CGA FAQ: Effect of arm-swing on gait

from BIOMCH-L 1/5/95
>Dear Colleagues,
>As part of a larger study, I recently evaluated the effects of a
>fixed arm position during gait as compared to a normal arm
>The study involved 79 individuals over the age of 60 under the
>two conditions.  The results were a bit surprising and
>speculative.  The only differences observed from normal to the
>fixed arm position were a decrease in maximum hip flexion, an
>increase in ankle dorsiflexion, a decrease in plantarflexion,
>and an increase in walking velocity.  Of all the changes, I
>really thought the walking velocity would decrease or not change
>at all.

>Anyway, I would appreciate your thoughts on these differences as
>well as any references on arm position and gait.  As always, I
>will post all replies.

    I noticed this recently while in a dance class. The upper
limb and the   lower limb are set to function in a rhythm by
default. But, if you disturb   that rhythm a period of chaos
occurs. (If fact, I messed up the steps  whenever the upper limb
motion did not match with that of the lower limb).  This would
probably be an interesting study ( Rhythms of motion of limbs).

    I am curious if the subjects were told to hold the upper limb
steady and   asked to walk and the data taken in the first
attempt. (Have you tried   training them and then see if there
was any difference ?). The increased   velocity could be for
balance reasons. (Of course it is easy to contemplate   once a
certain trend has been observed).

Interesting observations.  Did you fix one or both arms?  Was the
increase  in velocity accompanied by changes in cadence, step
length, or both?

One speculative thought is that increases in velocity could
relate to unsteadiness.

What was the fixed arm position?  Was it anterior to the
prevailing  location of the CG.  If so, the increase in walking
velocity could make  sense, more of a "fall forward" effect which
could increase gait pace.  Just a thought, interesting study.
Could possibly verify this by fixing arm  posterior to CG and
looking for velocity decrease.

Sounds like an interesting research project.  You said you were
surprised   at the results and it made me think about my own
personal experience.    Numerous times I've been told by others
that I walk like a "tin soldier"   (arms rigid at my side,
moving/swinging very little) and along with that I  keep up a
very fast pace.  The results from your investigation sounded very
similar to my experience and I would like to hear more about the
conclusions you come to.

This is my personal reflections based on clinical experience as a
physio   and athletics trainer regarding your interesting results
from measurements  of   gait with arms in a fixed position. It is
a common observation that elderly   subjects have limited arm
swing during gait, so may be I would not expect too    great a
difference. But still, arm swing is contributing to the
rotational   counterbalance of leg swing and also hip rotation
along a vertical axis, both  of which are essential for proper
stance leg extension. So I would actually  expect walking with
arms in a fixed position to prevent effective extension  of the
leg during stance phase. As a result, step length will tend to
decrease which your subjects may try to compensate by increasing
the cadence.  Why velocity should increase is however beyond me.
You haven't mixed cadence  and velocity by the way? If your
subjects were walking on a treadmill I presume the velocity was
constant throughout the tests???

But still, any arm swing effectively counteracts the rotational
reactive  forces caused by the swinging leg thereby inducing a
more linear progress of the torso.

As a result I imagine that the stance foot is allowed to complete
the  extension phase propelling the body forward. Also for an
effective hip  rotation If the arms are not swinging, effective
hip rotation is more.  An ineffective stride would probably
result in shorter stride length, thus  effecting both extension
and flexion of the leg.

Very interesting results.  I don't have any references but just
thinking about the result of the increase in walking velocity...

This is purely speculative and it could be wrong.  As people age
they   do have difficulties with postural adjustments.  Perhaps
not having     to move their arms may decrease instability in
some way and so   therefore, they may be able to walk faster???
It would be   interesting to see some graphs on the displacement
of  centre of  body   mass when they move their arms and when
they don't.  Awaiting   everyone else's thoughts on this.  All
the best.

 I'm very interested in your research. I have some questions

1.Why are the subjects over 60? Did you evaluate the effects on
younger adults? What's the age effect?

2. Did you observe the trunk rotations? How about thorax
rotations and pelvic rotations.

3. How are the arms fixed, flexed in front of chest or flexed at
both sides?   How long did the subjects practice walking with
fixed arms before measurements?

I'm doing measurements of thorax and pelvis rotations for
different gait   patterns including evaluating the effect of the
fixed arms. I do have   difficulties to collect data since the
range of motion is low. I would   like to know what you think.

rebecca craik published her dissertation work in a book edited by
dick   herman on gait circa 1980 i think.  her study looked at
the effects of   gait velocity on the interlimb coordination of
the arms with the legs.    that might be of assistance to you.

you might also want to check the literature for specific
diagnostic   populations who have UE weakness (ALS, some types of
MD).  the   relationship of the arm swing pattern on gait might
is certainly known in   clinically as anecdotal evidence.
perhaps its been documented.

My first thought is that there may be a transfer of freezing
d.o.f from the  upper body to the lower body, since this is a
"whole body" task. This would  explain the decrease in hip
flexion as well as the decrease in ankle   plantarflexion. About
the increase in walking velocity ... I have no clue,  sorry!



The data were collected (using a Motion Analysis Corp. 2-D
system) as part of a much larger project looking at the effects
of exercise on static and dynamic balance in an older population.
The arm was held in place across the abdomen by the subject so
that the iliac crest and greater trochanter markers remained
visible throughout the gait cycle.  Subjects were given ample
practice time and walked at a self-selected speed across a 30 ft.

The next step is to look at a college age population for
comparison (although they usually have a greater magnitude of arm
swing [both in ROM and ang. vel.]; thus minimizing interpolation

A special thanks to all who replied!  :)

John Sigg
Ithaca College
Ithaca, NY

Want to know more? Email the CGA list! [n/a]

Back to Clinical Gait Analysis home page