The interacting effect of cognitive and motor task demands on performance of [625709]

The interacting effect of cognitive and motor task demands on performance of
gait, balance and cognition in young adults
Tony Szturma,*, Pramila Maharjana, Jonathan J. Marottab,1, Barbara Shaya, Shiva Shresthaa,
Vedant Sakhalkara
aDepartment of Physical Therapy, School of Medical Rehabilitation, University of Manitoba, Winnipeg, MB, Canada
bDepartment of Psychology, University of Manitoba, P310 Duff Roblin Bldg, Canada
1. Introduction
Successful aging has become one of the most important aspects
of health care in the 21st century. As people live longer risks of
cumulative illness, chronic disability increase [1,2] . Mobility
limitations and cognitive impairments, both common with aging,
reduce levels of physical and mental activity, are prognostic of
future adverse health events, and are associated with an increased
fall risk [2]. Importantly, the link between cognitive impairment,
mobility limitations and the tendency to falls is recognized in the
literature [3].
Maintaining stability during walking through the environment
is a complex, multi-dimensional process requiring higher levelmotor control, and cognitive flexibility to address balance threats,
while attending to environmental demands and concurrent
cognitive tasks [2]. A key factor in locomotor control is executive
cognitive functioning and deficits are associated with increased
risk of falling [3,4] . Various dual task (DT) studies have affirmed
that difficulty in assigning attention to each task simultaneously
may contribute significantly to increased fall risks. Poor DT
performance in either the motor or cognitive task could be caused
by altered prioritization between the two tasks [5]. The most
common and consistent finding of DT studies has been the
reduction of gait speed [3], likely as a strategy for concurrent task
processing or to avoid stability threat. Reduced speed is commonly
observed in elderly, and when negotiating obstacles, irregular or
unpredictable terrain [6].
Dual-task studies have utilized cognitive tasks, like animal
enumeration or number subtraction that are typically only
assessed qualitatively, do not involve the visuomotor system
and are limited in recruitment of individual brain areas. Visual–
spatial processing of object locations/motions and their spatial
relations with respect to body and space are key aspects of balance
and locomotor skills, and evidence supports visual–spatial
processing as an important aspect of cognition to explore in
mobility decline [7,8] .Gait & Posture 38 (2013) 596–602
A R T I C L E I N F O
Article history:
Received 8 August 2012
Received in revised form 29 January 2013
Accepted 1 February 2013
Keywords:
Dual-taskTreadmill walking
StabilityLocomotor rhythm
Cognitive performanceA B S T R A C T
Mobility limitations and cognitive impairments, each common with aging, reduce levels of physical and
mental activity, are prognostic of future adverse health events, and are associated with an increased fall
risk. The purpose of this study was to examine whether divided attention during walking at a constant
speed would decrease locomotor rhythm, stability, and cognitive performance. Young healthy
participants (n = 20) performed a visuo-spatial cognitive task in sitting and while treadmill walking
at 2 speeds (0.7 and 1.0 m/s).Treadmill speed had a significant effect on temporal gait variables and ML-
COP excursion. Cognitive load did not have a significant effect on average temporal gait variables or COP
excursion, but variation of gait variables increased during dual-task walking. ML and AP trunk motion
was found to decrease during dual-task walking. There was a significant decrease in cognitive
performance (success rate, response time and movement time) while walking, but no effect due to
treadmill speed. In conclusion walking speed is an important variable to be controlled in studies that are
designed to examine effects of concurrent cognitive tasks on locomotor rhythm, pacing and stability.
Divided attention during walking at a constant speed did result in decreased performance of a visuo-
spatial cognitive task and an increased variability in locomotor rhythm.
/C223 2013 Elsevier B.V. All rights reserved.
* Corresponding author at: Department of Physical Therapy, School of Medical
Rehabilitation, R106 – 771 McDermot Avenue, Winnipeg, MB R3E 0T6, Canada.
Tel.: +1 204 787 4794; fax: +1 204 787 1227.
E-mail addresses: tony.szturm@med.university.ca ,
tony.szturm@med.umanitoba.ca (T. Szturm), maharjan.pramila@gmail.com
(P. Maharjan), jonathan_marotta@me.com (J.J. Marotta),
barb.shay@med.umanitoba.ca (B. Shay), drshivashrestha@gmail.com (S. Shrestha),
vedant.sak@gmail.com (V. Sakhalkar).
