motor patterns in human walking and running

7/30/2019 Motor Patterns in Human Walking and Running

1/13

doi:10.1152/jn.00081.200695:3426-3437, 2006. First published 22 March 2006;J NeurophysiolG. Cappellini, Y. P. Ivanenko, R. E. Poppele and F. LacquanitiMotor Patterns in Human Walking and Running

You might find this additional info useful...

62 articles, 28 of which can be accessed free at:This article cites

http://jn.physiology.org/content/95/6/3426.full.html#ref-list-1

26 other HighWire hosted articles, the first 5 are:This article has been cited by

[PDF][Full Text][Abstract]

, July , 2011; 106 (1): 91-103.J NeurophysiolFranois Hug, Nicolas A. Turpin, Antoine Couturier and Sylvain DorelConsistency of muscle synergies during pedaling across different mechanical constraints

[PDF][Full Text][Abstract], August , 2011; 106 (2): 999-1015.J Neurophysiol

Stacie A. Chvatal, Gelsy Torres-Oviedo, Seyed A. Safavynia and Lena H. Tingstepping postural behaviorsCommon muscle synergies for control of center of mass and force in nonstepping and

[PDF][Full Text][Abstract], September , 2011; 106 (3): 1525-1536.J Neurophysiol

LacquanitiFrancesca Sylos Labini, Yuri P. Ivanenko, Germana Cappellini, Silvio Gravano and FrancescounloadingSmooth changes in the EMG patterns during gait transitions under body weight

[PDF][Full Text][Abstract], September , 2011; 106 (3): 1363-1378.J Neurophysiol

Jinsook Roh, Vincent C. K. Cheung and Emilio BizziModules in the brain stem and spinal cord underlying motor behaviors

[PDF][Full Text][Abstract], October 1, 2011; 214 (19): 3305-3314.J Exp Biol

Nicolas A. Turpin, Arnaud Guvel, Sylvain Durand and Franois HugFatigue-related adaptations in muscle coordination during a cyclic exercise in humans

including high resolution figures, can be found at:Updated information and services

http://jn.physiology.org/content/95/6/3426.full.html

can be found at:Journal of NeurophysiologyaboutAdditional material and information

http://www.the-aps.org/publications/jn

This infomation is current as of November 8, 2011.

American Physiological Society. ISSN: 0022-3077, ESSN: 1522-1598. Visit our website at http://www.the-aps.org/.(monthly) by the American Physiological Society, 9650 Rockville Pike, Bethesda MD 20814-3991. Copyright 2006 by the

