Copyright 1997 by The Gemntological Society of America
Journal of Gerontology: MEDICAL SCIENCES
1997, Vol. 52A, No. 4, M232-M240
The Effects of Two Types of Cognitive Tasks
on Postural Stability in Older Adults
With and Without a History of Falls
Anne Shumway-Cook,1 Marjorie Woollacott,2 Kimberly A. Kerns,3 and Margaret Baldwin1
'Department of Physical Therapy, Northwest Hospital, Seattle.
department of Exercise and Movement Science, University of Oregon.
'Department of Psychology, University of Victoria, British Columbia.
Background. This study used a dual task design to investigate the effects of two different types of cognitive tasks on
stability (as measured by center of pressure displacement) in young vs older adults with and without a history of falls.
Methods. Two secondary cognitive tasks, a sentence completion and a visual perceptual matching task, were used to
produce changes in attention during quiet stance under flat vs compliant surface conditions in 20 healthy young adults,
20 healthy older adults, and 20 older adults with a history of imbalance and falls. Postural stability was quantified
using forceplate measures of center of pressure (COP). Speed and accuracy of verbal response on the cognitive tasks
were also quantified.
Results. During the simultaneous performance of a cognitive and postural task, decrements in performance were
found in the postural stability measures rather than the cognitive measures for all three groups. While no differences
were found between the young adults and the older healthy adults on the firm surface, no task condition, when task
complexity was increased (either through the introduction of a secondary cognitive task, or a more challenging postural condition such as standing on the compliant surface), significant differences in postural stability between the two
groups became apparent. In contrast to the young and healthy older adults, postural stability in older adults with a history of falls was significantly affected by both cognitive tasks.
Conclusion. Results suggest that when postural stability is impaired, even relatively simple cognitive tasks can further impact balance. Results further suggest that the allocation of attention during the performance of concurrent tasks
is complex', depending on many factors including the nature of both the cognitive and postural task, the goal of the
subject and the instructions.
P
REVIOUS research has shown that many aspects of
postural control decline with age, and that postural
deficits are a contributing factor to an increased likelihood
for falls in many older adults (1-10). Several studies have
suggested that decreased balance control either due to injury (11) or aging (12) may increase the attentional requirements associated with maintaining stability. But there has
been little research clarifying the possible mechanism by
which declining postural control requires increased attentional resources, nor the effect of increased attentional resources on the ability to maintain stability during the concurrent performance of postural and cognitive tasks.
It is possible that as adults age, increased attention is
used to compensate for deterioration within a sensory system. Numerous studies have documented age-related
changes in peripheral sensory structures in all three sensory
systems important for postural control (13-18). Researchers have hypothesized that as sensory systems deteriorate with age, increased attention is allocated to
"heighten" the signal coming from these systems in order
to gain the necessary information for postural control
(12,19). Thus, this is one possible mechanism by which declining postural control increases attentional demands in
older adults.
M232
Using a dual-task paradigm, researchers have begun to
study the attentional demands associated with sensory
motor processing for postural control. Several researchers
have documented that even highly practiced tasks such as
sitting, standing, and walking require some cognitive processing, and have found that the degree of processing
varies with the postural task (20,21). Thus, static equilibrium tasks such as sitting or standing quietly appear to require less attention than do dynamic equilibrium tasks such
as walking (20-23).
When the task of maintaining postural stability is done
concurrently with another task, which task will receive a
higher priority with respect to allocation of attentional resources? Researchers have shown that when a focal voluntary movement task is performed during unsupported qujet
stance (a postural stability task), the focal movement is delayed long enough to ensure the activation of anticipatory
postural adjustments which make certain that stability is
preserved (24,25). If external support is provided, reducing
the stability demand of the postural task, the reaction time
for the secondary motor task is shortened (24). This suggests that when there is competition between two motor
tasks, one of which is postural control, the maintenance of
stability is the priority. One might.presume that this same
COGNITIVE DEMANDS AND POSTURAL CONTROL
principle, "posture first," applies in all dual task conditions.
Thus, when a person is engaged in a combination of both
cognitive and postural tasks, attention would be allocated to
postural control ensuring stability and potentially affecting
performance on the cognitive task. Stelmach et al. (12)
found that recovery from postural instability induced by
rapid arm swinging was unaffected by the concurrent performance of a secondary mental arithmetic task in young
adults, but was impaired in elderly subjects. These results
support the "posture first" hypothesis in young adults but
raise the possibility that the priorities for allocating attention during the performance of dual tasks may change as
people age.
