Effect of Arm-swing on gait: What people said...

Dear all,

I'm passing this message on for Stephane <stephane.bouilland@wanadoo.fr> at the Fondation Franco-Américaine in France...

Dear Chris Kirtley, I would like to post the following message to the cga list :

Dear suscribers,

We recently examined a young hemiplegic girl when walking. She has nearly no problems any more with her leg and walks almost normally, but she still has lot of difficulties to move her arm. We are wondering if the small problems that occur at the lower limb level may be due to the fact that one arm does not move and stays alongside the trunk.Could someone explain me what is the effect of arm movement on the overall gait. Do we know if the arms move because the trunk is rotating or is it the contrary, that the trunk rotates because the arms are balancing?

Thank you very much much for your attention,

Stephane Bouilland

Just for openers, I seem to remember that Aurelio Cappozzo (J Biomech. 1993, 16:265-277) settled this one some years ago by showing that the angular momentum of the arms exactly balances that of the trunk.

But I'd be interested to know what others think. The question of whether preventing a subject from swinging their arms (in order not to obscure the hip markers) has any effect on the gait curves is one that often gets asked. I'd like to know a reference if this has ever been satisfactorily investigated (if not, it's a great idea for an undergrad project!).

By the way, this was discussed on BIOMCH-L some time ago - read the transcript here!


Dr. Chris Kirtley (Kwok Kei Chi) MD PhD Assistant Professor Department of Rehabilitation Sciences The Hong Kong Polytechnic University 

We just completed a study regarding the effects of restraining one arm on the 2D gait kinematics of elder walking: Sigg, JA, JC Ives, BC Gaba, and GA Sforzo. Effect of arm position on parameters of normal gait in older persons. Clinical Kinesiology. 51: 33-36, 1997.

The effects on a few aspects of gait were small (sorry, can't remember exactly which limb angles, etc. were affected--get reprints from lead author), but statistically signficant, and perhaps clinically relevant. My guess would be that a 3D analysis would show more gait changes as the arm swing seems to affect/counterbalance trunk rotation. The results we found may not be applicable to younger individuals. One final thing; the literature on this topic is fairly scattered in terms of years and type of journal, so don't plan on a quick computer search to find all the relevant material.

Jeff Ives, Ph.D. Dept. Exercise and Sport Sciences Hill Center Ithaca College Ithaca, NY 14850 USA Email: jives@ithaca.edu Phone: 607-274-1751 


I have seen for some time that arm swing is quite asymmetric in patients with chronic postural pain, either lower back or neck/shoulder type symptoms. What I have also observed is that in the vast majority of cases with verifiable leg length difference, the arm motion (from the shoulder) on the longer side is usually far less than on the shorter side. Correction of leg length difference often equalizes arm motion. Once arm motion is more symmetrical, as you might anticipate, there is a visual change in trunk rotation where is also becomes more symmetrical. This effect usually takes between 30-40 steps to become evident once proper limb lift is added.

I hope that this observation adds to the discussion.

Regards, Howard J. Dananberg, DPM < HowieDBPG@aol.com

From BIOMCH-L discussion June, 1998

Dear Biomech-L readers,

thanks a lot to all who responded to my questions on arm swing during
human walking. The replies are listed below.

Stefan Gruber
Dipl.-Ing. Stefan Gruber
Institute B of Mechanics
University of Stuttgart
Pfaffenwaldring 9
D-70550 Stuttgart      phone: ++49 711-685 7659
Germany                fax:   ++49 711-685 6400


As the arm swing is a natural action - exaggerated only in marching
actions, I believe that its action would be to maintain the body's
centre of gravity above the point of contact and thus minimise the
muscular effort required to balance the body.  It would be the same
effect in running - running with no arm swing would be decidedly

While I can't say that I have any references, perhaps a camera
positioned above the treadmill will show the relative positions of arms
and bodies when the subject is walking.

Alastair Campbell Ritchie
S7 level 3
National University of Singapore
Singapore 119260

Dear Stefan,

you may want to take a look at the publications by Robert Wagenaar and
Richard van Emmerich during the last eight years or so. They have
analysed human walking both in different types of patients and controls
using a dynamical systems approach. Publications give clues on the
experimental set-up, but also theoretical background of using arm swing
during human walking. It is interesting to note, by the way, that there
are two different swinging modes, dependent on walking speed. It is also
well known, that counterrotation of the upper trunk is necessary to keep
balance; obviously, swinging arms amplify this balancing motion pattern.

Hope this helps. regards,

Theo Smit
University Hospital Vrije Universiteit

Although I do not know the answers to your questions, I know someone who
probably does:

        Dr. David Webb
        Dept. of Anthropology and Sociology
        Kutztown University
        Kutztown, PA 19530 USA
        email: webb@kutztown.edu

Dr. Webb did his originial doctoral research on arm swinging in human
locomotion. So, if he can't answer your questions, he probably knows how
to find them.

