Purpose and Significance

The purpose of this study was to investigate the effects of different task prioritization

(PF vs. SF) on postural-suprapostural performance and its related cortical activity in

younger and older populations. The significance of the present study was addressed in the

academic and clinical aspects. In the academic aspect, this study provided a better insight

of the behavioral results and neural mechanism of attentional allocation under different

14

task prioritization in both younger and older populations. Especially, through this study,

we could clarify the applicability of “facilitatory hypothesis” or “posture-first principle”

with behavioral and cortical evidences (Figure 1). In clinical aspect, the results may

provide the clinical value for the physical therapists to instruct older adults who have

multi-tasking difficulty with a suitable movement strategy in their daily life and prevent

them from falling.

1.5 Hypotheses

1. Both postural and suprapostural performance are different between a

postural-suprapostural task with PF or SF strategy. In addition, the suitable task-priority

strategy for younger and older adults is different. These hypotheses would be

systematically tested by postural and suprapostural accuracy, postural regularity and

reaction time of the suprapostural task. We expected that optimal

postural-suprapostural overall performance was found with SF strategy in younger adults,

whereas optimal postural-suprapostural overall performance was found with PF

strategy in older adults.

2. Attentional resource allocation between postural and suprapostural tasks is different

depending the participants performing a postural-suprapostural task with PF or SF

15

strategy. This hypothesis would be tested by P1, N1, and P2 amplitudes of ERP

signals, for representing the allocated attention for posture and supraposture

respectively. We expected that P1, N1, and P2 amplitudes were significantly affected

between PF and SF strategies. Moreover, frontal area was found related to

information processing of working memory under dual-task condition and

motor-type suprapostural task was found related to parietal area.

^32,41,42

Therefore,

significant effects were expected found in frontal and parietal areas when adopting

PF and SF strategies.

16 Chapter 2 Methods

2.1 Participants

Thirty two healthy right-handed volunteers (16 younger adults, mean age: 24.4 ± 4.6

years; 16 older adults, mean age: 69.1 ± 2.7 years) without history of neurological,

vestibular, orthopedic, or cardiovascular disorders were recruited in this study. All

subjects had normal or corrected-to-normal vision. For older subjects, they were able to

ambulate independently without walking aids and had no history of falling. Besides, Mini

Mental State Examination (MMSE) score was measured for older adults and only the

subjects with more than 24 points were included (Appendix 1). Because the subjects were

asked to perform an suprapostural task while standing on a stabilometer (67-cm length ×

50-cm width × 24-cm height, anterior-posterior tilting angle: 0-100 degrees), the subjects

who were pregnant, had prior experience with tasks, unable to maintain balance on the

stabilometer for at least 80 seconds, or took any medications that could affect balance

were excluded from this study. Telephone interview with the subjects was done before

recruiting. Table 1 is the demographic data of both younger and older groups.

The protocol was approved by the research ethics board at the National Taiwan

17

University Clinical Trail Center (Appendix 2). Study procedure was explained by the

researcher for each subject and an inform consent was signed by the subjects prior to

participating in this experiment.

2.2 System Set-up and Data Recording

The experiment consisted of postural task and suprapostural task. Participants were

requested to perform a force-matching precision grip task with their right index and thumb

(suprapostural task) while standing on a stabilometer (postural task) (Figure 2). For the

postural task, participants were asked to maintain their balance on the stabilometer

(67-cm length × 50-(67-cm width × 24-(67-cm height, anterior-posterior tilting angle: 0-100 degrees)

with an inclinometer (Model: FAS-A, MicroStrain, USA) mounted on the center of the

stabilometer plate to measure the tilting angle of the stabilometer. The maximal anterior

tilting was recorded for each participant before the experiment and the 50% of the

maximal anterior tilting angle was set as the target angle for the postural task. For the

suprapostural task, participants were asked to execute a force-matching task, and the level

of force output was recorded with a load cell (15-mm diameter × 10-mm thickness, net

weight = 7 grams; Model: LCS, Nippon Tokushu Sokki Co., Japan). Maximum voluntary

contraction (MVC) of precision grip was also recorded before the experiment and the

