context
of dual task and falls, it is important to analyze both tasks
to investigate the level of interference between these two
motor tasks in FOA. Possible changes in the prehension
control, such as, reduced movement time, wrist velocity
and grip aperture velocity could be related to changes in
walking control. Changes in gait stability could be part of
a compensatory strategy to accommodate the control of
upper body movements toward an object in FOA. Furthermore, this combined motor task is different from other
dual task paradigms in the literature (Yamada et al.,
2011), because most studies have older adults perform
the secondary task during the entire pathway (e.g., carrying a tray). Thus, they do not need to change their motor
strategy to perform the secondary task, since they could

preprogram their movement from the beginning of the
walking task. However, to perform daily life activities,
older adults are required to change their walking patterns
to accommodate other tasks (e.g., prehension). Based on
these assumptions, our combined motor task can contribute to investigate the motor strategies used by FOA
when they have to disrupt the walking pattern to superimpose a voluntary, discrete task.
Based on these considerations, this study presents two
main research questions: (1) what are the changes in

prehension and walking when these tasks are combined
in FOA? (2) Do these changes occur as a function of the
manual task difficulty? To answer these questions, we
analyzed variables based on whole body center of mass
(COM) (including stability measures) and spatiotemporal
gait parameters to describe the possible changes in
walking of the FOA due to manual task difficulty. We
analyzed two steps before object grasping to investigate
the changes in walking during the approach phase. In
relation to reach-to-grasp, we analyzed the reaching and
grasping components, such as reaching duration and
velocity, and hand grip aperture and velocity. We also
investigated prehension variables in the upright stance to
identify changes in reach-to-grasp due to the addition of
walking. Therefore, the aim of this study was to
investigate the combined task of walking and prehension
with different levels of manual task difficulty in older
adults with and without a history of falls.

125

EXPERIMENTAL PROCEDURES

143

126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142

Participants

144

Thirty individuals participated in this study. They were
distributed in two groups (n = 15): older adults with no
history of falls (OA) (15 females); older adults who
experienced at least one fall in the 12-month period

prior to data collection (FOA) (15 females). Participants
were screened before starting the experimental task by
filling out a clinical questionnaire to check the history of
falls, health status, physical activity level (Baecke)
(Voorrips et al., 1991), cognitive function (Mini Mental
State Examination, MMSE) (Folstein et al., 1975) and balance performance (Mini-BESTest) (Maia et al., 2013).
Participants were excluded if they had cognitive impairment (<24 points in the MMSE), vestibular dysfunction,
and/or if they were unable to walk without assistance.
We invited participants through local media (newspaper,
television and radio). Forty-eight older adults returned to
our invitation. We did an initial contact by phone and we
asked them whether or not they experienced a fall in the
last 12 months, after explaining to them that a fall was
an event in which they came to the ground or to some
lower level unintentionally, regardless of the consequences of the fall. After this screening, 28 older adults
reported a recent history of fall and 20 older adults did
not experience a fall in the last 12 months. However,
regarding the FOA, seven individuals did not attend to
the inclusion criteria (visual problem [n = 1] and use of
assistive devices [n = 6]). Yet, six individuals refused to
participate in the study. For the OA, five individuals did
not attend to the inclusion criteria (neurological disorders
[n = 3] and musculoskeletal problems [n = 2]).
All participants had normal or corrected-to-normal
vision and no neurological/musculoskeletal disorders
that would affect task performance. The local ethics
committee approved all procedures and participants
signed a consent form before starting the experiment.

145

Experimental protocol

180

For data collection, we used an 8-camera motion analysis
system (MX-T40S, Vicon) with a sampling rate of 100 Hz.

181

Please cite this article in press as: Rinaldi NM, Moraes R. Older adults with history of falls are unable to perform walking and prehension movements
simultaneously. Neuroscience (2015), />
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163

164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179

182


NSC 16794

No. of Pages 12

29 December 2015

N. M. Rinaldi, R. Moraes / Neuroscience xxx (2015) xxx–xxx
183
184
185

186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215

216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236

Passive reflective markers were placed on participants’
skin at predefined landmarks according to the Plug-inGait Full Body model (Vicon) and two markers were
placed on the index finger and thumb, respectively.
Moreover, we positioned two video cameras (Bonita,
Vicon) in front of and on the left side of the participants.
We used the video camera images to determine dowel
contact (visual inspection), defined as the first contact of
the fingers with the dowel.

Participants performed two experimental tasks:
reaching-to-grasping a dowel (a wood cylinder,
diameter: 4.5 cm, height: 10 cm, mass: 105 g) during
quiet standing (stationary) and during walking. We
positioned the dowel on the top of a base made of wood
and placed over a support located approximately 3 m
from the starting position. We adjusted the height of the
support to height of the participants’ greater trochanter.
We positioned the support on the right side of the
walkway, with a distance corresponding to 50% of the
participants’ right arm length. This distance was also
used for the stationary task and, in this task,
participants stood behind the object at a distance of
30% of their right upper limb length. The base of the
dowel could be stable (wide base) and unstable (narrow
base). We made the manual task more difficult by
placing the dowel between two wooden obstacles such
that for each type of support there were three obstacle
conditions: no obstacle, short distance, and long
distance. The short and long distances of the obstacles
corresponded to three and five times the right hand
thickness, respectively. The combination of type of base
and obstacle resulted in six grasping conditions: stable
(SB) and unstable (UB) bases without obstacles, stable
base with obstacles at short (SSD) and long (SLD)
distances, and unstable base with obstacles at short
(USD) and long (ULD) distances (Fig. 1). Participants
also performed a baseline walking condition (WT)
without grasping (control condition). More details about
the experimental procedures are available in our

previous study (Rinaldi and Moraes, 2015).
We asked participants to reach-to-grasp the dowel as
they walked, without stopping, at a self-selected speed.
After grasping the dowel, participants were instructed to
hold it and walk normally until the end of the pathway.
For the stationary task, we instructed them to stand as
quietly as possible. In both tasks, participants were not
allowed to contact the obstacles and knock down the
support. Participants performed 21 trials for the walking
task, and 18 trials for the stationary task. We collected
these tasks in separate blocks and counterbalanced
them within each group. Trials were completely
randomized within each block. We repeated those trials
with errors at the end of each block without participants’
awareness.

237

Data analysis

238

The tridimensional coordinates of the individual markers
were digitally filtered using a 4th-order Butterworth filter
with a 6-Hz cut-off frequency. For the walking task, all
variables were calculated for the step at the period of
contact with the dowel (N) and two steps before contact

239
240

241
242

3

(N-2 and N-1) to verify possible gait adjustments during
the approach phase. For the WT condition, the dowel
was kept on the support to use as a reference to
identify the region corresponding to steps N-2, N-1,
and N.