1Tel: +1 204 474 7057.Contents lists available at SciVerse ScienceDirect
Gait & Posture
jo u rn al h om ep age: ww w.els evier.c o m/lo c ate/g aitp os t
0966-6362/$ – see front matter /C223 2013 Elsevier B.V. All rights reserved.
http://dx.doi.org/10.1016/j.gaitpost.2013.02.004

Virtual environments, viewed during treadmill walking, have
been used as an ecological approach to rehabilitation [9].
Computerized cognitive tasks and games have received interest
from researchers and clinicians, both as a model for learning a
broad range of cognitive tasks and as a means to examine training
and transfer of skills to daily life activities [10–12] . A treadmill
rehabilitation platform (TRP) was designed around a treadmill as it
is an excellent choice for conducting gait training with dual-tasks.
It can incorporate walking skills while interacting with computer-
generated cognitive activities viewed on a standard LCD display
[9]. DT treadmill walking has important advantages versus over
ground walking as gait variables are significantly influenced by
walking speed [13,14] and reduced gait speed is a highly consistent
strategy used during dual-task over-ground walking [3]. It is a
convenient method to determine steady-state walking speed. It
also allows gathering hundreds of consecutive steps in a few
minutes. Data from 5–10 strides (i.e. in gait laboratories or during
repeated walks over short, instrumented walkways) may reliably
measure gait speed, but is not sufficient for measures of gait
variability or periodicity, particularly during dual task walking and
for older adults with mobility limitations [15,16] .
The purpose of this study was to further explore the interplay
between cognitive and walking demands on task performance. Since
previous studies have shown that gait speed is an important factor
affecting gait parameters, the treadmill speed is held constant to
prevent a strategy of slowing walking speed. The first objective was
to evaluate the effect of walking speed on temporal gait parameters
and measures of walking stability, amplitude and variation of center
of foot pressure (COP) displacements and trunk motion. The
objective was to examine whether divided attention during walking
at a constant speed would decrease locomotor rhythm, stability, and
cognitive performance. This study addresses three hypotheses:
(1) Walking speed has a significant effect on temporal gait
variables, and measures representing walking stability.(2) Stability, locomotor, and cognitive performance will signifi-
cantly decline from single task to DT conditions during
constant speed.
(3) Cognitive performance will decline with increasing treadmill
speed.
2. Methods
Twenty healthy young adults aged 20–30 years (mean age 26.3 /C6 3.2 years)
participated. Participants were excluded if they had past neurological impairment,
musculoskeletal disorder or were taking medications that may have influenced their
walking.
3. Instrumentation and data recording
Fig. 1 illustrates the experimental set-up. Participants were
positioned on the treadmill 100 cm from the 30-inch monitor
connected to a computer running the cognitive game. Vertical foot
contact pressures were recorded from each foot using in-shoe
pressure insoles. (Vista Medical Ltd, WPG. MB). The pressure
insoles each consist of an array of 128 piezo-resistive sensors,
calibrated to 300 mm Hg (12-bit). Pressure signals from left and
right insoles were recorded at 35 Hz. The 3D Track STAR (Ascension
Tech, Burlington, VT, US) was used to record the position of the
trunk (80 Hz). The track STAR sensor was secured to the skin at the
second thoracic spinal process. A commercial motion mouse
(Gyration Air Mouse, USA) was secured to a head band and used as
the computer input device to control on-screen cursor motion with
head rotation (left–right). This Air Mouse has inertial sensors used
to derive angular position signals. With this simple method,
seamless and responsive hands-free interaction with the computer
application is made possible. In a similar manner, a number of
studies have used reaching or pointing tasks to evaluate
perception, attention, and higher-level cognitive decision-making
[17,18] . Visually guided head movements are among the most
natural, therefore these tasks are easily performed with minimal
Fig. 1. Experimental set-up. Participant is shown walking on treadmill while viewing a computer monitor and using head rotation (motion mouse) to interact with cognitive
game. Panel B presents trajectory of game paddle movements for one logged game file, 120 s duration. Each game event is 2 s in length; thus, recording includes a total of 60
game events (random presentation of different amplitudes and directions). Panel C presents overlay of individual game events segmented based on index times of target
appearance and disappearance Time zero is onset of target appearance, end of event is time when target disappears. Panel D: segmented game events shown in D are sorted
grouped in functional bins, which in this case represent medium amplitude player movements in leftward direction (upward trajectories), and rightward direction
(downward trajectories). Panel E illustrates analysis methods to quantify response time and movement time.T. Szturm et al. / Gait & Posture 38 (2013) 596–602 597

instruction. Computer controlled goal-directed movements can
provide an easily accessible way to track a wide range of cognitive
events while walking.