Thomas M. Greiner, Ph.D.
Assistant Professor
Department of Anatomy
New York Chiropractic College
Seneca Falls, NY 13148-0800  USA

Phone Office: (315) 568-3183
Phone Lab: (315) 568-3239
Fax: (315) 568-3017
EMail: tgreiner@nycc.edu


My Ph.D. (1989) was on the function of the upper limbs in human walking,
so I have some familiarity with the literature of that time.  I
my main project as:

Webb, David, Russell H. Tuttle and Michael Baksh, "Pendular Activity of
Upper Limbs During Slow and Normal Walking," American Journal of
Anthropology, Vol.93, pp.477-89, 1994.

1.  There are several references in that article which indicate that
arm-swinging is partly active, rather than fully passive.  This is
especially true of retraction (extension) at the shoulder, which
the use of the posterior deltoid muscle.  I strongly recommend finding
the references cited in my paper, and there may be several more by now.
(In fact, I do not recall the Hinrichs paper you mentioned in your
and I would appreciate your giving me the full bibliographic information
so I can look it up.  I have the following article from Hinrichs:
Hinrichs, Richard N., "Regression Equations to Predict Segmental Moments
of Inertia from Anthropometric Measurements: an Extension of the Data of
Chandler, et al.", Journal of Bone and Joint Surgery, Vol.18, No.8,
pp.621-4, 1985. || moments of inertia about transverse & longitudinal

2.  Elftman (among others) was convinced that the major function of
arm-swinging in human walking was indeed to balance rotational forces in
the human trunk, and I recommend his article to you:  Elftman, Herbert,
"The Function of the Arms in Walking", Human Biology, Vol.11, pp.529-35,
1939. Also, by reducing the rotation of the trunk, reaction forces
between the feet and the ground should also be reduced, at a given
since the body would be moving in a more efficient manner.

With regard to your treadmill setup, it would be nice to have a
from which ground reaction forces (GRF's) could be measured, but I don't
know where to get one, and it may be too late for you, anyway.  Others
have used force plates to measure GRF's in various directions, and this
information would be useful to you.  You might try to find the following
article:  Herman, R., R. Wirta, S. Bampton and F.R. Finley, "Human
Solutions for Locomotion:  I. Single Limb Analysis", in Herman,
Stein and Stuart, eds., Neural Control of Locomotion, Advances in
Behavioral Biology, Vol.18, Plenum Press, New York, 1976, pp.13-49. ||
lots of good basic info on v, stride len & freq, jt. angles, grnd reac
F's, stance/swing, some EMG.  Unfortunately, I don't recall anyone
measuring rotational GRF's, only linear ones.

My experience with treadmills has been quite good, overall.  I suspect
that your results will be perfectly usable, but I have a couple of
cautions:  (1)  the treadmill may maintain a more constant speed during
stride than a human normally would (i.e., humans slow down as we "vault"
over the supporting limb, and the treadmill may prevent that); (2)  if
you are using standard video, you will be able to sample at 30 frames
second, which might be a little slow for some of the details of
tending to smooth your kinematic data and obscure some fast
which a force plate would catch.

Finally, I have read some articles about 3-D kinematic models of human
walking, but most of them were more complex (mathematically) than I was
able to use.  You might try these:  Jackson, K.M., J. Joseph and S.J.
Wyard,"A mathematical model of arm swing during human locomotion",
of Biomechanics, Vol.11, pp.277-89, 1978. || EMG's of brach.,
tric., bic., delt., lat.dors., pec. maj.;  surface electrodes only;
incl.'s mus. torque, elastic torque, friction, link len., CoM (a la
Dempster) & accel/ton due to trunk mv/ment; Mochon, Simon and Tomas A.
McMahon, "Ballistic walking:  An improved model," Mathematical
Biosciences, Vol.52, pp.241-60, 1980. || passive model of swing phase is
fairly close to clinical data, o( , ) swing phase is virtually passive;
Bryant, J.T., H.W. Wevers and Philip J. Lowe, "One parameter model for
error in instantaneous centre of rotation measurements", Journal of
Biomechanics, Vol.17, No.5, pp.317-23, 1984. || gives a table and a
to correct experimental values for position of 'instantaneous center of
rotation', determined by X-rays or following pins in bones.

I hope this information is helpful to you.  Good luck with your
and please let me know when and where you publish your work.  I'm very


David Webb

The Hinrichs paper I mentioned is:

Richard N. Hinrichs: Whole Body Movement: Coordination of Arms and Legs
in Walking and Running. Multiple Muscle Systems - Biomechanics and
Movement Organization, Jack M. Winters and Savio L-Y. Woo (editors),
Springer-Verlag, 1990, chapter 45, p.694-705.

some other interesting ones are:

Richard N. Hinrichs and Peter R. Cavanagh and Keith R. Williams: Upper
Extremity Function in Running. I: Center of Mass and Propulsion
Considerations. International Journal of Sport Biomechanics, vol 3,
1987, p.222-241.

Richard N. Hinrichs, Upper Extremity Function in Running. II: Angular
Momentum Considerations, International Journal of Sport Biomechanics,
vol 3, 1987, p.242-263.