18

50% of the MVC force was set as the target force for the suprapostural task. The

participants needed to execute the thumb-index precision grip task in response to auditory

cues. The auditory cues consisted of 80-second sequences of tone pips, with a total of

fifteen warning-executive signal pairs. The interval between a warning tone (frequency:

800 Hz, duration: 100 ms) and an executive tone (frequency: 500 Hz, duration: 100 ms)

was 1.5 seconds for the first three warning-executive pairs, but was random presented at

different intervals of 1.5, 1.8, 2.1, 2.4, 2.7 or 3.0 seconds form the fourth to fifteenth

warning-executive pairs. The interval between the executive tone and the next warning

tone was 3.5 seconds. Participants performed a quick thumb-index precision grip (force

impulse duration < 0.5 second) to couple the peak precision force with the force target

when receiving the executive tone. In order to determine the reaction time (RT) of

force-matching, the initial activation of the first dorsal interosseous (FDI) muscle was recorded

with surface electromyogram (EMG) in a bipolar arrangement (Ag/AgCl, 1.1 cm in

diameter, Model: F-E9M-40-5, GRASS) and an AC amplifier (gain: 5000, cut-off

frequency: 1 and 300 Hz; Model: QP511, GRASS).

For recording cortical activation, electroencephalogram (EEG) data was recorded

, CP

3/4

19 z

, P

3/4

, T

3/4

, TP

7/8

, O

z

, and O

1/2

. The ground electrode was placed along the midline ahead

of F

z

and the recording references were placed on the mastoids of the both sides. In

addition, two electrodes were attached above the arch of the left eyebrow and below the

eye to monitor eye movements and blinks. The impedances of all electrodes were

maintained below 5 kΩ, and data was recorded with a band-pass filter set at 0.1 to 100

Hz with a notch filter at 60 Hz to remove the noise from the environment. Both behavioral

and cortical signals, including stabilometer movement, precision grip force, EMG of FDI

muscle, and EEG data, were synchronized with a sampling rate of 1 kHz.

2.3 Experimental Conditions and Procedures

This study was conducted in two separate days with one-week apart. Participants in

both age groups were randomly assigned to either PF or SF conditions in the first day and

to the other in the second day (Figure 3). In each experimental day, participants were

requested to perform three experimental tasks, including one postural-suprapostural task,

and two corresponding control tasks (a single corresponding postural task and a single

corresponding suprapostural task). There were six trials for each experimental task.

In most previous researches related to task prioritization, the lack of specification

instruction for how participants directing their attention when performing dual tasks was

20

a major limitation.

¹⁶

For the better improvement of task prioritization instruction, a

procedure derived from “optimum-maximum method” proposed by Navon (1990) was

used in this study for manipulating task prioritization.

⁴⁰

The optimum-maximum method

was used to guard subjects’ attention with specific instruction for both high-priority and

low-priority tasks.

^23,43

With this method, the high-priority task was designed the

“to-be-optimized” task, and low-priority task was the “to-be-maximized” task. Participants were

instructed to execute the high-priority task with their “optimum” level and to perform

their best on the low-priority task. Such a procedure required participants to optimize the

high-priority task and not to “give up” on the low-priority task. Besides, individually

determined performance standard and performance feedback were provided in the

high-priority task but not for low-high-priority task. Therefore, in this study, visual feedback about

the target and performance of stabilometer movement or force-matching task was used

for enhancing the prioritization of the attention (Figure 4). For example, participants in

the PF condition were instructed to pay their primary attention on the postural task with

maintaining the tilting angle of the stabilometer at the target angle precisely, and to

maximize the precision of force-matching task. Visual feedback of stabilometer target

angle and instantaneous stabilometer tilting angle was provided in the PF condition, but

visual information about the force-matching target and force output was not provided.