243

Spatiotemporal gait parameters. Step width and step
length were calculated as the absolute difference
between heel markers of the right and left feet at
subsequent heel contacts (HCs) in each direction,
medial–lateral (ML) and anterior–posterior (AP),
respectively. Step duration corresponded to the frame
difference between each HC divided by the sampling
frequency. The division of step length by step duration
resulted in step velocity.

248

COM variables and margin of dynamic stability (MDS).
Nexus software (Vicon) computed the tridimensional
COM coordinates based on the tridimensional
coordinates of the 39 markers, which defined a
15-segment model (Winter, 2005). COM velocity corresponded to the first derivative of the COM position

(central difference procedure). We identified the minimum
COM AP velocity during the approach phase until dowel
contact for all walking conditions. After that, we computed
the percentage of COM AP velocity reduction as the difference between minimum COM AP velocity in the WT
condition and in the walking combined with prehension
conditions (Fig. 3A). We also calculated the temporal difference between the time of dowel contact and minimum
COM AP velocity (Fig. 3A).
For the computation of the MDS, we first calculated
the extrapolated COM position (XcoM) (Hof et al.,
2005). Based on XcoM, MDS was calculated according
to Rinaldi and Moraes (2015), where metatarsal and heel
markers on both feet were used to define the outside
boundaries of the foot (foot edge) in the AP and ML directions. We calculated the MDS by the difference between
foot edge and XcoM position (AP and ML directions). A
positive value for the MDS indicates that XcoM is located
before the foot edge and the system is dynamically stable.

257

Reaching-to-grasping variables. We used the interval
between reaching onset and dowel contact to calculate
the reach-to-grasp variables. The methods used to
detect reaching onset for walking and stationary tasks
were the same as the ones used by Rinaldi and Moraes
(2015). The reaching variables analyzed were: movement
time (temporal difference between reaching onset and
dowel contact), peak wrist velocity (maximum value
obtained in the resultant wrist velocity curve), and timeto-peak wrist velocity (time of occurrence of the peak wrist
velocity adjusted to movement time, %). The peak wrist
velocity was calculated based on the relative position of

the right wrist to the right iliac crest (i.e., relative to the
person’s body position in space, as used by Carnahan
et al. (1996)). The grasping variables analyzed were:
peak grip aperture (maximum distance between the markers on thumb and index finger), time-to-peak grip aperture
(time of occurrence of peak grip aperture adjusted to
movement time, %), peak grip aperture velocity (maxi-

282

Please cite this article in press as: Rinaldi NM, Moraes R. Older adults with history of falls are unable to perform walking and prehension movements
simultaneously. Neuroscience (2015), />
244
245
246
247

249
250
251
252
253
254
255
256

258
259
260
261
262

263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281

283
284
285
286
287
288
289
290
291
292

293
294
295
296
297
298
299
300


NSC 16794

No. of Pages 12

29 December 2015

4

N. M. Rinaldi, R. Moraes / Neuroscience xxx (2015) xxx–xxx

Fig. 1. (A) Transverse view of the experimental set-up for the walking and stationary tasks. It shows the three steps (N-2, N-1, and N) selected for
data analysis and dowel’s location for walking and stationary tasks. (B) Illustration of the six prehension conditions. Dowel is shown in white,
obstacles in black and base in gray colors (Rinaldi and Moraes, 2015).

301
302
303
304
305
306

307
308

mum value obtained in the resultant velocity curve of the
distance thumb-finger, which was determined as the first
derivative of the thumb-finger distance), and time-to-peak
grip aperture velocity (time of occurrence of peak grip
aperture velocity adjusted to movement time, %).
We also analyzed the relationship between reaching
onset and gait events. For that, we calculated the
temporal difference between the last right HC before

dowel contact and reaching onset. Negative values
indicated that the participant touched the ground before
starting the reaching movement.

309

Statistical analysis

312

One-way analysis of variances (ANOVAs) were
computed to compare age, anthropometric (height and

313

Please cite this article in press as: Rinaldi NM, Moraes R. Older adults with history of falls are unable to perform walking and prehension movements
simultaneously. Neuroscience (2015), />
310

311

314


NSC 16794

No. of Pages 12

29 December 2015

N. M. Rinaldi, R. Moraes / Neuroscience xxx (2015) xxx–xxx
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332

333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348

body mass) and clinical characteristics (MMSE, MiniBESTest, and Baecke scores) between groups. For the
remaining data, five MANOVAs and three ANOVAs
were performed. Three-way MANOVAs (group [FOA,
OA] Â conditions [SB, SLD, SSD, UB, ULD, USD,
WT] Â step [N-2, N-1, N]) with repeated measures in the
last two factors were carried out for the following set of
dependent variables: (1) step length and step width; (2)
step duration and step velocity; and (3) MDS in the AP
and ML directions. Two-way ANOVAs (group Â
conditions) with repeated measures for the last factor
were carried out for the following dependent variables:
(1) temporal difference between HC and reaching onset;
(2) temporal difference between dowel contact and

minimum COM AP velocity; and (3) AP velocity
reduction. For reaching and grasping variables, two
three-way MANOVAs were calculated (group Â
conditions [SB, SLD, SSD, UB, ULD, USD] Â task
[walking and stationary]) with repeated measures in the
last two factors for the following set of dependent
variables: (1) movement time, peak wrist velocity, and
time-to-peak wrist velocity; and (2) peak grip aperture,
time-to-peak grip aperture, peak grip aperture velocity,
and time-to-peak grip aperture velocity. MANOVAs were
followed by univariate analyses, which revealed main
and interaction effects; therefore we focused on group
and interaction effects. Post hoc tests with Bonferroni
adjustments were performed for main and interaction
effects. We computed the effect size using the eta
squared (g2) parameter. The cut-off criteria for the effect
size (partial eta squared [g2]) were: small effect
(0.20 6 g2 < 0.50), medium effect (0.50 6 g2 < 0.80),
and large effect (g2 P 0.80) as suggested by Cohen
(1992). The level of significance was set at p 6 0.05.

349

RESULTS

350

We present results of the MANOVAs and ANOVAs in
Tables 1 and 2. Since there was no interaction between
step*condition and task*condition for any of the

dependent variables, we exclude the statistical results of
these interactions from both Tables 1 and 2.