4. Cognitive game task
Studies have used computer-based games to probe and
evaluate cognitive function [11]. The Useful Field of View (UFOV)
is a computer-based test that requires the ability to select relevant
information and ignore irrelevant information (cognitive inhibi-
tion) [19]. Studies have found that older adults with slower
cognitive speed of processing, as measured by the UFOV test,
experienced the greatest mobility loss [20]. A modified version of
the UFOV has been designed to evaluate visual–spatial processing
together with eye-head coordination. The goal of the test game is
to move a paddle (game sprite) to catch falling bright circle objects
(targets) moving vertically top to bottom, and to avoid triangle
shaped objects (distracters). The objects appear at fixed intervals
(2 s) and at random locations on the monitor. The game is
instrumented with an assessment module. This generates a logged
game file recording (80 Hz) the following signals associated with
player performance with respect to game events: (a) time index
and coordinates of each object and (b) position coordinates of the
game paddle (slaved to head rotation). Fig. 1C, presents trajectory
of head rotation of one game file (obtained in standing), and Fig. 1D
shows all segmented game events within one trial. These
contextual game events are sorted by direction and amplitude
to obtain multiple event groupings with similar movement
features (Fig. 1E). For a full description see Lockery et al. [21].
5. Protocol
Participants played the computer game using a standard optical
mouse for 2 min in sitting to familiarize themselves with the
cognitive task. The viewing height of the display during sitting and
walking was maintained by placing an adjustable stool on the
treadmill. A rest period of 2–3 min was given between test
conditions. Participants walked on a level treadmill for 10 min at
0.7 m/s for treadmill acclimation. During testing, participants
walked for 2 min at two treadmill speeds; 0.7 m/s (lower speed)
and 1.0 m/s (higher speed), singly, while performing the cognitive
game task. The order of treadmill speed, single and DT conditions
were randomized within a session to minimize potential order
effect. The cognitive tasks were also performed while standing on
the stationary treadmill (single task condition).6. Data analysis
Custom built scripts in MATLAB version 7.1 (The Math Works,
Natick, MA) processed the pressure data of each insole array into
footfall patterns. Time indices were computed for pressure onset
and offset, stance and swing phases for each right and left step, and
double support times. The average and coefficient of variation
(COV) of stance time (ST), swing time (SW), and double support
time (DS) were computed for each walk trial (45 steps per leg).
These gait variables have been identified based on associations
with falls, cognitive impairment [1] and balance impairments [22].
Center of pressure excursions in the anterior-posterior (AP) and
medial–lateral (ML) directions were calculated by summing the
contact forces recorded from each insole sensor. Based on the time
indices for stance onset and offset, the COP time’s series data were
segmented into individual right and left stance phases for each
step. The segmented stance COP trajectories were time normalized
and the average COP trajectory and standard deviation across the
trial was computed for each foot. Fig. 2 presents overlay of typical
time normalized COP trajectories for steps of one walk trial at each
speed. Traces are offset to a common baseline value of zero for
display purpose. Thickened lines represent ensemble average and
dotted lines standard deviation. Also presented are scatter plots
(means and SD) of stance and swing time for each of the 45 steps
obtained. The following variables were computed from the average
COP trajectory; (a) peak-to-peak COP displacement, (b) root mean
square (RMS), and (c) total path length (TLP) [23,24] . In addition
variance in step to step COP trajectories over each walk trial was
computed, defined as the average of the COP standard deviation.
Peak-to-peak amplitude and RMS of AP and ML trunk position data
were computed. Control of the motion of the trunk is a main
contributor to overall stability during walking [25,26] .
Different features of the test game events provide a basis for
objective quantification of cognitive functions. As illustrated in
Fig. 1E the following variables were determined: (1) game success
rate (percentage of target caught), (2) average motor response time
(time from appearance of the target to start of the paddle
movement), (3) average movement execution time (time between
movement initiation and final paddle position).