Maki and Whitelaw (26) found that postural responses in
both young and older adults were affected by the simultaneous performance of a secondary mental-arithmetic task
because subjects tended to lean slightly forward. Teasdale
et al. (27) found that both young and elderly subjects had
slower reaction times to an auditory stimulus when sensory
information from either vision or somatosensory systems
for postural control was decreased. This research does not
support a "posture first" hypothesis, because the concurrent
performance of a cognitive task did affect postural stability
and this effect was enhanced in older adults. Little is
known about the effect of competing cognitive and postural
demands on attentional resources during the performance
of multiple tasks in older adults with a recent history of
falls. We hypothesize that changes in attention associated
with the performance of concurrent secondary cognitive
tasks would affect postural stability more in older adults
with a history of falls, compared to healthy young and older
adults. We believe that this competition for limited attentional resources during the simultaneous performance of
postural and cognitive tasks contributes to instability and
falls in some older adults.
In order to test this hypothesis, we used a dual task
paradigm to investigate the effects of two types of cognitive
tasks on postural stability in young adults and older adults
with and without a history of falls. The two cognitive tasks
chosen, a language vs a visual spatial task, were designed to
examine the effect of two different types of information processing on postural stability. Two postural tasks included
standing on a firm vs compliant foam surface. Both postural
control and spatial orientation require visual processing;
however, it is not clear whether visual pathways used for information processing are the same for both tasks. We hypothesized that the greatest interference between cognitive
and postural tasks in our experimental paradigm would be
during the performance of a visual spatial orientation task
while standing on a compliant surface because of the competition for visual processing pathways in the two tasks. It
has been suggested that standing on compliant foam decreases the availability of reliable somatosensory cues for
postural control resulting in an increased reliance on visual
and vestibular inputs for maintaining stability (28).
METHODS
Subjects
We tested 20 young adults (mean age 31 ±6), 40 older
M233
adults, 20 with a self-reported history of falls (mean age 78
± 8), and 20 without (mean age 74 ± 6). Subjects were volunteers from the Seattle community who responded to a
newspaper advertisement. Criteria for classifying older
adults as fallers included: self-report of two or more falls in
the previous 6 months in the absence of a known neurological or musculoskeletal diagnosis. We did not systematically
determine the cause of previous falls, hence type of fall was
not used as a criterion for inclusion or exclusion in this
study. Table 1 displays the baseline characteristics for each
of the three groups. The two groups of older adults were
comparable with respect to age, gender, marital status, living situation, and mental status. Performance on the Short
Portable Mental Status Questionnaire (SPMSQ) (29) was
used to determine mental status. Frequency of imbalance,
defined as near falls, slips, trips, or stumbles experienced
by the subject, was determined by self-report.
The decision to recruit and classify older adults based on
prior history of falls was based on the assumption that people with a history of recent recurrent falls would most
likely have impaired balance. This assumption was then
verified by testing balance abilities directly in all subjects.
Results from the clinical testing, shown in Table 2, found
that older adults with a previous history of falls had significantly lower test scores (p < .05) on all clinical tests of balance compared to the older adults with no history of falls,
supporting the assumption that the two groups had significantly different balance skills.
Table 1. Baseline Characteristics of Subjects
Characteristic
n
Age — yr
Range
Sex
Female — n (%)
Male — n (%)
Living Situation — n (%)
Home
Retirement center (independent)
Score on SPMSQ — n (%)
No deficit (0-2 errors)
Mild deficit (3-4 errors)
Moderate deficit (5-7 errors)
Severe deficit (8-10 errors)
No. of Medications — n (%)
0-1
2-3
4
No. of Comorbidities — n (%)
0-1
2-3
4
Young
Adults
Older
Nonfallers
Older
Fallers
20
31 ± 6
24-44
20
74 ± 6
65-86
20
78 ± 8
65-94
10(50)
10(50)
11 (55)
9(45)
13(65)
7(35)
20(100)
14(70)
6(30)
20(100)
20(100)
20(100)
18(90)
1(5)
1(5)
18(90)
1(5)
1(5)
20(100)
13(65)
6(30)
1(5)
11(55)
4(20)
5(25)
10(50)
8(40)
2(10)
10 (50)
7(35)
11 (55)
1(5)
1(5)
2(10)
8(40)
10(50)
c
J
Frequency of Imbalance — n (%)
None
Monthly
Weekly
Daily
20(100)
8(40)
2(10)
M234
SHUMWAY-COOK ETAL.