Because the visual feedback was only provided for postural performance, the

21

corresponding control tasks of the PF condition were that 1) performing the postural task

on the stabilometer with visual feedback and did not execute the force-matching task, and

2) performing the force-matching task without visual feedback on a stable box (67-cm

length × 50-cm width × 24-cm height). In contrast, participants in the SF condition were

instructed to pay their major attention on the precision grip task with coupling the force

peak with the target precisely, and to maximize the precise tilting angle of the stabilometer.

Visual feedback of the force-matching target and force output was provided in the SF

condition, but visual information about the stabilometer and its target angle was not

provided. The corresponding control tasks of the PF condition were that 1) performing

the postural task on the stabilometer without visual feedback and did not execute the

force-matching task, and 2) performing the force-matching task with visual feedback on

a stable box (67-cm length × 50-cm width × 24-cm height). Besides, in order to remind

the force-matching target for the PF condition and the tilting angle target for the SF

condition, the visual feedback about the first 3 force-matching performances and the first

10-second stabilometer tilting angle with their target was provided in each trial for the PF

and the SF conditions, respectively. All the visual information was displayed on a 22-inch

computer monitor with 60 cm in front of the subjects at eye-level.

22

2.4 Data Analysis

2.4.1 Behavioral Data

For postural performance, the inclinometer data was conditioned with 6-Hz low-pass

filter and the units were converted to degrees. The inclinometer data from every executive

tone to next warning tone was selected for calculation of absolute postural error and

absolute postural approximate entropy (ApEn). The absolute postural error was presented

by calculating the root mean square (RMS) of the mismatch between the target angle and

the stabilometer tilting angle and then divided by the target angle, presenting as

RMS(SA-TA)

×100%

TA

(SA: stabilometer tilting-angle, TA: target angle). The absolute postural ApEn of the stabilometer tilting angle’s trajectory was used to represent the

variability property of the postural performance. According to previous study, the

calculation of postural ApEn was calculated after the trajectory of stabilometer tilting

angle normalized with standard deviation of time series, presenting as ApEn (m, r) =

log[C

m

(r)/C

m+1

(r)].

⁴⁴

Where m represents the length of the compared time windows and r

represents the tolerance range of the regularity.

^44-46

If a completely predictable time-series

with high regularity, value of C

m

(r) will be very close to C

m+1

(r), yielding a log-probability

(ApEn) of zero.

⁴⁴

In this study, m equaled 2 and the tolerance range of r was 0.15× the

23

standard deviation of the time series

⁴⁴

. The value of the ApEn was between 0 and 2. An

ApEn value of closer to 2 represented higher irregularities, or larger complexity of the

postural movement changes. In contrast, an ApEn value of closer to 0 represented greater

regularity.

⁴⁷

For suprapostural performance, the absolute force-matching error was presented as PPF-TF

×100%

TF (PPF: peak precision-grip force, TF: target force). The absolute force-matching RT of suprapostural task was recorded by calculating the time delay from the

presentation of executive tone to the EMG onset of FDI muscle. All behavioral

parameters of postural-suprapostural task were normalized in reference to its

corresponding control task.

software (NeuroScan Inc., EI Paso, TX, USA) and the off-line analysis was used for the

analysis. The DC shift of each channel on entire EEG data was corrected with third-order

24

correction. The eye movements and blinks were removed from the EEG data. After eye

movements were removed, the EEG data was low-pass filtered with cut-off frequency of

40 Hz (48 dB/octave), and segmented into epochs of 700 ms, including a 100 ms before

the onset of executive signals. The 100 ms-data prior the executive signals was used for

the baseline correction of each EEG epoch. A visual inspection for each epoch was

applied, and those epochs with artifacts, including excessive drift, eye movements or

blinks, were removed from analysis. Those epochs with adequate responses were

averaged. ERPs from the six trials of each task were group averaged separately at each

condition for each subject. According to the previous ERP studies, P1 amplitude was

reported associated with sensory input to attended task

³⁸

, N1 was associated with the

attention modulation related to postural control, and P2 was associated with the attention

modulation related to perceptual-motor suprapostural task,

^32,44

Therefore, in the present

study, we analyzed the peak amplitudes of P1 (70-110 ms), N1 (80-150 ms), and P2

(150-240 ms) components across all EEG electrodes to characterize the attention allocation

between postural and precision-grip tasks.

