Bimanual wheelchai p opulsion by people wi h se e e hemipa esis
a e s oke
B endan W. Smi h (Co esponding Au ho )
Depa men o Mechanical Enginee ing
Loyola Ma ymoun Uni e si y
[email p o ec ed]
(310) 258-5433
1 LMU D ., Los Angeles, CA 90045
Diana R. Bueno
COO a Exo i e Labo a o io I+D
[email p o ec ed]
Daniel K. Zonde an
Vice P esiden o Flin Rehabili a ion De ices, LLC
dzonde an@ lin ehab.com
Luis Mon ano
Depa men o Compu e Science and Sys ems Enginee ing
Uni e si y o Za agoza, Spain
[email p o ec ed]
Da id J. Reinkensmeye
Depa men s o Ana omy and Neu obiology, Mechanical and Ae ospace Enginee ing,
Biomedical Enginee ing, and Physical Medicine and Rehabili a ion
Uni e si y o Cali o nia, I ine
d [email protected]
Abs ac (250 wo d max)
Aim: Indi iduals who equi e manual wheelchai s a e s oke a e ypically augh o ambula e
wi h compensa o y p opulsion (i.e. using hei nonpa e ic a m and oo ), isking disuse o he
pa e ic a m. We in es iga ed whe he s oke su i o s can ins ead ambula e in a bimanual, le e -
d i en wheelchai ha equi es he pa e ic a m o con ibu e hal he p opulsi e inpu .
Me hods: Se en een indi iduals wi h ch onic s oke and se e e hemipa esis (uppe ex emi y
Fugl-Meye sco es be ween 10 and 24) pa icipa ed ac oss wo expe imen s. In he i s expe imen
pa icipan s (n=12) ambula ed in s aigh pa hs. In he second expe imen pa icipan s (n=12) also
pe o med u ns, using an imp o ed e sion o he wheelchai ha inco po a ed handb akes.
Twel e unimpai ed con ols also comple ed he second expe imen . Mo ion cap u e and EMG
we e used o compa e biomechanics be ween g oups.
Resul s: Al oge he , 15 o 17 pa icipan s wi h s oke could ambula e 30 m in s aigh pa hs, and
9 o 12 could u n 1800° en i ely unde he powe o hei pa e ic a m. Pa icipan s wi h s oke
exhibi ed la gely heal hy biomechanics, wi h minimal shoulde hiking/leaning o unk inclina ion.
Thei a m muscle EMG pa e ns we e simila o hose used by unimpai ed pa icipan s, excep ing
delayed elbow ex enso ac i a ion.
Conclusions: Indi iduals wi h se e e a m impai men in he ch onic s age o s oke e ain
su icien s eng h and coo dina ion wi h hei pa e ic a m o maneu e bimanual, le e -d i en
wheelchai s. We sugges bimanual, le e d i en p opulsion should be explo ed in s oke
ehabili a ion p ac ice as an al e na i e o compensa o y wheelchai p opulsion, as i has he
po en ial o exe cise heal hy mo emen syne gies, which may in u n help d i e use-dependen
mo o eco e y.
Keywo ds—s oke, a m, ehabili a ion, mobili y, syne gies, mo o con ol, muscle
ac i a ion, wheelchai .
In oduc ion
I con inues o be hypo hesized ha pe o ming a g ea e numbe o p ac ice mo emen s
du ing ehabili a ion a e s oke imp o es uppe limb mo o eco e y,1–4 as p edic ed by he heo y
o use-dependen neu al plas ici y.5 Ye , a majo challenge o bo h s oke pa ien s and s oke
ehabili a ion p o ide s is o success ully achie e he high numbe o p ac ice mo emen s ha is
heo e ically needed o exploi his neu al adap i e capaci y 6. In a sample o uppe limb
ehabili a ion he apy sessions in he US by Lang e al.,7 s oke pa ien s pe o med an a e age o
only 32 uppe limb p ac ice mo emen s pe session, an o de o magni ude less han he 400-600
epe i ions pe session equi ed in animal models o elici signi ican neu al adap a ion.6,8,9
Inc easing he numbe o p ac ice mo emen s is especially di icul o indi iduals wi h se e e
a m weakness (abou 80% o subacu e s oke pa ien s),10 since i is di icul o hem o pe o m
meaning ul exe cises on hei own. Such indi iduals a e ypically excluded om he apy
pa adigms such as cons ain -induced mo emen due o gene ally poo ou comes,11,12 and ha e
di icul y using equipmen such as hand cycles o weigh machines.13 Cu en obo ic he apy
de ices a e e ec i e14 bu oo expensi e o widesp ead use.
Because cos cons ain s limi one-on-one ime wi h he apis s, s oke pa ien s ypically
spend hou s each day si ing in manual wheelchai s wi h hei a ms es ing on hei laps o on a m
es s.15 Al hough a key goal o ea ly s oke ehabili a ion is o help pa ien s eco e he abili y o
walk, s oke su i o s o en use manual wheelchai s du ing inpa ien ehabili a ion o sea ing and
anspo a ion because o ini ial di icul ies wi h balance and walking.16–18 They a e augh o use
hei less-impai ed a m and leg o p opel hei wheelchai , a compensa o y s a egy ha may
con ibu e o asymme ic one.19,20 Al e na i ely, hey may eso o a powe ed wheelchai ha
does no p o ide o uppe ex emi y aining in addi ion o being expensi e and hea y. Ei he
way, ailing o ain he uppe ex emi y can lead o a cycle o lea ned disuse and dec easing
unc ional abili y.21,22 Push im-ac i a ed powe -assis wheelchai s (PAPAWs) help people wi h
a m weakness be e use a manual wheelchai and a e success ul p oduc s,23,24 bu hey canno be
used by people who canno app op ia ely posi ion he a ms o g ip and push he push im.
I people wi h hemipa esis a e s oke could use hei impai ed a ms o help p opel hei
wheelchai s, hey could pe o m many a m p ac ice mo emen s as hey mo ed abou in he no mal
cou se o hei day. Howe e , hey a e a p esen assumed o be unable o p opel manual
wheelchai s wi h hei weak uppe limbs. This assump ion s ems om he obse a ion ha hey do
no ha e enough hand unc ion o g ip and elease he push im, and because hey appea o no
e ain enough s eng h o mo o con ol o make he shoulde and elbow mo emen s needed o
p opel a wheelchai o wa d. I a de ice could assis hemipa e ic indi iduals in wheelchai
p opulsion, indi iduals could sel - ain hei a m as hey mo ed abou , much as cu en manual
wheelchai use s ain hei uppe ex emi y o a high le el o i ness in he no mal cou se o daily
ac i i y.