Table 2. Clinical Tests of Balance and Mobility
Elderly
Young
Balance
Self-perceptions (range 0-60)
Berg Balance Scale
(range 0-56)
Dynamic Gait Index (range 0-20)
Gait Velocity (miles/hr)
Mobility
Assistive Device
None (n)
Cane (n)
Walker (n)
Nonfallers
Fallers
60 ±0*
52 ± 14
39 ± 15
56 ± 0
20 ± 0
4.4 ± .33
53 ± 3
17 ± 4
2.8 ± .87
39 ±11
13 ± 5
2.4 ±.73
20
0
0
20
0
0
15
3
2
•Mean ± SD.
Procedures
Clinical tests of balance and mobility. — Following informed consent procedures as approved by the Institutional
Review Board at Northwest Hospital, Seattle, all subjects
underwent an assessment of balance and mobility skills.
This included measures to document functional abilities related to balance and mobility, assess underlying sensory
and motor strategies critical for these skills, and to determine potential sensory and motor impairments contributing
to instability and gait impairments (30). Subjects provided
a medical history and a self-report of fall and balance history. Subjects then completed the SPMSQ (29) and the Balance Self-Perceptions Test, a short questionnaire in which
subjects rate their perceived confidence when performing
common activities of daily living. Subjects were asked to
rate (using a scale 1-5) their degree of confidence in performing 20 basic activities of daily living (ADLs) and instrumental activities of daily living (IADLs) without fear of
loss of balance (31). The questionnaire was a modification
of one developed by Tinetti et al. (32) in their study examining the relationship between fear of falling and measures
of basic and instrumental ADLs.
A total of four clinical tests were chosen to evaluate
functional balance and walking skills. The Berg Balance
Scale (33) rates balance during the performance of 14 tasks
including sitting, standing, reaching, leaning over, turning,
and stepping. The Three-Minute Walk Test requires subjects to walk at their preferred pace for 3 minutes over a
300-foot indoor course (30). The course is carpeted and involves four different turns. Balance and gait deviations
were scored using the Performance Oriented Mobility Test
(6). The Dynamic Gait Index was used to evaluate the ability to adapt gait to changes in task demands including
changing speeds, head turns in the vertical or horizontal direction, stepping over or around obstacles, and stair ascent
and descent (30). All clinical tools have been shown previously to have good inter-rater and test-retest reliability.
Cognitive tasks. — Two secondary cognitive tasks, Sentence Completion (34), a language processing task, and
Judgment of Line Orientation (JOLO; 35,36), a perceptual
matching visual task, were used to produce changes in attention during the performance of a concurrent postural task. In
the JOLO task the subject is presented with an array of lines
numbered 1 to 11 set at different orientations. Above the
array are two unnumbered lines set at the same orientation
as two of the numbered array. Subjects are required to pick
the two numbers that correctly identify the orientation of
lines. In the Sentence Completion task the subject is presented with four blanks, some of which have a letter preceding them. The subject is required to create a four-word sentence by replacing each blank with a word. When a letter
precedes the blank, the word must start with that letter; any
word may be used in the blanks with stars. Thus, in the example "*
b
*
f
" an appropriate
response would be "the boy ran fast."
Cognitive tasks were presented on a Macintosh computer
placed at eye level with reference to the subject. A tape
recorder placed 12 inches from the subject was used to
record verbal responses to the cognitive tasks. Subjects
were instructed to stand as still as they could and complete
as many sentences (or line orientations) as possible within a
30-sec period. As soon as one sentence (or line orientation)
was completed, the examiner triggered the next stimulus.
Performance on cognitive tasks was determined by scoring
both speed (number of responses completed within 30 sec)
and accuracy. Accuracy on each trial of the JOLO task was
determined by whether or not the subject correctly identified the numbers which matched the orientation of the two
target lines presented. Thus, for each trial completed, subjects could score 0, 1, or 2 points. Final overall accuracy
was calculated as the ratio of number correct divided by
total number of possible points (e.g., 2 for each trial completed). Accuracy on each trial of the Sentence Completion
task was a function of both correctly using cued starting letters and creating a grammatically correct sentence. One
point was assigned for each word meeting criteria and 2
points were assigned for the sentence being grammatically
correct. Thus, for any sentence completed, there was a pOssibility of receiving 0 to 6 points. In our example, the sentence "the boy ran fast" would receive 6 points, "she baked
some cookies" would receive 5 points, and "the boy more
fun" would receive 4 points. Final accuracy was calculated
as the ratio of number of points received divided by total
number of possible points (i.e., 6 for each sentence completed). Change in performance on the cognitive tasks in
the two postural conditions was compared across the three
groups.
Postural conditions. — Changes in center of pressure in
the two surface conditions, firm vs compliant (medium density 2-inch Temper Foam) were used to examine postural
stability when there was competition between postural and
cognitive tasks. Center of pressure (total distance traveled)
was used as a measure of postural stability in stance. Figure
1 shows our experimental set-up. Subjects were instructed
in the two cognitive tasks, Judgment of Line Orientation
(JOLO) and Sentence Completion (Sentences), while sitting.