Stefan Gruber

Try these sources in regard to arm swing:

Ballesteros, M.L.F., Buchthal, F., & Rosenfalck, P.  (1965) The pattern
muscular activity during the arm swing of natural walking.  Acta
Scand., 63, 296-310.

Cappozzo, A., Figura, F., Leo, T., Marchetti, M.  (1978)  Movements and
mechanical energy changes in the upper part of the human body during
walking.  In E. Asmussen & K. Jorgensen (Eds.), Biomechanics VI-A (pp.
272-279).  Baltimore:  University Park Press.

Duchenne, G.B. (1959)  Physiology of motion (E.B. Kaplan, Translator).
Philadelphia:  W.B. Saunders.

Elftman H.  (1939)  The function of the arms in walking.  Human Biology,

Figura, F., Marchetti, M., & Leo, T.  (1985)  Kinematics of the female
body during ambulation at various speeds.  In D.A. Winter (Ed.)
IX-A (pp. 457-461).  Champaign, IL:  Human Kinetics.

Hogue, R. E. (1969)  Upper-extremity muscular activity at different
and inclines during normal gait.  Physical Therapy, 49, 963-972.

Murray, M.P. (1967, 1970, 1984)  There are 4 studies or articles
by Murray and colleagues.  The sources are as follows:  American Journal
Physical Medicine, Archives of Physical Medicine and Rehabilitation,
Physical Therapy, and Journal of Orthop. Res.

Hope these help.  I took these references from my dissertation "A
three-dimensional analysis of shoulder and pelvic kinematics during

Marilyn Miller
University of Wisconsin - La Crosse
134 Mitchell Hall
La Crosse, WI   54601

Dear Stefan,
I think you should search papers about speed skating.
The skater use the arm swing to accelerate
on short distances (500 m) and skate without arm swing
to save energy on long distances (10 000 m).

I hope it helps


      *   Dr.Christian Peham
      *   email: Christian.Peham@vu-wien.ac.at
      *   Clinic for Orthopaedics in Ungulates
      *   Locomotion Research Group
      *   University of Veterinary Medicine Vienna
      *   Phone: +43-1-250 77/5506; Fax: +43-1-250 77/5590
      *   Josef Baumanngasse 1; A-1210 Wien

Mr. Gruber,

What an interesting question !

Indeed, there is only very little literature available on that area. I
am very interested in that topic also. I am working on a bipedal walking
robot. I am using a ballistic principle (does McGeer/ McMahon ring a
bell ?). One of my goals is to stabilize an upper body in a limit cycle
on the pelvis while walking. The compensating motions of the upper body
with respect to the walking stability are very interesting matter. I am
not yet that far, but I will also start on extending my model with torso
motions soon.

Meanwhile there are two fields to get some information:
1. data from measurements from the biomechanical field:
        Winter D.A., Ruder G.K., MacKinnon C.D., 'Control of balance of
upper body during gait'. in: Multiple Muscle Systems: Biomechanics and
movement organisation (ed. Winters J.M., Woo, S.L-Y,). 1990
Springer-Verlag. pp: 534-541.

2. control theories from the robotic field.
       Work which involves ZMP-theories

I my opinion there has not yet been published a noteworthy paper for the
problem you are trying to answer. So keep on researching, we'll keep in
touch !


Ir. R.Q. van der Linde

Delft University of Technology
Faculty of Design, Engineering and Production
Laboratory of measurement and Control

Mekelweg 2
2628CD DELFT, The Netherlands
tel.nr: +31-152786585
fax:    +31-152784717

Email: R.Q.vanderLinde@wbmt.tudelft.nl

Hello Stefan,

You might want to look at the following papers:

Jackson, KM, Joseph, J and Wyard, SJ; A mathematical model of arm swing
during human locomotion. J Biomech. 1978; 11(6-7): 277-289.

Jackson, KM, Joseph, J and Wyard, SJ; The upper limbs during human
walking. Part I: Sagittal movement. Electromyogr Clin Neurophysiol. 1983
Sep; 23(6): 425-434.

Jackson, KM, Joseph, J and Wyard, SJ; The upper limbs during human
walking. Part 2: Function. Electromyogr Clin Neurophysiol. 1983 Sep;
23(6): 435-446.

I only have ther first paper handy, and can tell you that it's a 2-D
(sagittal plane) model of arm swing.

Hope this helps,


"Alterations in gait resulting from deliberate changes of arm-swing
amplitude and phase"
clinical biomechanics Vol 12,No 7/8,pp.516-521,1997.

                B. Auvinet

Dr. Gruber,
For references regarding your inquiries about arm swing in gait, see our
paper in Clinical Kinesiology:

Sigg, JA, JC Ives, BC Gaba, and GA Sforzo. Effect of arm position on
of normal gait in older persons. Clinical Kinesiology.  51:33-36, 1997.

We tried to make the references as complete as possible concerning
of how and why the arms swing as they do.

Jeff Ives, PhD
Dept. Exercise and Sport Sciences
Ithaca College
Ithaca, NY  14850  USA

Back to FAQ