publishes original articles on the function of the nervous system. It is published 12 times a yearJournal of Neurophysiology
http://jn.physiology.org/content/95/6/3426.full.html#ref-list-1http://jn.physiology.org/content/106/1/91.full.pdfhttp://jn.physiology.org/content/106/1/91.full.pdfhttp://jn.physiology.org/content/106/1/91.full.htmlhttp://jn.physiology.org/content/106/1/91.abstract.htmlhttp://jn.physiology.org/content/106/1/91.full.htmlhttp://jn.physiology.org/content/106/1/91.full.pdfhttp://jn.physiology.org/content/106/1/91.abstract.htmlhttp://jn.physiology.org/content/106/1/91.full.htmlhttp://jn.physiology.org/content/106/2/999.full.pdfhttp://jn.physiology.org/content/106/2/999.full.pdfhttp://jn.physiology.org/content/106/2/999.full.htmlhttp://jn.physiology.org/content/106/2/999.abstract.htmlhttp://jn.physiology.org/content/106/2/999.full.htmlhttp://jn.physiology.org/content/106/2/999.full.pdfhttp://jn.physiology.org/content/106/2/999.abstract.htmlhttp://jn.physiology.org/content/106/2/999.full.htmlhttp://jn.physiology.org/content/106/3/1525.full.pdfhttp://jn.physiology.org/content/106/3/1525.full.pdfhttp://jn.physiology.org/content/106/3/1525.full.htmlhttp://jn.physiology.org/content/106/3/1525.abstract.htmlhttp://jn.physiology.org/content/106/3/1525.full.htmlhttp://jn.physiology.org/content/106/3/1525.full.pdfhttp://jn.physiology.org/content/106/3/1525.abstract.htmlhttp://jn.physiology.org/content/106/3/1525.full.htmlhttp://jn.physiology.org/content/106/3/1363.full.pdfhttp://jn.physiology.org/content/106/3/1363.full.pdfhttp://jn.physiology.org/content/106/3/1363.full.htmlhttp://jn.physiology.org/content/106/3/1363.abstract.htmlhttp://jn.physiology.org/content/106/3/1363.full.htmlhttp://jn.physiology.org/content/106/3/1363.full.pdfhttp://jn.physiology.org/content/106/3/1363.abstract.htmlhttp://jn.physiology.org/content/106/3/1363.full.htmlhttp://jeb.biologists.org/content/214/19/3305.full.pdfhttp://jeb.biologists.org/content/214/19/3305.full.pdfhttp://jeb.biologists.org/content/214/19/3305.full.htmlhttp://jeb.biologists.org/content/214/19/3305.abstract.htmlhttp://jeb.biologists.org/content/214/19/3305.full.htmlhttp://jeb.biologists.org/content/214/19/3305.full.pdfhttp://jeb.biologists.org/content/214/19/3305.abstract.htmlhttp://jn.physiology.org/content/95/6/3426.full.htmlhttp://jn.physiology.org/content/95/6/3426.full.htmlhttp://jn.physiology.org/content/106/1/91.full.pdfhttp://jn.physiology.org/content/106/1/91.full.htmlhttp://jn.physiology.org/content/106/1/91.abstract.htmlhttp://jn.physiology.org/content/106/2/999.full.pdfhttp://jn.physiology.org/content/106/2/999.full.htmlhttp://jn.physiology.org/content/106/2/999.abstract.htmlhttp://jn.physiology.org/content/106/3/1525.full.pdfhttp://jn.physiology.org/content/106/3/1525.full.htmlhttp://jn.physiology.org/content/106/3/1525.abstract.htmlhttp://jn.physiology.org/content/106/3/1363.full.pdfhttp://jn.physiology.org/content/106/3/1363.full.htmlhttp://jn.physiology.org/content/106/3/1363.abstract.htmlhttp://jeb.biologists.org/content/214/19/3305.full.pdfhttp://jeb.biologists.org/content/214/19/3305.full.htmlhttp://jeb.biologists.org/content/214/19/3305.abstract.htmlhttp://jn.physiology.org/content/95/6/3426.full.html#ref-list-1


2/13

Motor Patterns in Human Walking and Running

G. Cappellini,1 Y. P. Ivanenko,1 R. E. Poppele,2 and F. Lacquaniti1,3,4

1Department of Neuromotor Physiology, Scientific Institute Foundation Santa Lucia, Rome, Italy; 2Department of Neuroscience,

University of Minnesota, Minneapolis, Minnesota; and 3Department of Neuroscience and 4Centre of Space Bio-medicine,

University of Rome Tor Vergata, Rome, Italy

Submitted 24 January 2006; accepted in final form 1 March 2006

Cappellini, G., Y. P. Ivanenko, R. E. Poppele, and F. Lacquaniti.

Motor patterns in human walking and running. J Neurophysiol 95:3426 3437, 2006; doi:10.1152/jn.00081.2006. Despite distinct differ-ences between walking and running, the two types of human locomo-tion are likely to be controlled by shared pattern-generating networks.However, the differences between their kinematics and kinetics implythat corresponding muscle activations may also be quite different. Weexamined the differences between walking and running by recordingkinematics and electromyographic (EMG) activity in 32 ipsilaterallimb and trunk muscles during human locomotion, and compared theeffects of speed (312 km/h) and gait. We found that the timing ofmuscle activation was accounted for by five basic temporal activationcomponents during running as we previously found for walking. Eachcomponent was loaded on similar sets of leg muscles in both gaits butgenerally on different sets of upper trunk and shoulder muscles. Themajor difference between walking and running was that one temporalcomponent, occurring during stance, was shifted to an earlier phase inthe step cycle during running. These muscle activation differencesbetween gaits did not simply depend on locomotion speed as shownby recordings during each gait over the same range of speeds (59km/h). The results are consistent with an organization of locomotionmotor programs having two parts, one that organizes muscle activa-tion during swing and another during stance and the transition toswing. The timing shift between walking and running reflects there-

fore the difference in the relative duration of the stance phase in thetwo gaits.