Subjects then performed two 30-sec trials of each cognitive
task while standing with feet shoulder distance apart, hands
COGNITIVE DEMANDS AND POSTURAL CONTROL
at side of body, on a firm or compliant surface. Subjects also
stood for two trials on firm vs compliant surface in a no task
context. Task order was randomized. Instructions to the subjects were to stand as still as possible and complete as many
responses as possible within the 30-sec period.
Apparatus. — The apparatus consisted of a static force
platform (Balance Master, NeuroCom, Clackamas, OR)
which measured force and moment components along the
x, y, and z axes. The signals were amplified and bandpass
filtered from 0-10 Hz (12 bits analog/digital conversion, 50
Hz sampling rate). Displacement of the center of pressure
was then computed and presented as total distance traveled
in mm for each 30-sec trial.
Analysis
Displacement of the center of pressure (distance traveled in
mm over 30-sec trial) was used to quantify postural stability
in the six conditions, firm surface/no task, firm surface/JOLO
task, firm surface/sentence task, compliant surface/no task,
compliant surface/JOLO task, compliant surface/sentence
Figure 1. The experimental set-up in which subjects stand on an instrumented forceplate (with and without compliant foam surface) and perform
secondary cognitive tasks presented on a computer placed at eye level in
front of them.
M235
task. Four of the older adults with a history of falls (18%) lost
their balance and were caught to prevent a fall during at least
one condition on the compliant surface. In order to include
their data, the distance traveled during the period in which
they were able to maintain their balance was determined and
extrapolated to the full 30 sec. We conducted separate analyses removing the 4 subjects who could not perform at least
one trial on the compliant surface. No differences in the overall results were found when these subjects were removed;
thus analyses are reported utilizing the full sample.
Descriptive statistics were used to characterize the subject sample; subjects were grouped according to fall history. A multivariate analysis of variance (MANOVA) was
used to analyze this factorial mixed design (37,38). Pearson
correlation was used to assess the relationship between
clinical measures of balance and mobility and postural
sway under the dual task conditions (39).
RESULTS
The Effect of Concurrent Cognitive Task
on Postural Stability
Following initial investigation of the data, square root
transformation of the raw displacement of center-of-pressure data was conducted to ensure homogeneity of variance, normalize the distributions, and reduce the impact of
outlying scores (40). Multivariate analysis of variance of
these transformed data revealed several significant effects
including main effect of group (F = 17.66, p < .0001), main
effect of surface (F = 143.52, p < .001), a Surface X Group
interaction (F = 7.19, p < .001), a main effect of task (F =
44.38, p < .001), and a Task X Group interaction (F =
11.79, p < .05). The interaction between the within-subject
factors was not significant, nor was the triple interaction of
Group X Surface X Task. Given the finding of significant
interactions between the within- and between-subject factors, simple main effects were investigated next (37,39).
The effect of surface. — Given the interaction of Group
with Surface, the effect of Group was examined in each
surface independently (averaging across tasks). On the firm
surface, there were no significant differences found between young adults and the healthy older adults, but the
older adults with a history of falls differed significantly
from both of these groups (p < .05, utilizing Scheffe" multiple range tests). On the compliant surface there were significant differences in postural stability between all three
groups (p < .05, utilizing Scheffe" multiple range tests), with
the older adults with a history of falls exhibiting the most
instability, and the younger adults being the most stable.
Figure 2 shows the effect of surface on postural stability
(averaged across the three task conditions) in the three
groups. All groups displayed a similar pattern of performance, showing significantly more instability on the compliant than on the firm surface; the stability of the older
adults with a history of falls was most affected (p < .001 for
all groups utilizing paired Mests).
The effect of task. — Given the interaction of Group with
Task, the effects of Group were examined for each task in-
M236
SHUMWAY-COOK ETAL.
dependently (averaging across surfaces). Again, using
Scheffe" multiple range tests, analyses revealed that in the
no-task condition, there was no significant difference in
stability between the young adults and the older healthy
adults; however, the older adults with a history of falls
swayed significantly more than either of the other two
groups. The addition of either the JOLO or Sentences tasks
produced a significant difference in stability among the
three groups (p < .05). The older adults with a history of
falls exhibited the most instability, while the younger adults
were the most stable. Figure 3 shows the effect of task (averaged across surface) for all three groups.
Utilizing planned comparisons (37), the pattern of performance across tasks within each group was examined.