7. Statistical analysis
A two-way repeated measures ANOVA was used to determine
the effects of treadmill speed and cognitive load (single vs. DT
conditions) on temporal gait variables, COP and trunk excursion
Fig. 2. Overlay plots of segmented time normalized AP-COP and ML-COP trajectories of each stance phase in one walk trial. Trajectories are offset to a common baseline value
of zero for display purpose. Thickened lines represent ensemble average and dotted lines standard deviation. Panel C presents typical plots of segmented game events of
medium leftward movements from one game trial obtained during walking. Top row are for 0.7 m/s treadmill speed and bottom for 1.0 m/s.T. Szturm et al. / Gait & Posture 38 (2013) 596–602 598

measures, and cognitive performance measures. The significance
level was set at alpha level of 0.5.
8. Results
Group means and standard error of means (SEM) for average and
COV of ST, SW, and DS are presented in Fig. 3. There was no
significant difference in gait variables (average or COV) between left
and right steps and therefore only results of analysis of right steps is
presented in Table A1 (Appendix A). Average and COV of ST, SW and
DS significantly decreased as a function of walking speed (p < 0.01).
There was no significant effect of cognitive load for walk only versus
DT walking on average ST, SW, or DS. In contrast COV of ST, SW, and
DS significantly increased from single to DT conditions (p < 0.01).
Group means (SEM) of COP peak-to-peak amplitude, RMS and
TPL are presented in Fig. 4. As presented in Table A2 (Appendix A)
all variables of COP excursion in ML direction significantly
increased with increasing treadmill speed (peak to peak:
p < 0.02; RMS: p < 0.01; TPL: p < 0.01). However there was no
significant effect of treadmill speed on AP-COP excursion variables.
As seen in Fig. 5 there was a significant effect of walking speed on
the variation of COP trajectories across steps within a trial. The
average standard deviation of ML-COP trajectories increased with
increasing treadmill speed (p < 0.01) but the opposite was
observed for AP-COP (p < 0.01). Cognitive load at either treadmill
speed had no significant effect on either COP excursion variables,
or average COP standard deviation.
Group means (SEM) of peak-to-peak amplitude and RMS of
trunk horizontal translation is shown in Fig. 5. As presented in
Table A2 (Appendix A) there was no significant effect of walking
speed on magnitude of linear trunk excursion in either AP or ML
direction. In contrast, peak-to-peak amplitude and RMS of trunk
horizontal translation in ML and AP directions were found to
significantly decrease while performing the cognitive tasks
compared to walking alone (p < 0.01).
Fig. 2D presents typical plots of segmented game movements
obtained at both speeds compared to standing (Fig. 1E). There was
a significant decrease in success rate (88–65%; p < 0.001; F(2, 18),25), and a significant increase in; (a) response time (410–680 ms;
p < 0.001; F(2, 18), 35), and (b) execution time (510–580 ms;
p < 0.01; F(2, 18), 16), while walking compared to standing. There
was no significant difference in these variables between the two
treadmill speeds.
9. Discussion
Treadmill speed had a significant effect on temporal gait
variables (average and COV) in keeping with previous studies. The
present results are consistent with Jordan et al. [13] and Kang and
Dingwell [27] who observed a decrease in gait variability with
increased walking speed. Measures of variability provide a
perspective on the consistency of locomotor rhythm, and are
often reported to represent walking stability. This view is
supported by the present findings wherein the magnitude and
variation in ML-COP displacements were influenced by treadmill
speed. Other studies have reported that walking speed affects
dynamic stability measures (e.g. Lyapunov exponents based on
trunk velocity/acceleration) [16,28] . To note AP-COP excursion was
not affected by treadmill speed, and in fact variation of AP-COP
trajectories across the 45 consecutive steps decreased with
increasing walking speed (as well as decreased stance/swing
duration). This may reflect the consistency of heel contact and a
more constant AP path length to toe off.