Mo eo e , he e a e likely o he bene i s o wheelchai use du ing subacu e s oke
ehabili a ion he apy. Acco ding o Blowe ,16 using a wheelchai g an s mobili y o pa ien s who
o he wise migh spend hei ime be ween ac i i ies elega ed o a ixed chai . He a gued ha his
may mo i a e pa ien s by acili a ing socializa ion and enhancing hei senses o eedom and sel -
e icacy and migh also exe cise hei spa ial awa eness. Al hough he e is mixed e idence ha
wheelchai use by s oke pa ien s leads o he de elopmen o asymme ic muscle one,20,25 a
bimanual wheelchai ha engaged bo h a ms symme ically could alle ia e such conce ns.
He e, we s udied i indi iduals wi h hemipa esis a e s oke could use hei mo e-impai ed
a m o powe one wheel o a wheelchai ha u ilizes Le e -Ac ua ed Rehabili a ion and Ambula ion
(LARA). A LARA-based wheelchai allows pa ien s o p opel a manual wheelchai by pumping a
le e wi hou equi ing g asping and eleasing o he push im. The le e is coupled o he wheel by
a a che -like, one-way clu ch ha ansla es o wa d pumps o he le e in o o wa d mo ion o he
chai . In he e sion o he chai es ed he e, he le e was elas ically connec ed o he ame o he
chai , c ea ing a mechanically esonan sys em. Ou p io analysis shows ha ope a ing a LARA-
based wheelchai a i s esonan equency could heo e ically educe he RMS o que equi ed o
sus ained o e -g ound mo emen by upwa ds o 50%.26
Ou LARA-based de ice is an adap a ion o he Resona ing A m Exe cise (RAE), a de ice
designed o epe i i e a m mo emen aining a e s oke.27 RAE consis s o a le e igidly
connec ed o one wheel o a wheelchai , a splin o a aching he impai ed a m o he le e , and
elas ic bands be ween he le e and he ame o he chai , which makes he le e sys em esonan .
To ain wi h RAE, he use mo es he a m o c ea e o wa d and backwa ds ocking wi h ze o ne
mo emen . Mo ing a he esonan equency o RAE ewa ds he use wi h la ge a m mo emen s.
In pilo s udies wi h RAE, s oke pa ien s wi h se e e a m impai men (uppe ex emi y
Fugl-Meye sco es < 15, ou o 66) we e able o s a RAE ocking and we e able o iden i y and
mo e a he sys em’s esonan equency.27 In addi ion, aining epea edly wi h RAE o e h ee
weeks educed a m impai men .28 These indings sugges ed o us ha such indi iduals migh be
able o ambula e o e g ound wi h a le e d i e wheelchai , and, based on he mo i a ional e ec
o mobili y,16 ha such o e g ound ambula ion may help encou age a m mo emen p ac ice. In
addi ion, a p e ious, la ge s udy examined he use o s e eo ypical, epe i i e lexion and
ex ension mo emen s o a m ehabili a ion du ing he acu e phase o s oke and also ound
signi ican , long- e m educ ions in a m impai men .29,30 Thus simple, epe i i e a m mo emen —
such as hose used o p opel a le e d i e chai —appea o p o oke bene icial mo o plas ici y.
A pilo s udy wi h LARA also yielded p omising esul s. Fou pa icipan s wi h se e e
pa esis a e ch onic s oke e ained su icien con ol o he shoulde and elbow o ambula e in a
LARA wheelchai 30 m o e g ound in s aigh lines.31 A second pilo s udy ound ha six
pa icipan s wi h se e e pa esis a e ch onic s oke apidly lea ned o use a e sion o LARA ha
suppo ed inc eased indoo maneu e abili y by using a manually clu ching ansmission ha was
mo e challenging o lea n.32
Howe e , a he onse o his s udy, i emained o be igo ously de e mined o wha ex en
indi iduals wi h se e e hemipa esis a e s oke e ain su icien mo o con ol o maneu e le e
d i e chai s in indoo en i onmen s, and whe he a m exe cise in such chai s is a biomechanically-
app op ia e ac i i y, ha could po en ially con ibu e o ehabili a ion o uppe ex emi y mo o
unc ion wi hou exace ba ing abno mal mo ing pa e ns. He e, we documen wo expe imen s o
answe hese ques ions. The i s expe imen ex ended he pilo esul s31 wi h addi ional subjec s.
The second expe imen measu ed a m join kinema ics and muscle ac i i y while ano he g oup o
ch onic s oke pa ien s pe o med bo h s aigh -line and u ning maneu e s wi h an upda ed
e sion o he LARA wheelchai ha included handb akes. I was o pa icula in e es o
de e mine how ea u es o a m con ol ela ed o ambula ion speed and whe he p opulsion wi h a
le e -d i en chai a e s oke we e associa ed wi h abno mal pa e ns o shoulde and a m
mo emen , which migh make i less desi able as a he apeu ic adju an .
Me hods
Fi s Expe imen
We de eloped a bimanual le e -d i en wheelchai o use as an inpa ien a m ehabili a ion
ool (Figu e 1, op le ). This wheelchai ea u es 1) adjus able a m suppo s a ached o a che ing
le e s ha suppo he pa e ic a m and posi ion i so ha he use ’s hand can push on he le e , 2)
s aps ha a ach he o ea m and hand o he le e s h ough he a m suppo s, and 3) elas ic bands
ha limi he ange o mo ion o he le e and make he le e esona e when pushed a ~1 Hz.
Use s d i e he wheelchai by hy hmically lexing and ex ending hei shoulde and elbow. The
a m suppo s and a che ing le e s ob ia e he need o he use o g asp and elease he push im.
The le e s also gi e use s mechanical ad an age compa ed o di ec ly pushing he push im. The
esonance p o ides u he mechanical ampli ica ion o he use ’s a m o ce.26
We hypo hesized ha his de ice would allow indi iduals wi h se e e a m weakness a e
s oke o use hei pa e ic a m o ambula e o e g ound in a s aigh line (No e: hence o h, we will
e e o he mo e-impai ed a m as he “pa e ic” a m, wi h he implici unde s anding ha he o he ,
“nonpa e ic” a m may also exhibi some impai men o mino pa esis). We es ed his hypo hesis
by asking 12 indi iduals wi h ch onic s oke (mean age 58.8±15.6 yea s; mean du a ion since
s oke 44±41 mon hs) o bimanually p opel a i s e sion o he LARA wheelchai . This g oup
was equally di ided by sex. Thei mean uppe ex emi y Fugl-Meye sco e anged om 10 o 23
(mean sco e 16.8±3.7) ou o he maximum 66, a se e e le el o a m impai men . Indi iduals wi h
FM sco es in his ange gene ally epo ne e using he impai ed a m o daily unc ional ac i i y
on a s anda dized a m use ques ionnai e, he Mo o Ac i i y Log.33,34 None o he pa icipan s had
enough hand unc ion o g asp he push im, and none could bimanually p opel a wheelchai ,
al hough hal o he pa icipan s we e egula use s o manual wheelchai s because o hei
pe sis ing gai impai men .