This analysis revealed that the pattern of performance
across tasks was similar for the young adults and older
adults without a history of falls. Neither of these groups
had significant, differences in stability between the no-task
and JOLO conditions, but did sway significantly more during the Sentences task (F = 22.15, p < .001; F = 24.09, p <
.001, respectively). In contrast, in the older adults with a
history of falls, both JOLO and Sentence Completion significantly increased sway over the no-task condition (F =
5.48, p < .03; F = 29.26, p < .0001, respectively) with the
greatest increase in postural instability found during performance of the Sentences task. These results can also be seen
in Figure 4, which shows the increase in center-of-pressure
displacement during the performance of the Sentences task
in two older subjects, a nonfaller and one with a history of
falls. These results are inconsistent with the hypothesis that
the greatest interference between cognitive and postural
tasks in our experimental paradigm would be during the
performance of a visual spatial orientation task (JOLO) because of the competition for visual processing pathways in
the two tasks.
The Effect of Postural Demands on Cognitive Performance
Figure 5 presents the number of responses completed on
the two cognitive tasks, on the firm vs compliant surface
for all three groups. Results indicate that young people
were significantly faster (number of responses completed
within 30 sec) at both tasks than either older group. No differences were found between the two groups of older adults
on speed of performance on either cognitive task. There
was no significant effect of surface condition on number of
responses completed for either cognitive task.
Figure 6 presents the percentage of correct responses on
the two cognitive tasks, on the firm vs compliant surface
for all three groups. Compared to the two groups of older
adults, young adults showed significantly more correct responses on the JOLO task but not the Sentence Completion
task. There was no significant difference in number of correct responses between the two groups of older adults.
There was no significant effect of surface condition on
number of correct responses for any of the groups.
Relationship Between Clinical Measures
and Dual Task Performance
For all conditions there was a high correlation between
performance on clinical measures of balance and mobility,
and center-of-pressure displacement in the dual task conditions in the older adults. Correlations ranged from -.4194 to
-.8351; all correlations were significant (p < .0001). Among
our older adults, those who scored the poorest on the clinical
measures of balance and mobility also had the largest displacement of center of pressure in the dual task conditions.
Effect of Surface
Effect of Task
36 -,
Pressure
36
32 32 28
28 -
Center
"o
24
24 -
"o
ment
n
20
20 -
b
16-
^
Older Adults with Falls
•
Older Adults w/o Falls
•
Young Controls
12 -
3 16 •
•
12 Firm
Compliant
Surface
Figure 2. The effect of surface conditions on postural stability in the
three groups of subjects — young adults, older adults with no history of
falls, and older adults with a history of falls.
No Task
JOLO
Older Adults with Falls
Older Adults w/o Falls
Young Controls
Sentences
Task
Figure 3. The effect of task on postural stability in the three groups of
subjects — young adults, older adults with no history of falls, and older
adults with a history of falls.
M237
COGNITIVE DEMANDS AND POSTURAL CONTROL
Older Non-Faller
Older Non-Faller
Firm Surface - No Task
Firm Surface - Sentence Completion
Sway
Older Faller
Older Faller
Firm Surface - No Cognitive Task
Firm Surface - Sentence Completion
Sway
Figure 4. Comparison of displacement of the COP during a no-task vs Sentence Completion task in an older adult with no history of falls and an older
adult with a history of falls.
Number of Responses Completed
Percent Correct/ Total Responses
• Compliant Surface
•Compliant Surface
Young
Non-
Fallara
JOLO
Young
Non-
Fatlara
SENTENCES
Figure 5. The number of responses completed on the two cognitive
tasks in the firm vs compliant surface for all three groups.
Young
Non-
Falters
JOLO
Young
Non-
Fallen
SENTENCES
Figure 6. The percentage of accurate responses on the two cognitive
tasks, in the firm vs compliant surface for all three groups.
M238
SHUMWAY-COOK ETAL.
DISCUSSION
This study used a dual task design to investigate the effects of two different types of cognitive tasks on stability
(as measured by center-of-pressure displacement) in young
vs older adults with and without a history of falls. Our results found that stability did not significantly differ between
the young adults and the healthy older adults. This is consistent with other authors who have shown either no difference or only marginal difference in stability between young
adults and healthy older adults (7,8). It is the older adults
with a history of falls who consistently show significant impairments in stability whether standing on a firm or compliant surface. While stability in all three groups decreased
when standing on the compliant surface, the older fallers
were most affected, suggesting that changing the availability of somatosensory inputs for postural control was most
challenging for the older fallers.