351
352
353
354

355

Sample characteristics

356

ANOVA showed difference between groups for the MiniBESTest (F1,28 = 112.7, p 6 0.001). FOA (22.2 pts)
scored lower than OA (27.3 pts), indicating impaired
balance. However, groups were similar in age
(FOA = 71.8 years|OA = 70.1 years), height (FOA =
1.55 m|OA = 1.54 m), body mass (FOA = 65.7 kg|
OA = 59.9 kg), MMSE (FOA = 27 pts|OA = 28.2 pts)
and physical activity level (FOA = 4.2 pts|OA =
4.1 pts). In addition, FOA presented a mean of 2.1 falls
in the last 12 months.

357
358
359
360
361
362

363
364
365

366

Spatiotemporal gait parameters

367

Step width was greater for FOA (0.09 m) than for OA
(0.06 m) (p = 0.003) (Fig. 2A). Step length was greater
in WT (0.58 m) than in all grasping conditions (0.47 m)
(p = 0.001) (Fig. 2B). Relative to the step effect, step

368
369
370

5

width was greater in step N (0.09 m) than N-2 (0.07 m)
and N-1 (0.06 m) (p = 0.001) (Fig. 2A). However, step
length was greater in step N-2 (0.54 m) than N-1
(0.48 m) and N (0.46 m) (p 6 0.001) (Fig. 2B).
Step duration was greater for FOA (0.62 s) than for OA
(0.52 s) (p = 0.03) (Fig. 2C). FOA (0.64 m/s) presented a
lower step velocity than OA (0.93 m/s) in step N
(p 6 0.001) (Fig. 2D). For the effect of condition, step
duration was greater for the stable base with obstacles

at short distance (0.61 s) and the stable base with
obstacles at long distance (0.58 s) conditions than for
walking through (0.51 s) (p 6 0.001) (Fig. 2C). However,
step velocity was lower in the grasping conditions
(0.88 m/s) than in walking through (1.15 m/s) (p 6 0.001)
(Fig. 2D). Step duration was lower in steps N-2 (0.52 s)
and N-1 (0.53 s) than in step N (0.66 s) (p = 0.002)
(Fig. 2C). Step velocity was greater in steps N-2 (1.04 m/
s) and N-1 (0.93 m/s) than in step N (0.78 m/s) (Fig. 2D).

371

COM variables and MDS

389

For the percentage of COM AP velocity, FOA (60%)
presented a greater reduction in AP velocity than OA
(30%) (Fig. 3B). For the effect of condition, COM AP
velocity reduction was greater in the obstacle (stable
base with obstacles at short distance: 56.5%; stable
base with obstacles at long distance: 45.7%; unstable
base with obstacles at short distance: 59.7%; unstable
base with obstacles at long distance: 47.9%) than in the
no-obstacle conditions (stable base: 28.9%; unstable
base: 32.4%) (p 6 0.001) (Fig. 3B). In absolute values,
the mean COM AP velocity for FOA was equal to
0.39 m/s and 0.75 m/s for the OA.
The temporal difference between minimum velocity
and dowel contact was greater for FOA (3.51 s) than

OA (2.58 s) (Fig. 3B). This result showed that FOA
exhibited a minimum velocity earlier than OA before
dowel contact (Fig. 3B). In addition, the minimum
velocity occurred earlier in the stable base condition
(2.71 s) than in the stable base at short distance
condition (3.31 s) (p = 0.001) (Fig. 3B).
In both directions, MDS was greater for FOA (AP:
0.07 m, ML: 0.04 m) than OA (AP: 0.02 m, ML: 0.01 m)
(p = 0.002) (Fig. 3C). The MDS AP was greater in the
stable base with obstacles at short distance (0.07 m)
and the unstable base with obstacles at short distance
(0.06 m) conditions than in walking through (0.01 m)
(p = 0.002) (Fig. 3C). In addition, the MDS AP was
greater in step N-1 (0.07 m) than in step N (0.03 m)
(p = 0.017) (Fig. 3C).

390

Reaching-to-grasping variables

419

FOA presented a greater movement time and a lower
peak wrist velocity (1.57 s|27.7 m/s) than OA (1.21 s|
31.7 m/s) (p = 0.001) (Fig. 4A, B). Movement time and
peak wrist velocity were greater for stationary (1.62 s|
0.46 m/s) than for the walking (1.16 s|0.38 m/s,
respectively) task (p 6 0.001) (Fig. 4A , B). In addition,
the time- to- peak wrist velocity occurred earlier in
walking (18.3%) than in the stationary (41.2%) task

(p 6 0.001) (Fig. 4C). Regarding the condition effect, the

420

Please cite this article in press as: Rinaldi NM, Moraes R. Older adults with history of falls are unable to perform walking and prehension movements
simultaneously. Neuroscience (2015), />
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388

391
392
393
394
395
396

397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418

421
422
423
424
425
426
427

428


NSC 16794

No. of Pages 12

29 December 2015

6

N. M. Rinaldi, R. Moraes / Neuroscience xxx (2015) xxx–xxx

Table 1. F- and p-values for main and interaction effects (condition, step and condition*step interaction) of the MANOVAs and the univariate ANOVAs
for the spatiotemporal gait parameters (step length, step width, step duration and step velocity), margin of dynamic stability in AP and ML directions,
temporal difference between minimum velocity and dowel contact, and AP velocity reduction (%)
Variables