We asked he pa icipan s o p opel he chai o wa d 3 m in a s aigh line 10 imes, wi h
a sho es b eak be ween each a emp . We measu ed he posi ion o he wheelchai du ing each
ial using a mo ion cap u e sys em (Phase Space) ha acked LEDs placed on he axis o o a ion
o each ea wheel o he wheelchai . We quan i ied he pa icipan s’ de ia ion om he a ge
s aigh line ajec o y as he oo mean squa e o he di e ence om a s aigh line pa h connec ing
he i s and las posi ions o he wheelchai .
Second Expe imen
Following he i s expe imen , we pe o med a second s udy using an imp o ed le e -
d i en wheelchai . In his s udy, indi iduals wi h uppe -ex emi y impai men om ch onic s oke,
and age-ma ched con ol subjec s maneu e ed in s aigh lines and pe o med 180⁰ u ns. This
ime, we measu ed he a m kinema ics and muscle ac i i y. Kinema ic da a we e eco ding using
he same mo ion cap u e sys em, his ime wi h ma ke s placed a key poin s on he pa icipan ’s
body as well as on he wheelchai . Muscle ac i i y da a we e eco ded ia EMG measu emen o
he muscles os ensibly associa ed wi h le e -d i en wheelchai ope a ion.
Pa icipan s
An age ma ched se o 12 pa icipan s wi h hemipa esis a e ch onic s oke (mean age
52.7±4.7 yea s; mean du a ion since s oke 34±18 mon hs) and 12 unimpai ed indi iduals (mean
age 51.3±4.6 yea s) pa icipa ed in he second s udy. Bo h pa icipan g oups consis ed o 5 males
and 7 emales. The s oke pa icipan s had uppe -ex emi y Fugl Meye (FM) sco es anging om
12 o 24 (mean sco e 17.6±1.3) ou o he maximum o 66.
Expe imen al p o ocol
Pa icipan s pe o med epea ed ials o wo di e en cou ses: 1) a 3-m s aigh -line
segmen and 2) a 1.5-m s aigh -line segmen ollowed by a he 180⁰ u n, ei he o he igh o o
he le . S oke pa icipan s pe o med 10 ials o he s aigh cou se, ollowed by 10 ials o he
180⁰ u n cou se. The u n was made in he di ec ion ha posi ioned hei pa e ic a m on he ou side
o he u n, engaging he pa e ic a m while he wheel powe ed by he nonpa e ic a m emained
equency, and e y simila esul s acqui ed. The esul s we e deemed obus o changes in cu o
equency, and 20 Hz was chosen.
EMG esul s we e p ima ily used o in es iga e he ela i e iming o ac i a ion o pai s o
muscles. This ela i e iming was ound o each ial, o a gi en pai o muscles, by inding he
delay ha maximized he c oss-co ela ion o he wo muscles’ smoo hed ac i i y cu es. To
ensu e ha ac i a ions we e compa ed wi hin indi idual le e pumps, he magni ude o his delay
was cons ained o be wi hin one hal o ha ial’s a e age pe iod. This me hod iden i ies he
a e age iming di e ence be ween he cen e s o he wo muscles’ ac i a ions. Obse ing aw da a
e ealed no clea end in muscle ac i a ion du a ions, so p ecise ini ia ion and e mina ion imes
o ac i i y we e no es ima ed.
EMG esul s we e also used o iden i y mo o syne gies, he e de ined as he p inciple
componen s o mo o coo dina ion a he muscle le el. By his de ini ion, syne gy ep esen s he
empo ally and spa ially co ela ed ac i a ions o a muscle.36,37 Each co ela ed muscula syne gy
ec o 𝐜𝐜𝑗𝑗=�c1j, … , cmj� is ac i a ed by i s co esponding neu al signal 𝒉𝒉𝑗𝑗(𝑡𝑡) which modula es
each syne gy. The e a e se e al algo i hms o de e mine 𝑪𝑪 and 𝑯𝑯. In his wo k, we ha e chosen
he me hod o Non-Nega i e Ma ix Fac o iza ion (NNMF) since i has been shown o pe o m
be e han PCA o iden i ying muscle syne gies om EMG da a.38 𝑪𝑪 and H we e ini ialized wi h
andom alues and i e a i ely upda ed using he mul iplica i e ules explained by Ajiboye and
Wei 36.
This me hod equi es selec ing he numbe o syne gies o be ex ac ed. Fo his wo k, we
used he c oss- alida ion me hod,36 which in ol es ex ac ing he minimum numbe o syne gies
(up o he numbe o measu ed muscles, 8 in his wo k) such ha he Va ia ion Accoun ed Fo
(VAF) index be ween he o iginal measu ed EMG (𝑽𝑽) and he econs uc ed EMG ma ix 𝑽𝑽
�=𝑪𝑪 ∙
𝑯𝑯 is o e 0.90. This index is compu ed as 𝑉𝑉AF = 1 − SVV
�
2/SV
2, whe e SVV
�
2 is he sum o he squa es
o he e o s be ween o iginal and p edic ed alue, and SV
2 is he o al a iance o he o iginal EMG
ac i a ion da a.
The p ima y syne gies o each pa icipan du ing each ask we e isually compa ed, and
common syne gies we e de ined o each combina ion o g oup (s oke o con ol) and ask
(s aigh -line o 180°- u n) as he syne gies obse ed mos consis en ly ac oss pa icipan s in ha
g oup o ha ask. A judgemen was made abou which pa icipan ep esen ed he mos ypical
exp ession o each common syne gy o each g oup and ask combina ion. O he pa icipan s in
he same g oup and ask we e conside ed o exp ess a common syne gy i any o hei p ima y
syne gies ma ched wi h ha common syne gy wi h a simila i y index 𝑆𝑆 > 0.7. Al hough a ully
au oma ic me hod o de ining common syne gies was no de eloped, his me hod was epea ed
wi h se e al al e na i e common syne gies o each g oup un il he g ea es ex en o simila i y
was ob ained.
Resul s
Expe imen 1: Can people wi h se e e a m hemipa esis bimanually p opel a le e
d i e chai in a s aigh line?
All 12 indi iduals in he i s s udy had se e e a m impai men (FM be ween 10 and 23).