We had hypothesized that when two tasks require the
same processing pathways there should be a decrement in
performance on one of the tasks when they are performed
simultaneously. Thus, one would expect that the greatest interference between cognitive and postural tasks in our experimental paradigm would be during the performance of a
visual spatial orientation task (JOLO) while standing on a
compliant surface because of the competition for visual
processing pathways in the two tasks. Results from our experiment found that the greatest interference was between
the sentence completion task and the task of standing on a
compliant surface. This could be due to the fact that while
the sentence completion task is a language task, it was presented visually, thus placing some demands on visual processing pathways. It is possible that had auditory pathways
alone been used for the processing of this task, less interference on the control of postural stability would have occurred. Another possibility is that the JOLO task performance could be enhanced by maintaining vertical stability;
thus, maintaining postural stability would have been important for both the stability and accurate task completion.
While no differences were found between the young
adults and the older healthy adults on the firm surface/notask condition, when task complexity was increased either
through the introduction of a secondary cognitive task or a
more challenging postural condition, differences between
these two groups began to emerge. This can be seen in the
significant decrease in stability in the healthy older adults
compared to the young adults during stance on the compliant surface. While there was no significant difference in
stability between the two groups during performance of the
visual perceptual matching task (JOLO), there was during
the performance of the Sentence Completion task. In contrast, postural stability in older adults with a history of falls
was significantly affected by both the visual perceptual and
language tasks.
These results suggest that when postural stability is impaired (as in the elderly fallers), even relatively simple cognitive tasks can further impact balance. It is also possible
that other factors could account for the differences between
the two older groups. In addition to significant differences
in balance abilities, the group of fallers tended to be older,
took more medications, had more comorbidities, and scored
lower on the SPMSQ test, compared to the group of nonfallers. Thus, it may be that differences in health and mental status, rather than balance per se, could account for differences between the two groups. In addition, since the
faller group had a larger percentage of females than the
nonfaller group, gender bias may have contributed to group
differences.
Contrary to our original hypothesis, our results suggest
that during the simultaneous performance of a cognitive
and postural task, decrements in performance were found in
the postural stability measures rather than the cognitive
measures. Interestingly, increasing the difficulty of the postural tasks (standing on a compliant surface) did not significantly impact performance on the cognitive tasks. We had
hypothesized that in a dual task context, allocation of attention would be directed toward maintaining stability at the
expense of cognitive performance. We referred to this as a
"posture first" hypothesis, suggesting a hierarchy exists in
the allocation of attentional resources with posture being a
first priority. Our data do not support the hypothesis that
posture is hierarchically more important than the performance of these secondary tasks, since during the performance of concurrent tasks, the decrement was found in postural stability rather than in the cognitive measures in all
three groups. We do not believe, however, that this finding
would be constant across all postural conditions. It is possible that the conditions used in this study were not challenging enough to require a change in priorities so that postural
control is ensured even at the expense of cognitive performance. The threat of injury was quite small. Subjects could
simply take a step to prevent a fall, or were caught by one
of the researchers.
We thus suggest a modification to the "posture first" attentional hierarchy. We suggest that the allocation of attention during the performance of concurrent tasks is complex,
depending on many factors including the nature of both the
cognitive and postural task, the goal of the subject, and the
instructions. In situations where the threat of injury is great
(e.g., walking on an icy street or standing at the edge of a
cliff), postural control would be the first priority for attentional resources. However, we suggest that in many situations the threat to stability is not so clear cut. In situations
where stability is not potentially injurious, performance oh
a secondary task could then take priority.
We think that hierarchies exit in both cognitive and
postural tasks with respect to the degree of attention^
demands. This is consistent with results from other researchers (20-23) who suggest that attentional demands increase with the difficulty of the postural tasks. For example,
sitting, standing, and walking represent a postural hierarchy
with respect to attentional demands. In a dual task context,
when postural demands on attentional resources are low,
cognitive tasks that are similarly low in attentional demands may not affect postural stability, but more demanding cognitive tasks might. This can be seen by the deterioration in postural stability during the performance of the
Sentence Completion task in comparison to the visual processing (JOLO) task, which did not affect stability. However, when postural demands are high (either due to the sta-
COGNITIVE DEMANDS AND POSTURAL CONTROL
bility requirements inherent in the task being performed or
because the individual has a limited capacity to maintain
postural stability either due to aging or disease), even relatively nondemanding cognitive tasks might adversely affect
postural stability. Again, this can be seen in the adverse effect on stability of both cognitive tasks in the older adults
with a history of falls.