Group

Condition

Step

Condition*group

Step*group

MANOVA

Wilk’s Lambda = 0.629,
F2,27 = 7.95, p = 0.002,
g2 = 0.371

Wilk’s
Lambda = 0.121,
F12,17 = 10.28,
p 6 0.0001,
g2 = 0.879

Wilk’s
Lambda = 0.232,
F4,25 = 20.66,
p 6 0.0001,
g2 = 0.768

Wilk’s
Lambda = 0.365,
F12,17 = 2.46,
p = 0.044,
g2 = 0.635

Wilk’s
Lambda = 0.717,
F4,25 = 2.47,
p = 0.070,
g2 = 0.283

F6,168 = 25.63,
p 6 0.0001,

g2 = 0.478
F6,168 = 3.39,
p = 0.017,
g2 = 0.108

F2.56 = 28.98,
p 6 0.0001,
g2 = 0.509
F2.56 = 8.78,
p = 0.002,
g2 = 0.239

F6,636 = 1.88,
p = 0.132,
g2 = 0.063
F6,636 = 2.46,
p = 0.061,
g2 = 0.081

F2,56 = 4.31,
p = 0.018,
g2 = 0.283
F2,56 = 0.55,
p = 0.525,
g2 = 0.133

Wilk’s
Lambda = 0.120,
F12,17 = 10.38,
p 6 0.0001,

g2 = 0.880

Wilk’s
Lambda = 0.163,
F4,25 = 32.16,
p 6 0.0001,
g2 = 0.837

Wilk’s
Lambda = 0.393,
F12,17 = 2.19,
p = 0.068,
g2 = 0.325

Wilk’s
Lambda = 0.675,
F4,25 = 3.04,
p = 0.037,
g2 = 0.607

F6,168 = 4.31,
p = 0.007,
g2 = 0.133
F6,168 = 52.03,
p 6 0.0001,
g2 = 0.650

F2,56 = 13.73,
p 6 0.0001,
g2 = 0.329

F2,56 = 64.32,
p 6 0.0001,
g2 = 0.697

F6,168 = 1.55,
p = 0.206,
g2 = 0.053
F6,168 = 1.57,
p = 0.193,
g2 = 0.054

F2.56 = 3.78,
p = 0.060,
g2 = 0.119
F2.56 = 5.08,
p = 0.014,
g2 = 0.154

Wilk’s
Lambda = 0.338,
F2,27 = 12.06,
p 6 0.0001,
g2 = 0.662

Wilk’s
Lambda = 0.451,
F4,25 = 7.61,
p 6 0.0001,
g2 = 0.549

Wilk’s
Lambda = 0.604,
F12,17 = 0.93,
p = 0.541,
g2 = 0.596

Wilk’s
Lambda = 0.919,
F4,25 = 0.55,
p = 0.698,
g2 = 0.081

F6,168 = 4.86,
p 6 0.0001,
g2 = 0.148
F6,168 = 2.51,
p = 0.061,
g2 = 0.082

F2,56 = 6.78,
p = 0.002,
g2 = 0.195
F2,56 = 1.54,
p = 0.227,
g2 = 0.052

F6,168 = 0.559,
p = 0.521,
g2 = 0.020
F6,168 = 1.292,

p = 0.282,
g2 = 0.044

F2,56 = 0.546,
p = 0.582,
g2 = 0.019
F2,56 = 1.443,
p = 0.245,
g2 = 0.049

F1,28 = 9.681,
p = 0.004,
g2 = 0.257

F2,28 = 6.391,
p 6 0.0001,
g2 = 0.186

–

F5,140 = 0.78,
p = 0.545,
g2 = 0.027

–

F1.28 = 14.13,
p 6 0.0001,
g2 = 0.335

F5.140 = 43.67,
p 6 0.0001,
g2 = 0.609

–

F5,140 = 1.07,
p = 0.360,
g2 = 0.037

–

Follow-up univariate
Step length
F1,28 = 3.09,
p = 0.09,
g2 = 0.094
Step width
F1,28 = 10.80,
p = 0.003,
g2 = 0.278
MANOVA

Wilk’s Lambda = 0.67,
F2,27 = 6.59,
p = 0.005,
g2 = 0.328

Follow-up univariate
Step duration

F1,28 = 10.63,
p = 0.030,
g2 = 0.275
Step velocity
F1,28 = 10.62,
p = 0.020,
g2 = 0.293
MANOVA

Wilk’s Lambda = 0.528,
F2,27 = 12.06,
p 6 0.0001,
g2 = 0.472

Follow-up univariate
MDS AP
F1,28 = 5.85,
p = 0.022,
g2 = 0.173
MDS ML
F1,28 = 18.53,
p 6 0.0001,
g2 = 0.398
ANOVA
Temporal
difference
ANOVA
% AP velocity
reduction

429
430
431
432
433
434
435
436
437
438
439
440

movement time was lower for conditions without
obstacles (stable base: 1.25 s|unstable base: 1.28 s)
than with obstacles (stable base with obstacles at short
distance: 1.55-s|unstable base with obstacles at short
distance: 1.56 s) (p 6 0.001) (Fig. 4A). In addition, the
movement time was greater for conditions with
obstacles at short (stable base with obstacles at short
distance: 1.55-s|unstable base with obstacles at short
distance: 1.56 s) versus long distances (stable base with
obstacles at long distance: 1.36-s|unstable base with
obstacles at long distance: 1.36 s) (p 6 0.001) (Fig. 4A).
Peak wrist velocity and time-to-peak wrist velocity were

greater in the stable base (0.45 m/s|32%) than in the
stable base with obstacles at short distance (0.39 m/s|
26%) condition (p = 0.010) (Fig. 4B, C). For the
unstable base, peak wrist velocity and time- to- peak

wrist velocity were lower for the unstable base with
obstacles at short distance (0.39 m/s|27%) than for the
unstable base with obstacles at long distance (0.43 m/s|
33%) condition (p = 0.008) (Fig. 4B, C).
The temporal difference between right HC and
reaching onset was greater for FOA (0.36 s) than for OA
(À0.14 s). This indicated that FOA touched the ground
before reaching onset (Fig. 4D).

Please cite this article in press as: Rinaldi NM, Moraes R. Older adults with history of falls are unable to perform walking and prehension movements
simultaneously. Neuroscience (2015), />
441
442
443
444
445
446
447
448
449
450
451
452


NSC 16794

No. of Pages 12

29 December 2015

7

N. M. Rinaldi, R. Moraes / Neuroscience xxx (2015) xxx–xxx

Table 2. F- and p-values for main and interaction effects (condition, step and condition*step interaction) of the MANOVAs and the univariate ANOVAs
for reaching (movement time, peak wrist velocity, time-to-peak wrist velocity, temporal difference) and grasping (peak grip aperture, time-to-peak grip
aperture, peak grip aperture velocity, time-to-peak grip aperture velocity) variables
Variables

Group

Condition

Task

Task*group

MANOVA

Wilk’s
Lambda = 0.618,
F3,25 = 5.15,
p = 0.007,
g2 = 0.402

Wilk’s
Lambda = 0.110,
F3,25 = 7.04,
p = 0.0001,

g2 = 0.883

Wilk’s
Lambda = 0.121,
F3,25 = 60.27,
p 6 0.0001,
g2 = 0.877

Wilk’s
Lambda = 0.880,
F3,25 = 1.13,
p = 0.354,
g2 = 0.110

F1,27 = 4.37,
p = 0.046,
g2 = 0.159
F1,27 = 15.57,
p = 0.0001,
g2 = 0.385
F1,27 = 2.68,
p = 0.113,
g2 = 0.098