2.5 Statistical Analysis

The task prioritization conditions (PF condition, SF condition) and age groups

25

(younger group, older group) effects on behavioral and electrophysiological parameters

of postural and suprapostural tasks, including the normalized force-matching error,

normalized force-matching RT, normalized postural error, normalized postural ApEn, and

ERP amplitudes of P1, N1, and P2 components were compared with 2 × 2 mixed analysis

of variance (ANOVA). When necessary, post hoc least significant difference (LSD)

comparisons were performed. The level of significance was set at p < 0.05. Signal

processing of behavioral data and statistical analysis was completed by using MatLab v.

R2008a (Mathworks, Natick, MA, USA) and the statistical package for SPSS statistics v.

17.0 (SPSS Inc., Chicago, IL, USA).

26 Chapter 3 Results

3.1 Behavioral Performance

3.1.1 Error and Regularity of Postural Performance

Figure 5 shows the absolute and normalized postural error of SF and PF conditions

in the younger and older groups. ANOVA results suggested that normalized postural error

was subject to task prioritization (F

1, 30

= 12.99, p < 0.01) and age difference (F

1, 30

= 11.28,

p < 0.01) without interaction (F 1, 30

= 0.30, p = 0.59). Larger normalized postural error

was observed in the PF condition than that in the SF condition for both younger and older

groups (p < 0.05). Besides, normalized postural error was larger in the older group than

that in the younger group across task prioritization conditions (p < 0.05). The normalized

postural error of SF condition in the younger group was below 100% (84.51 ± 3.86%),

but the others were above 100%, indicating that younger adults had better postural

performance during the postural-suprapostural dual-task condition than that during the

single postural task condition. For postural regularity, Figure 6 displayed the absolute and

normalized postural ApEn results of SF and PF conditions in the younger and older

27

groups. ANOVA results showed a significant main effect of task prioritization (F

1, 30

4.41, p < 0.05) and age difference (F

1, 30

= 18.82, p < 0.001) on the normalized ApEn

values without a significant interaction (F

1, 30

= 2.21, p < 0.15). Post-hoc testing showed

a larger normalized ApEn in the younger group than that in the older group (PF condition:

younger (102.87 ± 1.58%) > older (92.16 ± 1.65%)), p <0.01; SF condition: younger

(103.87 ± 1.70%) > older (97.99 ± 2.12%), p <0.05), indicating that younger adults had

higher postural irregularity when performed a postural-suprapostural task than older

adults. Also, we noted that normalized ApEn was above 100% in the younger for both PF

and SF conditions, but was below 100% in the older group, indicating that addition of the

force-matching task led to an opposite effect on postural regularity between younger and

older groups. On the other hand, the task prioritization effect on normalized ApEn was

only shown in the older group with larger value in the SF condition than that in the PF

condition (p < 0.05).

3.1.2 Error and Reaction Time of Force-matching Task

For suprapostural performance, force-matching error of PF and SF conditions in

younger and older groups is shown in Figure 7. ANOVA results suggested that normalized

force-matching error was subject to task prioritization (F

1, 30

= 12.31, p < 0.01), but not to

28

age effect (F

1, 30

= 2.25, p = 0.14) with no significant interaction effect (F

1, 30

= 1.69, p <

0.20). Post-hoc evaluation revealed that normalized force-matching error in older group

was higher in PF condition than that in SF condition (p < 0.05). Besides, all normalized

force-matching errors were above 100% (younger group: PF condition = 118.90 ± 5.63%,

SF condition = 103.16 ± 5.49%; older group: PF condition = 139.88 ± 11.57%, SF

condition = 105.65 ± 5.31%), indicating that force-matching error tended to increase

when subjects were requested to perform a force-matching task and kept their balance on

a stabilometer concurrently compared to perform the force-matching task in a stable

posture (stand on a stable box).