Balance retraining programs designed to reduce the likelihood for injurious falls in older adults have focused primarily on modifying many of the sensory and motor impairments affecting postural control (31,41^43). The results
of this study provide an increased understanding of attentional demands associated with declining postural control,
and their effect on stability when additional cognitive tasks
are being performed. This increased understanding could
serve as a basis for the development of new balance retraining programs that focus on training stability under the context of multiple tasks. The concept of multiple hierarchies
in both postural and cognitive tasks could be exploited by
the clinician when developing sequential tasks designed to
challenge and improve balance. Initially, tasks that are low
on the postural hierarchy could be retrained; as balance
improves, secondary cognitive tasks of varying difficulty
could be introduced to further challenge balance and enhance the development of postural stability.
ACKNOWLEDGMENTS
This investigation was supported by grant R01 AG5317 to M. Woollacott from the National Institute on Aging.
Address correspondence to Dr. Anne Shumway-Cook, Department of
Physical Therapy, Northwest Hospital, 1550 N. Meridian Avenue, Seattle,
WA 98133. E-mail:[email protected]
REFERENCES
1. Tinetti ME, Speechley M, Ginter SF. Risk factors for falls among elderly persons living in the community. N Engl J Med 1988;319:
1701-7.
2. Campbell AJ, Borrie MJ, Spears GF. Risk factors for falls in a community-based prospective study of people 70 years and older. J
Gerontol Med Sci 1989;44:M112-7.
3. Dunn JE, Rudberg MA, Furner SE, Cassel CK. Mortality, disability
and falls in older persons: the role of underlying disease and disability. Am J Public Health 1992,82:395-400.
4. Lipsitz LA, Jonsson PV, Kelley MM, Koestner JS. Causes and correlates of recurrent falls in ambulatory frail elderly. J Gerontol Med Sci
1991;46:M 114-22.
5. Speechley M, Tinetti M. Assessment of risk and prevention of falls
among elderly persons: role of the physiotherapist. Physiotherapy
Canada 1990;2:75-9.
6. Tinetti ME. Performance oriented assessment of mobility problems in
elderly patients. J Am Geriatr Soc 1986;34:119-26.
7. Woollacott MH, Shumway-Cook A, Nashner LM. Aging and posture
control: changes in sensory organization and muscular coordination.
Int J Aging Hum Dev 1986;23:97-114.
8. Horak F, Shupert C, Mirka A. Components of postural dyscontrol in
the elderly: a review. Neurobiol Aging 1989; 10:727-45.
9. Stelmach GE, Worrington CJ. Sensorimotor deficits related to postural stability: implications for falling in the elderly. Clin Geriatr Med
1985; 1:679-94.
10. Stelmach GE, Teasdale N, DeFabio RP, Phillips J. Age related decline
in postural control mechanisms. Int J Aging Hum Dev 1989;29:
205-23.
11. Geurts ACH, Mulder TW, Nienhuis B, Rijken RAJ. Dual-task assessment of reorganization in persons with lower limb amputation. Arch
Phys Med Rehabil 1991 ;72:1059-64.
M239
12. Stelmach GE, Zelaznik HN, Lower D. The influence of aging and attentional demands on recovery from postural instability. Aging
1900;2:155-61.
13. Kosnik W, Winslow L, Kline D, Rasinski K, Sekuler R. Visual
changes in daily life throughout adulthood. J Gerontol Psych Sci
1988;43:P63-70.
14. Sloane P, Baloh RW, Honrubia V. The vestibular system in the elderly.
Am J Otolaryngol 1989;1:422-9.
15. Ochs AL, Newberry J, Lenhardt ML, Harkins SW. Neural and
vestibular aging associated with falls. In: Birren JE, Schaie KW, eds.
Handbook of the psychology of aging. New York: Van Nostrand
Reinhold, 1985:378-99.
16. Whanger A, Wang HS. Clinical correlates of the vibratory sense in elderly psychiatric patients. J Gerontol 1974;29:39-45.
17. Pitts DG. The effects of aging on selected visual functions: Dark
adaptation, visual acuity, stereopsis, and brightness contrast. In: Sekular R, Kline D, Dismukes K, eds. Modern aging research: aging and
human visual function. New York: Alan R. Liss, 1982:131-60.
18. Rosenhall U, Rubin W. Degenerative changes in the human vestibular
sensory epithelia. Acta Otolaryngol 1975;79:67—81.
19. Posner MI, Nissen MJ, Klein RM. Visual dominance: an informationprocessing account of its origin and significance. Psychol Rev 1976;
83:157-71.
20. Lajoie Y, Teasdale N, Bard C, Fleury M. Attentional demands for
static and dynamic equilibrium. Exp Brain Res 1993;97:139-44.