F5,135 = 21.78,
p 6 0.0001,
g2 = 0.452
F5,135 = 6.89,
p 6 0.0001,
g2 = 0.226

F5,135 = 7.04,
p 6 0.0001,
g2 = 0.168

F1,27 = 35.27,
p 6 0.0001,
g2 = 0.573
F1,27 = 21.15,
p 6 0.0001,
g2 = 0.458
F1,27 = 120.95,
p 6 0.0001,
g2 = 0.817

F1,27 = 0.77,
p = 0.389,
g2 = 0.030
F1,27 = 1.08,
p = 0.307,
g2 = 0.053
F1,27 = 2.29,
p = 0.142,
g2 = 0.049

F1,28 = 11.98,
p = 0.002,
g2 = 0.300
Wilk’s
Lambda = 0.572,
F4,24 = 4.48,

p = 0.008,
g2 = 0.428

F5,140 = 1.92,
p = 0.145,
g2 = 0.064
Wilk’s
Lambda = 0.033,
F20,8 = 11.84,
p = 0.001,
g2 = 0.976

–

–

Wilk’s
Lambda = 0.283,
F4,24 = 15.22,
p 6 0.0001,
g2 = 0.717

Wilk’s
Lambda = 0.658,
F4,24 = 3.12,
p = 0.034,
g2 = 0.342

F1,27 = 0.18,
p = 0.674,

g2 = 0.007
F1,27 = 12.43,
p = 0.002,
g2 = 0.315
F1,27 = 4.55,
p = 0.042,
g2 = 0.144
F1,27 = 8.38,
p = 0.007,
g2 = 0.237

F5,135 = 1.87,
p = 0.167,
g2 = 0.065
F5,135 = 9.05,
p 6 0.0001,
g2 = 0.251
F5,135 = 2.77,
p = 0.064,
g2 = 0.093
F5,135 = 26.56,
p 6 0.0001,
g2 = 0.496

F1,27 = 2.58,
p = 0.120,
g2 = 0.087
F1,27 = 14.45,
p = 0.001,
g2 = 0.349

F1,27 = 52.13,
p = 0.001,
g2 = 0.659
F1,27 = 0.16,
p = 0.695,
g2 = 0.006

F1, 27 = 0.11,
p = 0.746,
g2 = 0.004
F1, 27 = 5.77,
p = 0.023,
g2 = 0.176
F1, 27 = 6.08,
p = 0.020,
g2 = 0.184
F1, 27 = 3.16,
p = 0.087,
g2 = 0.105

Follow-up univariate
Movement time

Peak wrist velocity

Time-to-peak wrist velocity

ANOVA
Temporal difference

MANOVA

Follow-up univariate
Peak grip aperture

Time-to-peak grip aperture

Peak grip aperture velocity

Time-to-peak grip aperture velocity

453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471

472

Peak grip aperture was unaffected by group, condition
and task (Fig. 5A). However, time-to-peak grip aperture
was lower for FOA (73.2%) than for OA (86.8%) in the
walking task (Fig. 5B). Time-to-peak grip aperture was
lower for conditions without obstacles than for
conditions with obstacles at both short and long
distances (stable base: 72.2%|stable base with
obstacles at long distance: 78.6%|stable base with
obstacles at short distance: 78.7%|unstable base:
71.9%|unstable base at long distance: 79.9%|unstable
base at short distance: 79.1%) (p 6 0.001) (Fig. 5B). In
addition, time-to-peak grip aperture was greater in
walking (79.9%) than in stationary (73.6%) task
(p = 0.001) (Fig. 5B).
The peak grip aperture velocity and the time-to-peak
grip aperture velocity were lower for FOA (0.12 m/s|
37.2%) than for OA (0.15 m/s|47.3%) (p 6 0.001)
(Fig. 5C, D). The peak grip aperture velocity was greater
in the stationary (0.18 m/s) than in the walking task
(0.09 m/s) (p = 0.01) (Fig. 5C). For the condition effect,

the time- to- peak grip aperture velocity was lower in
conditions without obstacles than in conditions with
obstacles at short and long distances (stable base:
31.6%|stable base with obstacles at long distance:
45.2%|stable base with obstacles at short distance:
48.2%|unstable base: 29.6%|unstable base with
obstacles at long distance: 47.7%|unstable base with

obstacles at short distance: 52.2%) (p 6 0.001) (Fig. 5D).

473

DISCUSSION

481

We investigated the motor performance of walking
combined with prehension at varying levels of manual
task difficulty in older adults with and without a history of
falls. FOA performed worse than OA in the
MiniBESTest, which indicates poor balance and
mobility. This result supports the findings that FOA have
poor balance control (Cebolla et al., 2015). History of falls
did influence motor performance during the combined

482

Please cite this article in press as: Rinaldi NM, Moraes R. Older adults with history of falls are unable to perform walking and prehension movements
simultaneously. Neuroscience (2015), />
474
475
476
477
478
479
480

483

484
485
486
487
488
489


NSC 16794

No. of Pages 12

29 December 2015

8

N. M. Rinaldi, R. Moraes / Neuroscience xxx (2015) xxx–xxx

Fig. 2. Mean and standard error for step width (A), step length (B), (C) step duration and (D) step velocity for fallers and non-fallers for all
experimental conditions (SB: stable base, SLD: stable base with obstacles at long distance, SSD: stable base with obstacles at short distance, UB:
unstable base, ULD: unstable base with obstacles at long distance, USD: unstable base with obstacles at short distance, WT: walking baseline).

490
491

492
493
494
495
496

497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523

task of walking and prehension, and will be discussed
below.

FOA exhibited a more conservative walking strategy
and decoupled the combined task when compared to
OA
Walking performance in FOA was observed to be
impaired. These impairments included: reduced step
velocity and increased step width and duration when
compared to OA. These findings are in agreement with
other studies on gait changes in FOA (Barak et al.,
2006; Kirkwood et al., 2011; Toebes et al., 2012). The
slowness of FOA may indicate the need for extra time
to pick up necessary sensory information to assist with
movement performance (Chapman and Hollands, 2007).
Although the spatial-temporal gait parameters could allow
inferences about the control of stability, the MDS is a
more appropriate measurement to investigate dynamic
stability control. Our results showed an increase in the
MDS of FOA to ensure body stability. Consequently,
FOA preferred a more conservative strategy to increase
their dynamic stability compared to OA.
When walking was combined with prehension, FOA
presented a greater reduction in AP velocity than OA.
Moreover, FOA exhibited the minimum AP velocity
earlier than OA before dowel contact. Considering the
AP velocity, FOA almost stopped walking to perform the
prehension task. Altogether, these results suggest that
FOA decouple the walking and prehension tasks, which
may represent a loss of automaticity to superimpose a
discrete motor task on walking. Similarly to our task, in
activities that require movement transitions in sequence,