Ye , all we e able o p opel he wheelchai o wa d a leas a o al o 15 m o e 10 ials. Some
pa icipan s ini ially s uggled wi h he ask, p e en ing hem om eaching he a ge o al
dis ance o 30 m, bu 10 o he 12 indi iduals success ully comple ed he 3-m ask by he las ial
(Figu e 3a). By he las ial, hese 10 indi iduals mo ed in s aigh ajec o ies, while he o he 2
consis en ly u ned owa d hei pa e ic side.
The a e age speed ac oss all subjec s inc eased signi ican ly om he i s o he las ial,
eaching 0.2±0.2 m/s on he las ial (pai ed - es , p = 0.014). The a e age speed o each subjec
ac oss all ials was also signi ican ly co ela ed wi h hei FM sco e o uppe ex emi y
impai men (Figu e 3b; linea eg ession, 2 = 0.44, p = 0.019). The pa icipan ’s de ia ion om
he s aigh -line pa h dec eased ac oss ials, wi h he a e age de ia ion on he en h ial being
0.05 ± 0.05 m, al hough his dec ease in e o was no signi ican . The pa icipan s pe o med an
a e age o 237 a m mo emen s du ing he cou se o he expe imen ( ange = [54, 722]). A ideo
showing he pa icipan s p opelling hemsel es wi h he de ice can be ound a
h p://you u.be/HzZjBIOxLzg. In a pos -expe imen in e iews, se e al pa icipan s no ed ha
achie ing sel -p opulsion wi h hei pa e ic a m was mo i a ing and empowe ing, since i allowed
hem o use his a m in a meaning ul way, o en, hey epo ed, o he i s ime in yea s.
Expe imen 2: Can people wi h se e e a m hemipa esis u n a le e d i e chai wi h
hei hemipa e ic a m?
In he second s udy, 12 indi iduals wi h s oke olun ee ed o p opel an imp o ed e sion
o he wheelchai while unde going mo ion cap u e and EMG eco ding; we also inco po a ed a
u ning maneu e in which pa icipan s squeezed a handb ake wi h hei nonpa e ic hand and
pumped he chai wi h hei pa e ic hand. Se en o he pa icipan s in his second s udy had
p e iously pa icipa ed in he i s s udy desc ibed abo e. All indi iduals in his s udy also had
se e e a m impai men , de ined as FM sco e o 24 o below.
In his s udy, 9 o 12 pa icipan s comple ed he en i e y o he expe imen (10 s aigh -line
ials and 10, 180⁰ u n ials). The o he 3 comple ed he 10 s aigh -line ials bu no mo e han
one 180⁰ u n ial. These pa icipan s ook he oppo uni y hey we e o e ed o end he expe imen
a any ime. One exp essed ha hey we e expe iencing subs an ial p e-exis ing shoulde pain. All
12 age-ma ched con ols comple ed he ull 30 ials. Equipmen p oblems caused loss o mo ion
cap u e da a o one pe son wi h s oke and one age-ma ched con ol. Excluding he pa icipan s’
whose da a was los , he mean age o he s oke pa icipan s was 52.5 ± 5.2 yea s (49.4 ± 5.8 yea s
o hose who comple ed all 20 ials) and he mean age o he unimpai ed con ols was 53.1 ± 4.8
yea s.
Consis en wi h esul s om he i s s udy, all pa icipan s wi h s oke in his expe imen
we e able o p opel hemsel es ac oss 3 m o 10 ials (Figu e 3). Thus, combining he i s and
second expe imen , 15 o 17 pa icipan s wi h FM sco es less han 24 demons a ed hey could
p opel a bimanual, le e -d i en wheelchai 30 m. A e p opelling he chai hese i s 30 m, 9 o
12 pa icipan s in he second expe imen also p opelled hemsel es ac oss 1.5 m and pe o med a
180⁰ u n o 10 ials; hese u ns we e d i en solely by hei pa e ic a ms. In o al, hese
pa icipan s demons a ed hey could bo h p opel he chai 45 m and u n in he chai a o al o
1800° (i.e. 5 ull o a ions).
Biomechanics Pa 1: How did he indi iduals wi h s oke achie e le e -based
p opulsion?
Wheelchai speed and le e pumping beha io a ied subs an ially be ween pa icipan s in
he second expe imen . S oke pa icipan s we e on a e age subs an ially slowe han unimpai ed
con ols, ye , in his second expe imen , he e was only a ma ginal co ela ion be ween FM sco e
and a e age speed (Figu e 4a; linea eg ession, p = 0.22). The lack o signi ican co ela ion may
be explained in ha al hough he 3 slowes pa icipan s had among he lowes FM sco es, he 3
as es pa icipan s did no ha e dis inc ly high FM sco es. Gi en ha eg ession analysis is poo ly
condi ioned o da a se s whe e a ew pa icipan s pe o m d ama ically di e en ly han he es , i
is unclea i he co ela ion in he i s expe imen (wi hou handb akes) and lack o co ela ion in
he second expe imen (wi h handb akes) desc ibes a ue e ec o handb akes on ope a ing he
LARA wheelchai , o i i is simply an a i ac o which pa icipan s happened o achie ed
especially as ambula ion du ing each expe imen .
The ques ions hen a ise: wha ac o s do explain how indi iduals wi h hemipa esis a e
ch onic s oke achie e le e -based p opulsion, and wha explains as e o slowe p opulsion
speeds i speed is no simply a unc ion o impai men le el. Two ac o s we e in es iga ed: he
equencies a which pa icipan s pumped he le e s, and he anges o mo ion o elbow mo emen
hey exhibi ed. Simila esul s we e obse ed du ing bo h s aigh -line and 180⁰- u n ials. Resul s
om s aigh -line ials a e shown he e. Resul s om 180⁰- u n ials we e simila and a e no
shown he e.
Simila wide anges o pump equency we e obse ed o bo h s oke pa icipan s and
con ols (Figu e 4b). Pump equency was co ela ed wi h speed, bu his ela ionship di e ed
be ween s oke pa icipan s and con ols (Figu e 5b), implying ha o he dynamics we e a play.
Elbow ange o mo ion, on he o he hand, was mildly co ela ed wi h FM sco e, and subs an ially
lowe o s oke pa icipan s han o con ols (Figu e 4c). Elbow ange o mo ion was also loosely
co ela ed wi h speed (Figu e 5c), sugges ing a p edic i e ela ionship mode a ed by o he ac o s.