21. Bardy BG, Laurent M. Visual cues and attention demands in locomotor positioning. Percep Mot Skills 1991;72:915-26.
22. Teasdale N, Lajoie Y, Fleury M, Bard C. Cognitive processes involved for regulating balance when walking. Soc Neurosci Abstr
1993; 19:990.
23. Teasdale N, Bard C, LaRue, J, Fleury M. On the cognitive penetrability of posture control. Exp Aging Res 1993; 19:1-13.
24. Cordo P, Nashner L. Properties of postural adjustments associated
with rapid arm movements. J Neurophysiol 1982;47:287-302.
25. Horak FB, Anderson M, Esselman P, Lynch K. The effects of movement velocity, mass displaced and task certainty on associated postural adjustments made by normal and hemiplegic individuals. J Neurol
Neurosurg Psychiatry 1984;47:1020-8.
26. Maki BE, Whitelaw RS. Influence of experience, expectation and
arousal on posture control strategy and performance. In: Woollacott
M, Horak F, eds. Posture and gait: control mechanisms. Eugene, OR:
University of Oregon Books, 1992; 1:123-6.
27. Teasdale N, Stelmach GE, Breunig A. Postural sway characteristics of
elderly persons under normal and altered visual and support surface
conditions. J Gerontol Med Sci 1991;41:M 197-223.
28. Shumway-Cook A, Horak F. Assessing the influence of sensory interaction on balance. Phys Ther 1986;66:1548-50.
29. Pfeiffer E. Short portable mental status questionnaire. J Am Geriatr
Socl975;23:433^1.
30. Shumway-Cook A, Woollacott M. Motor control: theory and practical
applications. Baltimore: Williams & Wilkins, 1995.
31. Shumway-Cook A, Gruber W, Baldwin M, Liao S. The effects of exercise on balance, mobility and risk for falls in community-dwelling
older adults. Phys Ther 1997;77:46-57.
32. Tinetti ME, Baker DI, McAvay G et al. A multifactorial intervention
to reduce the risk of falling among elderly people living in the community. New Engl J Med 1994;331:821-7.
33. Berg K. Measuring balance in the elderly: Validation of an instrument
(Dissertation). Montreal, Canada: McGill University, 1993.
34. Ekstrom RB, French JW, Harmon HH. Kit of factor-referenced cognitive tests. Princeton, NJ: Educational Testing Service, 1976.
35. Benton AL, Varney NR, Hamshere KdeS. Judgment of line orientation test manual. Iowa City: University of Iowa, 1977.
36. Benton AL, Hamshere KdeS, Varney NR, Spreen O. Contribution to
neuropsychological assessment: a clinical manual. New York: Oxford
University Press, 1983.
37. Hand DJ, Taylor CC. Multivariate analysis of variance and repeated
measures: a practical approach for behavioral scientists. London:
Chapman and Hall, 1987.
38. Tabachnick BG, Fidell LS. Using multivariate statistics. New York:
Harper & Row, 1983.
39. Kirk, KE. Experimental design: procedures for the behavioral sciences. Pacific Grove, CA: Brooks/Cole Publishing, 1995.
M240
SHUMWAY-COOK ETAL.
40. Pedhazur EJ. Multiple regression in behavioral research: explanation
and prediction. New York: Holt, Rinehart & Winston, 1982.
41. Hu M-H, Woollacott MH. Multisensory training of standing balance
in older adults: I. Postural stability and one leg balance. J Gerontol
MedSci 1994;49:M52-61.
42. Moore S, Woollacott M. Sensory organization training for older
adults with impaired balance. Manuscript submitted for publication.
43. Fiatarone MA, Marks EC, Ryan ND, Meredith CN, Lipsitz LA, Evans
WJ. High-intensity strength training in nonagenarians: effects on
skeletal muscle. JAMA 1990;263:3029-34.
Received February 28, 1996
Accepted November 6, 1996
ANNOUNCING~ln your mailboxes now!
A Bonus Issue
for AJJ GSA Members and for Subscribers
to The Journals of Gerontology, Series B:
Psychological and Social Sciences
A Monograph of 10 Peer-Reviewed Papers Entitled
Asset and Health Dynamics
Among the Oldest Old (AHEAD):
Initial Results from the Longitudinal Study
George C. Myers, PhD, Guest Editor
This publication of The Gerontological Society of America was made possible by
generous support from the National Institute on Aging, The University of Michigan, and
the Office of the Assistant Secretary for Planning and Evaluation (ASPE). Additional
copies may be purchased at $10.00 each including postage and handling. Send check
or credit card data to The Gerontological Society of America, P.O. Box 79151,
Baltimore, MD 21279-0151. For bulk discounts, call GSA at 202/842-1275.