We hypothesized that an increase in walking speed would
result in an increase in trunk motion. However, the present
results did not show any significant effect of speed on linear ML
or AP trunk motion. In contrast to this finding, Kavanagh [25]
found a significant increase in ML trunk motion with increasing
over ground walking speed. The difference in these findings
might be explained by the difference in walking speeds or may
reflect a consequence of treadmill walking. In the present study
speeds were 0.7–1.0 m/s, whereas, in the Kavanagh study it was
much greater (0.9–1.7 m/s). The treadmill’s safety rails, front
panel and monitor provided stationary visual cues that could
help orientate and stabilize the body location and trunk position
during walking.
Fig. 3. Presents group means and standard error of means (SEM) of trial average and COV of ST, SW, and DS at the two speeds, walk-only and DT walks conditions.T. Szturm et al. / Gait & Posture 38 (2013) 596–602 599

The present findings demonstrate that ML and AP trunk motion
was reduced while tracking and interacting with moving images
on a monitor. Dingwell et al. [29] examined the effect of
performing a visual Stroop task on trunk motion during treadmill
walking. Variability of trunk velocity in all three directions
decreased during the DT condition. Similar findings have been
reported by Doi et al. [26] that ML trunk acceleration was
significantly decreased during DT over ground walking using a
color Stroop task with images projected onto the wall. A decrease
in trunk motion would minimize head motion and thus help to
stabilize gaze while tracking and interacting with moving targets
or when reading words. In this regard Lambert et al. [30] has shown
that visual acuity does not decrease during treadmill walking in
young healthy participants, although visual acuity decreased
significantly in patients with a peripheral vestibular loss.
In the present study the head rotational pointing movements
used to interact with the visual targets were ramp movements,
duration approximately 600 ms, and amplitudes less than 30.
Studies which have examined tracking visual targets (up to 258)
during treadmill walking using eye movements only compared to
eye-head and trunk rotation show little lateral deviation of the COP
from the heading direction when they performed the tracking task
with eye or head rotation, whereas, trunk rotations led to a doubling
of ML-COP deviation [24]. Also of note with respect to the present
cognitive outcome measures, response time was found to increase
during walking. The head is stationary during this time period i.e.
from initial appearance of objects to start of head movement.
Thus far, most DT walking studies have been limited to a single
cognitive performance indicator (example correct response
number) while performing the cognitive tasks during a walk of
a few meters on a walkway. In most of these studies, both walking
speed and cognitive performance decreased (for review see Al
Yahya et al. [3]). In the present study during DT walking it took
Fig. 4. Present group means and (SEM) of peak-to-peak amplitude, RMS and path length of COP displacement at the two speeds, walk only and DT walks conditions. Bottom
right histogram presents group means (SEM) of the average SD of COP trajectories within a trial.
Fig. 5. Presents group means (SEM) of peak-to-peak amplitude and RMS of trunk
linear displacement at the two speeds, walk only and DT walks conditions.T. Szturm et al. / Gait & Posture 38 (2013) 596–602 600

longer to initiate movements to specified targets (presence of
distracters), movement times to reach the final target position
were longer, and the number of targets caught decreased.
Cognitive load also did not have a significant effect on average
temporal gait variables or magnitude of COP excursion. At first
glance this appears different from the previous research findings,
but given a constant treadmill speed it would be expected that
average values of temporal gait variables would be consistent
among conditions performed at the same speed. A significant
increase in variability of stance time, swing time, and double limb
support time during DT walking was seen, consistent with results
of recent studies [3]. Analysis of gait variability (e.g. standard
deviations) only quantifies the average magnitude of differences
across all strides, regardless of temporal order. Other analyses
reflective of dynamic stability include Lyapunov exponents and
Floquet multipliers. These measures have been shown to be
sensitive to change in treadmill walking speed and DT conditions
[16,27] . These methods however require large numbers of
consecutive steps (i.e. hundreds) therefore are often conducted
during treadmill walking. In the present study 45 steps, using
2 min of walking data, were collected.
The application of computer tasks can provide a broad range of
executive cognitive functions. Employing computer tasks, and
parsing subjects’ actions and choices can provide recordings of
multiple contextual events. This permits quantification of process
measures in addition to objective measures of cognitive perfor-
mance [17]. Thus one can make principled comparisons of the
influence that cognitive demands have on stability, gait and fall
risk. This will provide a better understanding of the functional
consequences of decline in physical and mental skills with age and
in early stages of disease, and help in making choices for
prevention, treatments, and lifestyle decisions.