Howe e , combining hese wo ac o s p oduced a much mo e p edic i e ela ionship, speci ically
he p oduc o pump equency and elbow ange o mo ion was highly co ela ed wi h speed
(Figu e 5d). O no e, elbow ange o mo ion and pump equency we e hemsel es no co ela ed
(Figu e 5e). These esul s imply ha wo unco ela ed mo o capaci ies de e mine an indi idual’s
ambula ion speed: 1) hei abili y o quickly ac i a e and deac i a e hei a m muscles and 2) hei
abili y o ex end he elbow h ough a wide ac i e ange o mo ion.
We suspec ed ha some pa icipan s migh ha e achie ed as e speeds using abno mal
compensa o y mo emen s, such as unk inclina ion o shoulde leaning o hiking. On a e age,
s oke pa icipan s showed no inc eased unk mo ion o e con ols; on an indi idual basis,
howe e , he h ee slowes pa icipan s exhibi ed la ge unk mo ion (Figu e 4d). This sugges s
ha compensa ion by unk inclina ion is a means by which slowe pa icipan s achie ed
mo emen , bu mos pa icipan s did no employ his compensa ion. Ampli ude o shoulde
ansla ion, which would be expec ed o cap u e leaning o hiking, di e ed minimally be ween
s oke pa icipan s and con ols, and was unco ela ed wi h FM sco e (Figu e 4e). No was g ea e
shoulde mo emen obse ed o slowe pa icipan s. This sugges s ha le e -based p opulsion
does no encou age abno mal shoulde mo emen in indi iduals wi h hemipa esis a e ch onic
s oke, e en o hose o whom p opulsion is di icul .
Biomechanics Pa 2: How does s oke a ec he muscle ac i a ion pa e ns used
du ing le e -based p opulsion?
In gene al, bo h ch onic s oke and con ol pa icipan s exhibi ed a cascade o muscle
ac i a ion du ing only he push phase o le e -based p opulsion, and minimal ac i a ion du ing he
eco e y phase when he mo emen was assis ed by he e u n ac ion o he sp ing a ached o he
le e (Figu e 6). This di e s om he subs an ial muscle ac i i y obse ed du ing bo h push and
eco e y phase du ing push- im wheelchai p opulsion.39,40 This inding implies ha LARA
educes in he complexi y o ope a ing manual wheelchai . This aises he ques ions o how he
iming o hese cascading muscle pulses compa es be ween s oke pa icipan s and con ols and
whe he his migh help explain di e ences in hei p opulsion speed. We also in es iga ed he
occu ence o common muscle ac i a ion syne gies ac oss pa icipan s.
The ela i e ac i a ion iming be ween he iceps and o he muscles bes di e en ia ed he
ac i a ion pa e ns o s oke and con ol pa icipan s (Figu e 7a). Fo con ols pa icipan s, he
ac i i y o he iceps signi ican ly p eceded ha o mos o he muscles ( - es , p alues shown in
Table 2 “Con ols”), sugges ing hey ini ia ed le e mo emen by ac i a ing he iceps. Fo
ch onic s oke pa icipan s, ac i a ion o he iceps did no signi ican ly p ecede ha o any o he
muscle ( - es , p alues shown in Table 2 “S oke”). Compa ing hese ela i e imings be ween
s oke pa icipan s and con ols, i was ound ha he imings o all muscles excep he an e io
del oid we e signi ican ly di e en be ween g oups (pai ed - es , nonequal a iance, p alues
shown in Table 2 “Di e ence”). This compa ison suppo s ha a signi ican shi in iming o he
iceps occu s ollowing s oke, and ha i no longe leads he o he muscles.
The ela i e iming o he biceps and iceps was in es iga ed as a p edic o o pa icipan s’
p opulsion speed, based on ou inding ha ampli ude o elbow mo emen was co ela ed wi h
pa icipan s’ p opulsion speed. Elbow ex ension os ensibly equi es ac i a ing he iceps wi hou
in e e ence o he biceps. Ou obse a ion ha he iceps no longe leads he biceps a e s oke
migh help explain ch onic s oke pa icipan s’ educed elbow ex ension ela i e o con ol
pa icipan s. To elucida e his connec ion, each pa icipan ’s bicep ac i a ion iming was compa ed
o hei elbow ampli ude (Figu e 7b). A ma ginally signi ican end was ound be ween biceps
iming and elbow ampli ude o s oke pa icipan s (linea eg ession, R2 = ?, p = 0.052). A non-
signi ican co ela ion exis ed o con ol pa icipan s (linea eg ession, p = 0.15). Ex apola ing
his end would sugges ha someone wi h hemipa esis a e ch onic s oke could inc ease hei
ange o elbow mo emen , and hei p opulsion speed by lea ning o ac i a e hei iceps
p eemp i ely o hei biceps and o he a m muscles.
Along wi h delayed iceps ac i a ion, s oke pa icipan s appea o ha e exhibi ed ea ly
apezius ac i a ion (no shown di ec ly, bu implied by Figu e 7a). Using a simila analysis as
desc ibed o he iceps, he apezius was ound o signi ican ly lead he pec o alis majo ( - es ,
p = 0.030) and he an e io del oid ( - es , p = 0.030) o ch onic s oke pa icipan s. This may
sugges ha s oke pa icipan s compensa e o hei impai men by p eemp ing hei a m
mo emen wi h posi ioning o b acing o he shoulde , e en i kinema ic da a did no show
inc eased ampli ude o shoulde mo emen .
EMG syne gy analysis sugges ed la gely simila muscle ac i a ion pa e ns ac oss he wo
g oups, wi h some disc epancy in he exp ession o syne gies in ol ing he iceps. Fo each
pa icipan , wo p ima y syne gies on a e age desc ibed mo e han he 90% o hei a iabili y in
muscle ac i a ion in each ask condi ions. The syne gies o each pa icipan we e isually
compa ed, and wo common syne gies (Figu e 7c) we e de ined o each combina ion o g oup
(s oke o con ol) and ask (s aigh -line o 180°- u n). Fo hose who exp essed hem, hese
common syne gies we e ema kably simila be ween s oke pa icipan s and con ols du ing bo h
he s aigh -line and 180⁰- u n ials (Figu e 7c). O e all, hese common syne gies we e exp essed
wi h 52% consis ency ac oss s oke pa icipan s (i.e. as one o hei wo p ima y syne gies) and
wi h 63% consis ency ac oss con ols (Table 3). This sugges s ha he mos common muscle
ac i a ion pa e ns we e la gely sha ed by s oke pa icipan s and con ols, bu ha subs an ial
a ia ion in hese pa e ns exis ed wi hin bo h g oups. The common syne gies ha mos in ol ed
he iceps (𝑤𝑤1 o s aigh -line ials and 𝑤𝑤2 o 180⁰- u n ials), we e also exp essed wi h 52%
consis ency ac oss s oke pa icipan s, bu wi h 79% consis ency ac oss con ols. Tha is, iceps
ac i a ion was gene ally syne gized wi h he same muscles o con ols. In con as , only abou
hal o s oke pa icipan s ac i a ed hei iceps wi hin any such sha ed such syne gy. Tha syne gy
did, howe e , closely esemble he iceps ac i a ion syne gy exp essed by con ols.