Figure 8 displays the RT of force-matching task of PF and SF conditions in younger

and older groups. Similar as force-matching error, all normalized force-matching RT

values were above 100% (younger group: PF condition = 110.79 ± 3.50%, SF condition

= 107.70 ± 1.87%; older group: PF condition = 102.51 ± 4.12%, SF condition = 102.36

± 2.80%), indicating that RT would be longer when subjects were requested to perform a

force-matching task and kept their balance on a stabilometer concurrently compared to

perform the force-matching task in a stable posture. However, the RT of force-matching

did not vary with either task-priority strategy or age difference (task-priority effect: F =

0.48, p = 0.50; age effect: F = 3.15, p = 0.09).

29

3.2 ERP Amplitudes

Figure 9 displays the typical ERP waveforms of younger group and older group in

postural-suprapostural tasks. It is interesting to find that the ERP characteristics were

different between the younger and older groups. In the younger group, only the N1 and

P2 waves presented after the presentation of the executive signals across

postural-suprapostural conditions (Figure 9(a)); however, the P1, N1, and P2 waves were all

observed in sequence after the presentation of the executive signals in the older group

(Figure 9(b)). Therefore, for statistical analysis of ERP amplitude, N1 and P2 amplitudes

were analyzed via a 2 (task prioritization: PF vs. SF) × 2 (age: younger vs. older) mixed

ANOVA, with repeated measure on the first variable, while P1 amplitudes was analyzed

via a paired t-test to examine the task prioritization effect for the older adults.

3.2.1 Task Prioritization Effect on ERP Amplitudes

Figures 10(a-e) are typical ERP recordings showing the effects of task prioritization

P1, N1, and P2 amplitudes. ANOVA results suggested that in the younger group, the N1

z

: F

1, 30

= 6.61, p < 0.05) were subject to a significant task prioritization effect.

1, 30

= 4.47, p < 0.05). Further post-hoc analysis indicated

that P2 amplitudes on T

5

, P

3

, P

Z

, and O

1

electrodes were greater in the SF condition than

), and right temporal (FT

8

and T

4

) areas with SF strategy (p < 0.05)(Figure 11(c)).

ANOVA results suggested that the N1 amplitudes of the electrodes around parietal (CP

3 1, 30

= 21.26, p < 0.001; CP

Z

: F

1, 30

3

, and P

Z

) were larger in the PF condition than that in the SF

31

condition (p < 0.05)(Figure 11(d). On the other hand, the P2 amplitudes of electrode FT

8

had a significant main effect of task prioritization (F

1, 30

= 5.16, p < 0.05). Besides, some

electrodes showed significant interaction effect between task prioritization and age

factors around right frontal (F

8

: F

1, 30

= 4.39, p < 0.05; FT

8

: F

1, 30

= 5.26, p < 0.05) and

temporal (T

在文檔中作業優先性對姿勢-上姿勢作業與大腦活動的影響：年齡效應 (頁 29-0)

14

15

32,41,42

16

Chapter 2 Methods

17

18

1/2

z

3/4

7/8

7/8

z

3/4

7/8

z

3/4

z

3/4

19

z

3/4

3/4

7/8

z

1/2

z

20

16

40

23,43

21

22

RMS(SA-TA)

×100%

TA

m

m+1

44

44-46

m

m+1

44

23

44

47

24

38

32,44

25

26

Chapter 3 Results

1, 30

1, 30

p < 0.01) without interaction (F 1, 30

27

1, 30

1, 30

1, 30

1, 30

28

1, 30

1, 30

29

3

1, 30

3

1, 30

3

1, 30

3

1, 30

30

3

1, 30

z

1, 30

z

1, 30

^32,41,42

¹⁶

⁴⁰

^23,43

⁴⁴

^44-46

⁴⁴

⁴⁴

⁴⁷

³⁸

^32,44