In conclusion, walking speed is an important variable to be
controlled in studies that are designed to examine effects of
concurrent cognitive tasks on variables representing locomotor
rhythm, pacing and stability. Divided attention during walking at a
constant speed did result in decreased performance of a visuo-
spatial tracking task, an increased variability in locomotor rhythm.
It also appears that gaze control is an important priority during DT
tasks that depend on processing of visual information.
Funding
None.
Conflict of interest
There is no conflict of interest.
Appendix AReferences
[1] Santos-Eggimann B, Karmaniola A, Seematter-Bagnoud L, Spagnoli J, Bula C,
Cornuz J, et al. The lausanne cohort Lc65+: a population-based prospective
study of the manifestations, determinants and outcomes of frailty. BMC
Geriatrics 2008;8:20.
[2] Buchman AS, Boyle PA, Leurgans SE, Barnes LL, Bennett DA. Cognitive function
is associated with the development of mobility impairments in community-
dwelling elders. The American Journal of Geriatric Psychiatry 2011;19(6):
571–80.
[3] Al-Yahya E, Dawes H, Smith L, Dennis A, Howells K, Cockburn J. Cognitive
motor interference while walking: a systematic review and meta-analysis.
Neuroscience and Biobehavioral Reviews 2011;35(3):715–28.
[4] Hall CD, Echt KV, Wolf SL, Rogers WA. Cognitive and motor mechanisms
underlying older adults’ ability to divide attention while walking. Physical
Therapy 2011;91(7):1039–50.
[5] Hobert MA, Niebler R, Meyer SI, Brockmann K, Becker C, Huber H, et al. Poor
trail making test performance is directly associated with altered dual task
prioritization in the elderly – baseline results from the TREND study. PloS ONE
2011;6(11):e27831.
[6] Verghese J, LeValley A, Hall CB, Katz MJ, Ambrose AF, Lipton RB. Epidemiology
of gait disorders in community-residing older adults. Journal of the American
Geriatrics Society 2006;54(2):255–61.
[7] Murray ME, Senjem ML, Petersen RC, Hollman JH, Preboske GM, Weigand SD,
et al. Functional impact of white matter hyperintensities in cognitively normal
elderly subjects. Archives of Neurology 2010;67(11):1379–85.
[8] Bagurdes LA, Mesulam MM, Gitelman DR, Weintraub S, Small DM. Modulation
of the spatial attention network by incentives in healthy aging and mild
cognitive impairment. Neuropsychologia 2008;46(12):2943–8.
[9] Feasel J, Whitton MC, Kassler L, Brooks FP, Lewek MD. The integrated virtual
environment rehabilitation treadmill system. IEEE Transactions on Neural
Systems and Rehabilitation Engineering 2011;19(3):290–7.
[10] Rand D, Weiss PL, Katz N. Training multitasking in a virtual supermarket: a
novel intervention after stroke. The American Journal of Occupational Therapy
Official Publication of the American Occupational Therapy Association
2009;63(5):535–42.
[11] Jimison HB, McKanna J, Ambert K, Hagler S, Hatt WJ, Pavel M. Models of
cognitive performance based on home monitoring data. IEEE Engineering in
Medicine and Biology Society Conference 2010;2010:5234–7.
[12] Burgess PW, Alderman N, Forbes C, Costello A, Coates LM, Dawson DR, et al.
The case for the development and use of ‘‘ecologically valid’’ measures of
executive function in experimental and clinical neuropsychology. Journal of
the International Neuropsychological Society JINS 2006;12(2):194–209.
[13] Jordan K, Challis JH, Newell KM. Walking speed influences on gait cycle
variability. Gait & Posture 2007;26(1):128–34.
[14] Stoquart G, Detrembleur C, Lejeune T. Effect of speed on kinematic, kinetic,
electromyographic and energetic reference values during treadmill
walking. Neurophysiologie Clinique Clinical Neurophysiology 2008;38(2):
105–16.
[15] Hollman JH, Childs KB, McNeil ML, Mueller AC, Quilter CM, Youdas JW.
Number of strides required for reliable measurements of pace, rhythm and
variability parameters of gait during normal and dual task walking in older
individuals. Gait & Posture 2010;32(1):23–8.
[16] Bruijn SM, van Dieen JH, Meijer OG, Beek PJ. Statistical precision and sensitiv-
ity of measures of dynamic gait stability. Journal of Neuroscience Methods
2009;178(2):327–33.