In insic Mo i a ion: A e indi iduals wi h s oke mo i a ed by hei pe o mance in
le e -d i en wheelchai s?
All pa icipan s answe ed a 10-ques ion su ey wi h ele an ques ions d awn om he
In insic Mo i a ion In en o y (IMI), which has been used in s udies o obo -aided a m
ehabili a ion.41 Ques ions ela ed o mo i a ion, e o , and pe cei ed compe ence (Table 4). All
ques ions p omp ed a nume ic esponse be ween 1 o “s ong disag eemen ” and 7 o “s ong
ag eemen ”. Pe haps no su p isingly, s oke pa icipan s belie ed ha p ac icing longe wi h he
wheelchai would be mo e bene icial o hem han did con ol pa icipan s (pai ed - es , p =
0.0043). They also belie ed ha ambula ing in he chai equi ed mo e e o han did con ol
pa icipan s (pai ed - es , 0.025). No signi ican di e ence was ound be ween he esponses o
s oke and con ol pa icipan s on any o he emaining ques ions. Nei he g oup el pa icula ly
anxious o bo ed du ing he expe imen . Bo h g oups indica ed ha hey wan ed o do he ac i i y,
possibly consis en wi h he esul s ha bo h g oups ound he ask un and el ha hei
pa icipa ion was impo an . Bo h g oups we e sa is ied wi h hei pe o mance despi e being
somewha less con iden ha hey we e good a he ac i i y. Bo h g oups sha ed a simila le el o
con idence ha hey became compe en a he ac i i y du ing his session. Su p isingly s oke
pa icipan s did no eel signi ican ly less compe en a ambula ing in he chai han did con ol
pa icipan s.
Discussion
The assump ion ha indi iduals wi h se e e a m weakness a e s oke canno bimanually
p opel hemsel es in a manual wheelchai is inco ec . He e, we ound ha a o al o 15 o 17
pa icipan s wi h s oke (10 o 12 om he i s expe imen , and all 5 new pa icipan s om he
second) p opelled a manual, le e -d i en wheelchai 30 o 45 m in a s aigh line, demons a ing
hei abili y o coo dina e he mo emen o hei a ms o mo e in a s aigh line. Using a handb ake
du ing he second expe imen , 9 o 12 s oke pa icipan s comple ed en 180⁰ u ns ( o a o al o
1800⁰) comple ely unde he powe o he pa e ic a m. Thus, wi h app op ia e mechanical
adap a ions, e en indi iduals wi h o he wise li le unc ional use o he a m gene a ed enough o ce
in he equi ed spa io empo al pa e n o p opel and u n a le e d i e chai . This con i ms ou
p elimina y indings ha ch onic s oke su i o s e ain his capaci y,31,32 bu in a la ge sample.
Mo eo e , while ambula ing du ing his s udy, pa icipan s egula ly pe o med be ween 400-600
a m mo emen s du ing each 1-hou session. This dose o exe cise exceeds ha e en o ecen
a emp s o p o ide a high dose o mo emen p ac ice in s oke ehabili a ions.2 I emains o be
seen i his so o a m mo ion is he apeu ically meaning ul, bu his s udy yields some p omising
esul s abou his possibili y, coupled wi h p e ious s udies on he he apeu ic e ec i eness o
simple, epe i i e a m mo emen REFS. We discuss now se e al ea u es o he le e -d i en
p opulsion, he biomechanics o he a m and unk, and hen u u e di ec ions o esea ch.
Fea u es o le e -d i en wheelchai ambula ion
The s oke pa icipan s mo ed a abou hal he speed o unimpai ed con ols du ing
s aigh line mo emen and o a ed he chai a abou one- hi d he speed o con ols du ing u ns.
While hei u ning was slowe gene ally, he as e hal o s oke pa icipan s mo ed in s aigh
lines a abou he same speed as he slowe hal o con ols. Toge he hese esul s indica e ha
Con lic o In e es
Daniel Zonde an and Da id Reinkensmeye a e co- ounde s o and hold equi y in Flin
Rehabili a ion De ices, a company ha is comme cializing ehabili a ion echnology. Daniel
Zonde an is cu en ly employed a Flin , and Da id Reinkensmeye has ecei ed paymen o
consul ing om Flin . Da id Reinkensmeye holds equi y in Hocoma, a manu ac u e o
ehabili a ion echnology. The e ms o Da id Reinkensmeye ’s in e es s ha e been e iewed by
he U.C. I ine Con lic o In e es commi ee.
Re e ences
1. Lohse KR, Lang CE, Boyd LA. Is mo e be e ? Using me a-da a o explo e dose- esponse
ela ionships in s oke ehabili a ion. S oke 2014; 45: 2053–2058.
2. Lang CE, S ube MJ, Bland MD, e al. Dose esponse o ask-speci ic uppe limb aining
in people a leas 6 mon hs pos s oke: A phase II, single-blind, andomized, con olled ial.
Ann Neu ol 2016; 80: 342–354.
3. Bell JA, Wolke ML, O ez RC, e al. T aining in ensi y impac s mo o ehabili a ion e icacy
ollowing unila e al ischemic insul o he senso imo o co ex in C57BL/6 mice.
Neu o ehabil Neu al Repai 2015; 29: 590–598.
4. F ench B, Thomas LH, Lea hley MJ, e al. Repe i i e ask aining o imp o ing unc ional
abili y a e s oke BT - Coch ane Da abase Sys Re . Coch ane Da abase Sys Re . Epub
ahead o p in 2016. DOI: 10.1002/14651858.CD006073.pub3.www.coch anelib a y.com.
5. Sawaki L. Use-dependen plas ici y o he human mo o co ex in heal h and disease. IEEE
Eng Med Biol Mag 2005; 24: 36–9.
6. Je e s MS, Ka hikeyan S, Gomez-Smi h M, e al. Does S oke Rehabili a ion Really
Ma e ? Pa B: An Algo i hm o P esc ibing an E ec i e In ensi y o Rehabili a ion.
Neu o ehabil Neu al Repai 2018; 32: 73–83.
7. Lang CE, Macdonald JR, Reisman DS, e al. Obse a ion o amoun s o mo emen p ac ice
p o ided du ing s oke ehabili a ion. A ch Phys Med Rehabil 2009; 90: 1692–8.