such as sit-to-walk, FOA have been observed to also
divide this motor task into two phases (i.e., sit-to-stand
and stand-to-walk) to achieve a more upright position

before initiating gait to ensure body stability (Chen et al.,
2013).
FOA began moving their hand toward the dowel
360 ms after HC, whereas OA started at 140 ms before
HC. The behavior of OA was more similar to young
adults (Rinaldi and Moraes, 2015), although hand movement in young adults started 460 ms before HC. The
beginning of hand movement 460 ms before HC represents the appropriate timing to exploit the upper limb forward momentum (Rinaldi and Moraes, 2015). The short
time for OA could be related to a smaller range of motion
of their upper limbs. For FOA, the beginning of hand
movement after HC is one more indication of the decoupling of walking and prehension. FOA may prefer to slow
down, almost to the point of terminating gait, before initiating hand movement toward the dowel. These changes
indicate a conservative motor strategy to achieve the goal
of the task. Older adults with fear of falling present an
extended anticipatory postural adjustment (APA) during
gait initiation under dual-task condition (Uemura et al.,
2012) to compensate for deficits in balance control. Thus,
FOA have to compensate for balance deficits when performing the combined task by dividing it in two phases:
a preparatory phase to gain stability followed by the
grasping execution.

524

History of falls affected prehension movement

549

Changes in motor performance of FOA were not
restricted to walking alone, but also occurred in the
prehension task. FOA presented slower movement time
and lower peak wrist velocity, peak grip aperture
velocity, and time-to-peak grip aperture. To the best of
our knowledge, this is the first study to show
modifications in prehension movement in FOA, which
indicates a generalized slowing down in movement

550

Please cite this article in press as: Rinaldi NM, Moraes R. Older adults with history of falls are unable to perform walking and prehension movements
simultaneously. Neuroscience (2015), />
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541

542
543
544
545
546
547
548

551
552
553
554
555
556
557


NSC 16794

No. of Pages 12

29 December 2015

N. M. Rinaldi, R. Moraes / Neuroscience xxx (2015) xxx–xxx

9

Fig. 3. Time series of COM AP velocity for one faller and one non-faller in walking combined to grasping and walking through conditions (A). The
vertical dashed lines indicate minimum velocity and dowel contact, respectively. The AP velocity reduction (%) is indicated by the difference
between the combined condition (thick line) and the baseline walking condition (thin line). The temporal difference is calculated by the difference

between the time of minimum velocity and dowel contact, mean and standard error for AP velocity reduction (left side), temporal difference between
minimum velocity and dowel contact (right side) (B) and margin of dynamic stability in AP (left side) and ML direction (right side) for experimental
conditions (SB: stable base, SLD: stable base with obstacles at long distance, SSD: stable base with obstacles at short distance, UB: unstable base,
ULD: unstable base with obstacles at long distance, USD: unstable base with obstacles at short distance, WT: walking through).
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580

performance. This slowness suggests that FOA need
more time to gain sensory information to accomplish the

manual task successfully, as has been suggested for
walking tasks (Chapman and Hollands, 2007). Although
most of studies that investigate the performance of FOA
showed changes in walking (primary task) (Hall et al.,
2011), we also found changes in the secondary motor
task (prehension). We suggest that FOA may have problems in switching attention between two motor tasks due
to neuromuscular problems, which may be related to a
loss of function in the frontal subcortical pathways
(Viswanathan and Sudarky, 2011).
Additionally, the time-to-peak grip aperture occurred
earlier for FOA than for OA only during the walking task,
indicating a strategy to increase the time available for
online control of hand configuration before dowel
contact. The need for this additional time in FOA
appeared only during the combined task. This may have
occurred due to the complexity of planning a movement
while concurrently performing another movement. Thus,
they preprogrammed less their grasping movement in
the walking task and relied more on online control for
performing grasping successfully.

Gait and prehension were mutually modified due to
the level of difficulty of the manual task in both
groups

581

Most of the studies have shown a decline in gait
performance of older adults during dual task conditions
(Hall et al., 2011). However, some studies did not show

when older adults changed their walking pattern due to
the addition of a secondary task. Our results show that
older adults made modifications in the walking pattern in
the step of object grasping and the previous step before
it. Similarly, Rinaldi and Moraes (2015) found that young
adults made modifications in their walking due to the addition of prehension one-step before it. In this context, the
level of difficulty of the prehension task influenced the gait
of older adults. We found a decrease in both step length
and speed with the addition of the prehension task. Yet,
the different levels of difficulty of the manual task affected
more some variables; COM AP velocity reduction was
more pronounced for the conditions with obstacles,
whereas the MDS increased for conditions with obstacles
at short distance as compared to WT. Thus, participants
adjusted their walking to accommodate the accurate exe-

584

Please cite this article in press as: Rinaldi NM, Moraes R. Older adults with history of falls are unable to perform walking and prehension movements
simultaneously. Neuroscience (2015), />
582
583

585
586
587
588
589
590
591

592
593
594
595
596
597
598
599
600
601
602


NSC 16794

No. of Pages 12

29 December 2015

10

N. M. Rinaldi, R. Moraes / Neuroscience xxx (2015) xxx–xxx

Fig. 4. Mean and standard error for movement time (A), peak wrist velocity (B), time-to-peak wrist velocity (C) and temporal difference between
heel contact and reaching onset (D) for fallers and non-fallers for all experimental conditions (SB: stable base, SLD: stable base with obstacles at
long distance, SSD: stable base with obstacles at short distance, UB: unstable base, ULD: unstable base with obstacles at long distance, USD:
unstable base with obstacles at short distance).

Fig. 5. Mean and standard error for peak grip aperture (A), time-to-peak grip aperture (B), peak grip aperture velocity (C), and time-to-peak grip
aperture velocity (D) for fallers and non-fallers for all experimental conditions (SB: stable base, SLD: stable base with obstacles at long distance,

SSD: stable base with obstacles at short distance, UB: unstable base, ULD: unstable base with obstacles at long distance, USD: unstable base with
obstacles at short distance).