[17] Song JH, Nakayama K. Hidden cognitive states revealed in choice reaching
tasks. Trends in Cognitive Sciences 2009;13(8):360–6.
[18] Chapman CS, Gallivan JP, Wood DK, Milne JL, Culham JC, Goodale MA. Reaching
for the unknown: multiple target encoding and real-time decision-making in a
rapid reach task. Cognition 2010;116(2):168–76.
[19] Ball K, Edwards JD, Ross LA. The impact of speed of processing training on
cognitive and everyday functions. Journal of Gerontology Series B Psychologi-
cal Sciences and Social Sciences 2007;62:19–31.
[20] Wood JM, Chaparro A, Lacherez P, Hickson L. Useful field of view predicts
driving in the presence of distracters. Optometry and Vision Science 2012;89(4):
373–81.Table A1
Results of analysis of variance: effect of speed and DT on gait parameters.
F-statistics (df).
Averages Coefficient of variance
(COV)
Speed DT Speed DT
Stance time p < 0.01
F(2, 18) 48.5NS p < 0.01
F(2, 18) 30.6p < 0.01
F(2, 18) 27.3
Swing time p < 0.01
F(2, 18) 19.5NS p < 0.01
F(2, 18) 6.25p < 0.01
F(2, 18) 18.5
Double support time p < 0.01
F(2, 18) 39.8NS p < 0.01
F(2, 18) 29.3p < 0.01
F(2, 18) 8.5Table A2
Results of analysis of variance, effects of speed and DT on COP parameters and trunk
translations. F-statistics (df).
Center of pressure (COP) Trunk translations
Speed DT Speed DT
AP ML AP ML AP ML AP ML
Pk-Pk NS p < 0.01
F(2, 18) 5.3NS NS NS NS p < 0.01
F(2, 18) 11.2p < 0.01
F(2, 18) 4.3
RMS NS p < 0.01
F(2, 18) 7.2NS NS NS NS p < 0.01
F(2, 18) 21.3p < 0.01
F(2, 18) 12.6
TPL NS p < 0.01
F(2, 18) 15.2NS NST. Szturm et al. / Gait & Posture 38 (2013) 596–602 601

[21] Lockery D, Peters JF, Ramanna S, Shay BL, Szturm T. Store-and-feed forward
adaptive gaming system for hand-finger motion tracking in telerehabilitation.
IEEETransactions onInformationTechnology in Biomedicine 2011;15(3):467–73.
[22] Hollman JH, McDade EM, Petersen RC. Normative spatiotemporal gait param-
eters in older adults. Gait & Posture 2011;34(1):111–8.
[23] Desai A, Goodman V, Kapadia N, Shay BL, Szturm T. Relationship between
dynamic balance measures and functional performance in community-dwell-
ing elderly people. Physical Therapy 2010;90(5):748–60.
[24] Duysens J, Duysens JP, Bastiaanse CM, van Sprundel M, Schubert M, Smits-
Engelsman BC. How trunk turns affect locomotion when you are not looking
where you go. Human Movement Science 2008;27(5):759–70.
[25] Kavanagh JJ. Lower trunk motion and speed-dependence during walking.
Journal of Neuroengineering and Rehabilitation 2009.[26] Doi T, Asai T, Hirata S, Ando H. Dual-task costs for whole trunk movement
during gait. Gait & Posture 2011;33(4):712–4.
[27] Kang HG, Dingwell JB. Separating the effects of age and walking speed on gait
variability. Gait & Posture 2008;27(4):572–7.
[28] Dingwell JB, Marin LC. Kinematic variability and local dynamic stability of
upper body motions when walking at different speeds. Journal of Biomechan-
ics 2006;39(3):444–52.
[29] Dingwell JB, Robb RT, Troy KL, Grabiner MD. Effects of an attention demanding
task on dynamic stability during treadmill walking. Journal of Neuroengineer-
ing and Rehabilitation 2008;5:12.
[30] Lambert S, Sigrist A, Delaspre O, Pelizzone M, Guyot JP. Measurement of
dynamic visual acuity in patients with vestibular areflexia. Acta Oto-Laryn-
gologica 2010;130(7):820–3.T. Szturm et al. / Gait & Posture 38 (2013) 596–602 602