8. Kleim JA, Ba bay S, Nudo RJ. Func ional Reo ganiza ion o he Ra Mo o Co ex
Following Mo o Skill Lea ning. J Neu ophysiol 1998; 80: 3321–3325.
9. Bi kenmeie RL, P age EM, Lang CE. T ansla ing animal doses o ask-speci ic aining
o people wi h ch onic s oke in 1-hou he apy sessions: A p oo -o -concep s udy.
Neu o ehabil Neu al Repai 2010; 24: 620–635.
10. Langho ne P, Coupa F, Pollock A. Mo o eco e y a e s oke: a sys ema ic e iew. Lance
Neu ol 2009; 8: 741–754.
11. Taub E, Uswa e G, King DK, e al. A placebo-con olled ial o cons ain -induced
mo emen he apy o uppe ex emi y a e s oke. S oke 2006; 37: 1045–1049.
12. Wol SL, Wins ein CJ, Mille JP, e al. E ec o Cons ain -Induced Mo emen . 2006; 296:
2095–2104.
13. Zeh EP, Loadman PM, Hundza SR. Neu al con ol o hy hmic a m cycling a e s oke. J
Neu ophysiol 2012; 108: 891–905.
14. Meh holz J, Pohl M, Pla z T, e al. Elec omechanical and obo -assis ed a m aining o
imp o ing ac i i ies o daily li ing, a m unc ion, and a m muscle s eng h a e s oke.
Coch ane Da abase Sys Re . Epub ahead o p in 3 Sep embe 2018. DOI:
10.1002/14651858.CD006876.pub5.
15. Je e DU, La ham NK, Smou RJ, e al. Physical he apy in e en ions o pa ien s wi h
s oke in inpa ien ehabili a ion acili ies. Phys The 2005; 85: 238–48.
16. Blowe P. The ad an ages o he ea ly use o wheelchai s in he ea men o hemiplegia.
Clin Rehabil 1988; 2: 323–325.
17. Ch is ina M. Mumma. Pe cei ed Losses Following S oke. Rehabil Nu s 1986; 11: 19–24.
18. Ba ke DJ, Reid D, Co C. Accep ance and meanings o wheelchai use in senio s oke
su i o s. Am J Occup The 2004; 58: 221–230.
19. Ba e JA, Wa kins C, Plan R, e al. The COSTAR wheelchai s udy: a wo-cen e pilo
s udy o sel -p opulsion in a wheelchai in ea ly s oke ehabili a ion. Collabo a i e S oke
Audi and Resea ch. Clin Rehabil 2001; 15: 32–41.
20. Ashbu n A, Lynch M. Disad an ages o he ea ly use o wheelchai s in he ea men o
hemiplegia. Clin Rehabil 1988; 2: 327–331.
21. Coope R, Qua ano L, Axelson P, e al. Resea ch on physical ac i i y and heal h among
people wi h disabili ies: a consensus s a emen . J Rehab Res De 1999; 36: 142–154.
22. Schweigho e N, Han CE, Wol SL, e al. A unc ional h eshold o long- e m use o hand
and a m unc ion can be de e mined: p edic ions om a compu a ional model and
suppo ing da a om he Ex emi y Cons ain -Induced The apy E alua ion (EXCITE)
T ial. Phys The 2009; 89: 1327–1336.
23. Coope RA, Fi zge ald SG, Boninge ML, e al. E alua ion o a push im-ac i a ed, powe -
assis ed wheelchai . A ch Phys Med Rehabil 2001; 82: 702–8.
24. Nash MS, Koppens D, an Haa en M, e al. Powe -Assis ed Wheels Ease Ene gy Cos s and
Pe cep ual Responses o Wheelchai P opulsion in Pe sons Wi h Shoulde Pain and Spinal
Co d Inju y. A ch Phys Med Rehabil 2008; 89: 2080–2085.
25. Pome oy VM, Mickelbo ough J, Hill E, e al. A hypo hesis: sel -p opulsion in a wheelchai
ea ly a e s oke migh no be ha m ul. Clin Rehabil 2003; 17: 174–80.
26. Zonde an DK, Smi h B, Reinkensmeye DJ. Le e -ac ua ed esonance assis ance (LARA):
A wheelchai -based me hod o uppe ex emi y he apy and o e g ound ambula ion o
people wi h se e e a m impai men . In: IEEE In e na ional Con e ence on Rehabili a ion
Robo ics. 2013.
27. Zonde an DK, Pala ox L, He nandez J, e al. The Resona ing A m Exe cise : design and
pilo es ing o a mechanically passi e ehabili a ion de ice ha mimics obo ic ac i e
assis ance. J Neu oenginee ing Rehabil 2013; 10: 39.
28. Zonde an DK, Augsbu ge R, Bodenhoe e B, e al. Machine-based, sel -guided home
he apy o indi iduals wi h se e e a m impai men a e s oke: A andomized con olled
ial. Neu o ehabil Neu al Repai 2015; 29: 395–406.
29. Feys HM, De Wee d WJ, Selz BE, e al. E ec o a he apeu ic in e en ion o he
hemiplegic uppe limb in he acu e phase a e s oke: a single-blind, andomized, con olled
mul icen e ial. S oke 1998; 29: 785–792.
30. Feys H, De Wee d W, Ve beke G, e al. Ea ly and epe i i e s imula ion o he a m can
subs an ially imp o e he long- e m ou come a e s oke: a 5-yea ollow-up s udy o a
andomized ial. S oke 2004; 35: 924–9.
31. Smi h BW, Zonde an DK, Lo d TJ, e al. Feasibili y o a bimanual, le e -d i en
wheelchai o people wi h se e e a m impai men a e s oke. Con P oc . Annu In Con
IEEE Eng Med Biol Soc IEEE Eng Med Biol Soc Annu Con 2014; 2014: 5292–5295.
32. Sa igul-Klijn Y, Lobo-P a J, Smi h BW, e al. The e is plen y o oom o mo o lea ning
a he bo om o he Fugl-Meye : Acquisi ion o a no el bimanual wheelchai skill a e
ch onic s oke using an unmasking echnology. IEEE In Con Rehabil Robo 2017; 50–55.
33. S ewa JC, C ame SC. Pa ien - epo ed measu es p o ide unique insigh s in o mo o
unc ion a e s oke. S oke 2013; 44: 1111–6.
34. an de Lee JH, Becke man H, Knol DL, e al. Clinime ic p ope ies o he Mo o Ac i i y
Log o he assessmen o a m use in hemipa e ic pa ien s. S oke 2004; 35: 1404–1410.
35. De Luca CJ, Donald Gilmo e L, Kuzne so M, e al. Fil e ing he su ace EMG signal:
Mo emen a i ac and baseline noise con amina ion. J Biomech 2010; 43: 1573–1579.