603
604
605
606
607
608
609

cution of the manual task that may challenge their body
stability, as found in young adults (Rinaldi and Moraes,
2015). As a result, they adopted a more conservative
strategy to allow them to allocate more attention to the
grasping task and avoid errors. Previous studies also
found changes in the walking behavior of older adults
for the secondary tasks with the highest level of difficulty

(Nordin et al., 2010; Hall et al., 2011). In synthesis,
reaching-to-grasping affected postural control by changing the gait behavior.
Different conditions and tasks influenced prehension.
Participants presented a greater movement time and a
lower peak wrist velocity for conditions with obstacles
(short and long distance). Relative to grasping, the time-

Please cite this article in press as: Rinaldi NM, Moraes R. Older adults with history of falls are unable to perform walking and prehension movements
simultaneously. Neuroscience (2015), />
610
611

612
613
614
615
616


NSC 16794

No. of Pages 12

29 December 2015

N. M. Rinaldi, R. Moraes / Neuroscience xxx (2015) xxx–xxx

655

to-peak grip aperture and the time-to-peak grip aperture
velocity increased in the presence of obstacles.
Therefore, level of difficulty influenced the performance
of reaching and grasping components in older adults.
When considering the reaching component, older adults
presented differences compared to young adults (Rinaldi
and Moraes, 2015). Young adults had an increased
movement time and an earlier time-to-peak wrist velocity
only for the most difficult condition (unstable base with
obstacles at short distance). It was observed that the type
of base and presence of obstacles influenced reaching
movement of older adults, but not young adults. The
results of the present study are in agreement with studies

that demonstrated that aging modified the performance of
goal-directed movements with a decrease in movement
speed (Ren et al., 2013) and increase in movement time
(Roy et al., 1996). Changes in reaching control of older
adults due to manual task level of difficulty can be thought
of as a conservative strategy to achieve the task successfully. Grasping is a complex action which requires the
involvement of higher level motor control structures
(Cicerale et al., 2014) and it has been shown that older
adults present a reduction in neural activity (Beurskens
et al., 2014), consequently grasping becomes impaired,
because both stationary and walking tasks overloaded
the central executive areas of the brain.
Finally, movement time, peak wrist velocity and peak
grip aperture velocity were greater for the stationary
condition than for the walking task. The walking task
was more difficult for older adults, because of the
challenge to body stability, leading to the adoption of a
conservative strategy for prehension control. In addition,
time-to-peak grip aperture occurred earlier in the
stationary condition than in the walking condition. This
result indicated that walking changed the control of
grasping, thus older adults allocated more attention for
dynamic stability control. Consequently, the time to
adjust the hand configuration decreased, which
indicates a prioritization of the primary task (walking).

656

Limitations

657

One limitation of the present study is that the starting point
of walking and the location of the object were always the
same throughout the experiment. This design could
facilitate an anticipatory control strategy, because
participants could pre-plan their walking behavior before
starting each trial. For future studies, we are already
planning additional experiments were we will manipulate
the starting point and the starting limb. Moreover, the
prehension task represents a challenge for the walking
stability, and during the approaching phase, older adults
adopted a conservative walking strategy to perform
reaching and grasping successfully, based on
anticipatory control. However, for future studies, we are
planning to have more unpredictable environments,
such that participants will not know when and where
they have to grasp an object. In this case, it will be
possible to investigate how the motor performance and
dynamic stability control is changed when the action
needs to be adjusted on-line.

617
618
619
620
621
622
623
624

625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654

658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675

11

CONCLUSION

676

Our results showed that FOA exhibited a more
conservative walking strategy. They also decoupled the
combined task when compared to OA, since they need
to increase body stability to be able to perform grasping

successfully. History of falls affected the prehension
movement. Our combined task allowed for identification
of changes in the motor control strategies adopted by
FOA for grasping while walking. In addition, motor
patterns (walking and grasping) of older adults (fallers
and non-fallers) were mutually modified due to the level
of difficulty of the manual task.
This study manipulated a combined task that is widely
used by older adults during daily life activities. The older
adults with a history of falls presented important
changes in their action strategies probably in order to
divide attention between the two motor tasks, which
could impair their performance of daily life activities.
Interestingly, our results show modifications in the
prehension control of FOA. This suggests that it may be
possible to use manual tasks as an assessment tool for
fall risk prediction. However, future studies need to
address this more careful and deeply. Based on these
findings and the fact that prehension movement is
widely used during daily life activities, we suggest that
preventive and rehabilitation programs should also
emphasize movement exercises that could improve the
control of upper limbs, especially while performing
locomotor tasks.

677

UNCITED REFERENCE

705

Lin and Liao (2011).

678
679
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694
695
696
697
698
699
700
701
702
703
704

706

REFERENCES

707

Barak Y, Wagennar RC, Holt KG (2006) Gait characteristics of elderly
people with a history of falls: a dynamic approach. Phys Ther 86
(11):1501–1510.
Beauchet O, Annweiler C, Dubost V, Allali G, Kressig RW,
Bridenbaugh S, Berrut G, Assal F, Herrmann (2009) Stops
walking when talking: a predictor of falls in older adults? Eur J
Neurol 16:786–795.
Beurskens R, Helmich I, Rein R, Bock O (2014) Age-related changes
in prefrontal activity during walking in dual-task situations: a fNIRS
study. Int J Psychophysiol 92(3):122–128.
Carnahan H, McFadyen BJ, Cockell DL, Halverson AH (1996) The
combined control of locomotion and prehension. Neurosci Res
Commun 19(2):91–100.
Cebolla EC, Rodacki AL, Bento PCB (2015) Balance, gait,
functionality and strength: comparison between elderly fallers
and non-fallers. Braz J Phys Ther 19(2):146–151.
Chapman GJ, Hollands MA (2007) Evidence that older fallers
prioritise the planning of future stepping actions over the
accurate execution of ongoing steps during complex locomotor
tasks. Gait Posture 26:59–67.
Chen T, Chang CC, Chou LS (2013) Sagittal plane center of mass
movement strategy and joint kinetics during sit-to-walk in elderly
fallers. Clin Biomech 28:807–812.

Chu YH, Tang PF, Peng YC, Chen HY (2013) Meta-analysis of type
and complexity of a secondary task during walking on the
prediction of elderly falls. Geriatr Gerontol Int 13:289–297.
Cicerale A, Ambron E, Lingnau A, Rumiati RI (2014) A kinematic
analysis of age-related changes in grasping to use and grasping
to move common objects. Acta Psychol 151:134–142.