36. Ajiboye AB, Wei RF. Muscle syne gies as a p edic i e amewo k o he EMG pa e ns
o new hand pos u es. J Neu al Eng; 6. Epub ahead o p in 2009. DOI: 10.1088/1741-
2560/6/3/036004.
37. Sa a ynia S, To es-O iedo G, Ting L. Muscle Syne gies: Implica ions o Clinical
E alua ion and Rehabili a ion o Mo emen . Top Spinal Co d Inj Rehabil 2011; 17: 16–24.
38. T esch MC, Cheung VCK, D’A ella A. Ma ix Fac o iza ion Algo i hms o he
Iden i ica ion o Muscle Syne gies: E alua ion on Simula ed and Expe imen al Da a Se s.
J Neu ophysiol 2006; 95: 2199–2212.
39. Schan z P, Bjö kman P, Sandbe g M, e al. Mo emen and muscle ac i i y pa e n in
wheelchai ambula ion by pe sons wi h pa a-and e aplegia. Scand J Rehabil Med 1999;
31: 67–76.
40. Chow JW, Millikan TA, Ca l on LG, e al. Kinema ic and Elec omyog aphic Analysis o
Wheelchai P opulsion on Ramps o Di e en Slopes o Young Men Wi h Pa aplegia. A ch
Phys Med Rehabil 2009; 90: 271–278.
41. No ak D, Nagle A, Kelle U, e al. Inc easing mo i a ion in obo -aided a m ehabili a ion
wi h compe i i e and coope a i e gameplay. J Neu oeng Rehabil 2014; 11: 64.
42. Michaelsen SM, Dannenbaum R, Le in MF. Task-speci ic aining wi h unk es ain on
a m eco e y in s oke: Randomized con ol ial. S oke 2006; 37: 186–192.
43. Ci s ea MC, Le in MF. Compensa o y s a egies o eaching in s oke. B ain 2000; 123:
940–953.
44. Ba ke RN, B aue S, Ca son R. T aining-induced changes in he pa e n o iceps o biceps
ac i a ion du ing eaching asks a e ch onic and se e e s oke. Exp B ain Res 2009; 196:
483–496.
45. Veege HEJ, Van De Woude LH V, Rozendal RH. Load on he Uppe Ex emi y in Manual
Wheelchai P opulsion. J EIec omyog aphy Kinesiol 1991; I: 270–280.
46. Buneo CA, Soech ing JF, Flande s M. Muscle ac i a ion pa e ns o eaching: he
ep esen a ion o dis ance and ime. J Neu ophysiol 1994; 71: 1546–1558.
47. Lum PS, Bu ga CG, Sho PC. E idence o s eng h imbalances as a signi ican con ibu o
o abno mal syne gies in hemipa e ic subjec s. Muscle and Ne e 2003; 27: 211–221.
48. Beebe JA, Lang CE. Absence o a p oximal o dis al g adien o mo o de ici s in he uppe
ex emi y ea ly a e s oke. Clin Neu ophysiol 2008; 119: 2074–2085.
49. Yang YS, Koon z AM, T iolo RJ, e al. Su ace elec omyog aphy ac i i y o unk muscles
du ing wheelchai p opulsion. Clin Biomech 2006; 21: 1032–1041.
Table 1. Mo ion cap u e ma ke posi ions.
Table 2. P- alues o muscle ac i a ion iming ela i e o iceps.
Bold indica es p < 0.05.
Table 3. Exp ession o he i s wo common kine ic syne gies ac oss pa icipan s.
Con ol
Pa icipan s
Line
Tu n
𝑤𝑤1
𝑤𝑤2
𝑤𝑤1
𝑤𝑤2
S1
C1
C2
C1
C2
S2
C1
C2
C1
C2
S3
C2
C2
S4
C2
C2
S5
C1
C2
S6
C2
C1
C2
S7
C2
C2
S8
C2
C1
C2
S9
C1
C2
S10
C2
C2
S11
C1
C1
S12
C1
C1
S oke
Pa icipan s
Line Tu n
𝑤𝑤1
𝑤𝑤2
𝑤𝑤1
𝑤𝑤2
P1
C2
C2
P2
C2
C2
P3
C1
P4
C1
C2
C1
C2
P5
C1
C2
C1
P6
C1
P7
C1
C2
C1
P8
C1
C1
P9
C1
C2
P10
C1
C1
P11
P12
𝑊𝑊1 and 𝑊𝑊2: Fi s wo (i.e. p ima y) common syne gies.
𝐶𝐶𝑖𝑖: Indica es ha he pa icipan exp essed 𝑊𝑊1 o 𝑊𝑊2 as hei 𝑖𝑖- h indi idual syne gy.
Emp y cell: The pa icipan did no exp ess his common syne gy as ei he hei 1s o 2nd indi idual syne gy.
G ey cell: The pa icipan did no comple e he espec i e ask.
Figu e 6. Raw EMG om ch onic s oke and unimpai ed pa icipan . A 20-400Hz bandpass il e was used o
condi ion he signal and emo e mo ion a i ac , a 125ms window mo ing RMS was used o es ima e muscle ac i i y.
S oke Pa icipan
Con ol Pa icipan
Figu e 7. A: Muscle ac i a ion iming ela i e o iceps. A e age ac i a ion imings o selec ed muscles ela i e
o he ac i a ion o he iceps (a = 0). Posi i e alues indica e ha he muscle was ac i a ed la e han he iceps.
Timing is measu ed in e ms o ac ion o ha a e age pe iod leng h du ing he ial. Unimpai ed con ols led wi h
iceps ac i a ion, a pa e n no main ained a e s oke. Following s oke, he apezius ends o lead he mo emen .
B: Co ela ion o biceps- iceps iming wi h elbow mo emen . Ch onic s oke pa icipan s’ iming o he biceps
ela i e o he iceps was co ela ed wi h he ampli ude o hei elbow mo emen s. A simila bu weake co ela ion
was ound o unimpai ed con ols. This esul is consis en wi h he heo y ha la ge elbow ampli udes equi e he
iceps o ex end he elbow du ing he a m mo emen , bu ha his ex ension equi es he iceps o con ac wi hou
co-con ac ion o he biceps. C: The common muscle syne gies 𝒘𝒘𝟏𝟏 and 𝒘𝒘𝟐𝟐, o each expe imen al ask. Ba s
indica e he mean no malized ac i a ion o each muscle in ol ed in he syne gy ac oss subjec s who exp essed ha
syne gy. E o ba s indica e he s anda d de ia ion abou hese means. Syne gies o s oke pa icipan s a e shown in
cyan, and syne gies o con ol pa icipan s in blue
A
B
C