708
709
710
711
712
713
714
715
716
717
718
719
720
721
722
723
724
725
726
727
728
729
730

731
732
733
734
735
736

Please cite this article in press as: Rinaldi NM, Moraes R. Older adults with history of falls are unable to perform walking and prehension movements
simultaneously. Neuroscience (2015), />

NSC 16794

No. of Pages 12

29 December 2015

12
737
738
739
740
741
742
743
744
745
746
747
748
749

750
751
752
753
754
755
756
757
758
759
760
761
762
763
764
765
766
767
768
769
770
771
772
773
774
775
776
777
778
779

780
781

828
829
830

N. M. Rinaldi, R. Moraes / Neuroscience xxx (2015) xxx–xxx

Cohen J (1992) A power primer. Psychol Bull 112(1):155–159.
Delbaere K, Crombez G, Vanderstraeten G, Willems T, Cambier D
(2004) Fear-related avoidance of activities, falls and physical
frailty. A prospective community-based cohort study. Age Ageing
33(4):368–373.
Diermayr G, Mcisaac TL, Gordon AM (2011) Finger force
coordination underlying object manipulation in the elderly – a
mini-review. Gerontology 57:217–227.
Folstein MF, Folstein SE, Mchugh PR (1975) ‘‘Mini-mental state”. A
practical method for grading the cognitive state of patients for the
clinician. J Phychiatr Res 12(3):189–198.
Gorniak SL, Zatsiorsky VM, Latash ML (2011) Manipulation of a
fragile object by elderly individuals. Exp Brain Res 212:505–516.
Hall CD, Echt KV, Wolf SL, Rogers WA (2011) Cognitive and motor
mechanisms underlying older adults’ ability to divide attention
while walking. Phys Ther 91:1039–1050.
Hausdorff JM, Rios DA, Edelberg HK (2001) Gait variability and fall
risk in community-living older adults: a 1-year prospective study.
Arch Phys Med Rehabil 82:1050–1056.
Hawkes TD, Siu KC, Silsupadol P, Woollacott MH (2012) Why does
older adults’ balance become less stable when walking and

performing a secondary task? Examination of attentional
switching abilities. Gait Posture 35:159–163.
Hof AL, Gazendam MGJ, Sinke WE (2005) The condition for dynamic
stability. J Biomech 38:1–8.
Huang MH, Brown SH (2015) Effects of task context during standing
reach on postural control in young and older adults: a pilot study.
Gait Posture 41:276–281.
Kirkwood RN, Moreira BS, Vallone MLDC, Mingoti SA, Dias RC,
Sampaio RF (2011) Step length appears to be a strong
discriminant gait parameter for elderly females highly concerned
about falls: a cross-sectional observational study. Physiotherapy
97:126–131.
Lin SI, Liao CF (2011) Age-related changes in the performance of
forward reach. Gait Posture 33:18–22.
Maia AC, Rodrigues-de-Paula F, Magalha˜es LC, Teixeira RL (2013)
Cross-cultural adaptation and analysis of the psychometric
properties of the Balance Evaluation Systems Test and
MiniBESTest in the elderly and individuals with Parkinson’s
disease: application of the Rasch model. Braz J Phys Ther 17
(3):195–217.
Menant JC, Schoene D, Sarofim M, Lord SR (2014) Single and dual
task tests of gait speed are equivalent in the prediction of falls in
older people: a systematic review and meta-analysis. Ageing Res
Rev 16:83–104.

Nordin E, Moe-nilssen R, Ramnemark A, Lundin-olsson L (2010)
Changes in step-width during dual-task walking predicts falls. Gait
Posture 32:92–97.
O’loughlin J, Robbitaille Y, Boivin J, Suissa S (1993) Incidence of and
risk factors for falls and injurious falls among community-dwelling

elderly. Am J Epidemiol 137:342–354.
Perracini M, Ramos LR (2002) Fall-related factors in a cohort of
elderly community residents. Rev Saude Publica 36(6):709–716.
Ren J, Wu YD, Chan JS, Yan JH (2013) Cognitive aging affects motor
performance and learning. Geriatr Gerontol Int 13(1):19–27.
Rinaldi NM, Moraes R (2015) Gait and reach-to-grasp movements
are mutually modified when performed simultaneously. Hum Mov
Sci 40:38–58.
Roy EA, Weir PL, Leavitt JL (1996) Constraints on prehension: a
framework for studying the effects of aging. Adv Psychol Res
114:279–314.
Springer S, Giladi N, Peretz C, Yogev G, Simon ES, Hausdorff JM
(2006) Dual-tasking effects on gait variability: the role of aging,
falls, and executive function. Mov Disord 21(7):950–970.
Stelmach GE, Zelaznik HN, Lowe D (1990) The influence of aging
and attentional demands on recovery from postural instability.
Aging (Milano) 2:155–161.
Tideiksaar R (1996) Preventing falls: how to identify risk factors,
reduce complications. Geriatrics 51:43–55.
Toebes MJP, Hoozemans MJM, Furrer R, van Dieen JH (2012) Local
dynamic stability and variability of gait are associated with fall
history in elderly subjects. Gait Posture 36:1050–1056.
Uemura K, Yamada M, Nagai K, Tanaka B, Mori S, Ichihashi N (2012)
Fear of falling is associated with prolonged anticipatory postural
adjustment during gait initiation under dual-task conditions in
older adults. Gait Posture 35:282–286.
Verghese J, Kuslansky G, Holtzer R, Katz M, Xue X, Buschke H,
Pahor M (2007) Walking while talking: effect of task prioritization
in the elderly. Arch Phys Med Rehabil 88:50–53.
Viswanathan A, Sudarky L (2011) Balance and gait problems in the

elderly. Handb Clin Neurol 103:623–634.
Voorrips LE, Ravelli AC, Dongelmans PC, van Staveren WA (1991) A
physical activity questionnaire for the elderly. Med Sci Sports
Exerc 29:117–121.
Winter DA (2005) Biomechanics and motor control of human
movement. Hoboken, New Jersey: Wiley. 384.
Yamada M, Aoyama T, Nakamura M, Tanaka B, Nagai K, Tatematsu
N, Uemura K, Nakamura T, Tsukoyama T, Tchihaski N (2011)
The reliability and preliminary validity of game-based fall risk
assessment in community-dwelling older adults. Geriatr Nurs 32
(3):188–194.

(Accepted 20 December 2015)
(Available online xxxx)

Please cite this article in press as: Rinaldi NM, Moraes R. Older adults with history of falls are unable to perform walking and prehension movements
simultaneously. Neuroscience (2015), />
782
783
784
785
786
787
788
789
790
791
792
793
794

795
796
797
798
799
800
801
802
803
804
805
806
807
808
809
810
811
812
813
814
815
816
817
818
819
820
821
822
823
824

825
826
827