I N T R O D U C T I O N

Walking and running are generally considered as distinctgait modes with strikingly different mechanics and energetics.Humans change gait to increase locomotion speed while savingenergy. Thus oxygen consumption is lower for walking thanfor running below the transition speed, whereas it is higherabove this speed (Margaria 1976). In walking, the body vaultsup and over each stiff leg in an arc, analogous to an invertedpendulum (Cavagna et al. 1976) (Fig. 1A). Kinetic energy in

the first half of the stance phase is transformed into gravita-tional potential energy, which is partially recovered as the bodyfalls forward and downward in the second half of the stancephase. Running, instead, is analogous to bouncing on a pogostick (Full and Koditschek 1999; Raibert 1986) (Fig. 1A). As aleg strikes the ground, kinetic and gravitational potential en-ergy is temporarily stored as elastic strain energy in muscles,tendons, and ligaments and then is nearly all recovered duringthe propulsive second half of the stance phase. The walkinggait may also be defined by the existence of a double support

phase during stance, whereas running has a flight phaseduring which neither limb is in ground contact.

Walking and running are the two most common forms ofhuman gait. Although they share some basic kinetics andkinematics, the two gaits are also distinctly different so thetransition from walking to running is obvious. In fact, bothkinematics and kinetics change abruptly in going from awalking gait to a running gait (Hreljac 1993; Minetti et al.

1994; Nilsson et al. 1985). For example, the gait transition isaccompanied by an abrupt decrease in ground contact time by35% and an 50% increase in peak ground reaction force.There are also a number of gait parameters that change mono-tonically with increasing speed during both walking and run-ning, including increased step length and cycle duration anddecreased stance duration (Nilsson et al. 1985). Many of thesechanges are associated with increasing intensity of muscleactivation (Ivanenko et al. 2006; Prilutsky and Gregor 2001;Winter and Yack 1987).

The mechanisms that underlie the changes associated withspeed and gait changes are still not well understood. In general,the mechanics of locomotion and the associated muscle activity

have been more thoroughly studied in the human than has theneural control. For the latter, we must rely largely on extrap-olations from animal models. A number of experimental find-ings from animals and humans suggest that different locomo-tion patterns may result primarily from peripheral factors, likemuscle performance or sensory feedback (Pearson 2004; Smithet al. 1993) and/or by a reconfiguration of the neural network(Gillis and Biewener 2001; Grillner 1981) or by modulating theparameters of a basic network (Collins 2003; de Leon et al.1994; Golubitsky et al. 1999; Grillner and Zangger 1979;Orlovsky et al. 1999; Pribe et al. 1997).

An important result from the animal studies was the dem-onstration that the entire range of speeds and locomotion gaitscan be generated in the cat by varying only the intensity of

stimulation of the mesencephalic locomotor region (MLR)(Mori et al. 1989; Shik 1983; Shik et al. 1966). The gaitdetermined by this central command was automaticallyadapted to the external conditions (such as slope of the road orbody loading) by peripheral mechanisms. Increasing only thestrength of the MLR stimulation could increase the speed offorward progression, and at stronger stimulation levels, theanimal changed the locomotion gait from out-of-phase coordi-nation (walk or trot) to in-phase coordination (run or gallop).Areas anatomically similar to the MLR in the cat have also

Address for reprint requests and other correspondence: Y. P. Ivanenko,Dept. of Neuromotor Physiology, Scientific Institute Foundation Santa Lucia,306 via Ardeatina, 00179 Rome, Italy (E-mail: [email protected]).

The costs of publication of this article were defrayed in part by the paymentof page charges. The article must therefore be hereby marked advertisementin accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

J Neurophysiol 95: 34263437, 2006;doi:10.1152/jn.00081.2006.

3426 0022-3077/06 $8.00 Copyright 2006 The American Physiological Society www.jn.org


3/13

been found to serve a similar function in fish, reptiles, birds,and primates (Jordan 1991; Mori et al. 1996).

The question remains, however, as to whether this descend-ing control modulates a basic motor program for walking andrunning (for instance, a set of nonlinear oscillators with bifur-cations at critical transitional points) or a separate set ofoscillators for each distinct gait. In either case, a gait transitionmight be subject to feedback concerning critical peripheral

factors. A recent line of investigation relevant to the neuralgeneration of locomotor patterns is based on the identificationof elementary units of muscle activation (Bizzi et al. 1991,2002; Davis and Vaughan 1993; dAvella and Bizzi 2005;Giszter et al. 2001; Hart and Giszter 2004; Hultborn 2001;Ivanenko et al. 2003; Kargo and Giszter 2000; Patla et al. 1985;Ting and Macpherson 2005; Tresch and Bizzi 1999). Accord-ing to this approach, locomotor programs may be considered asa set of characteristic timings of muscle activation (Ivanenko etal. 2005). In fact, the same five basic activation componentscan account for the electromyographic (EMG) waveforms of32 ipsilateral leg, trunk, and shoulder muscles during walkingat speeds between 1 and 5 km/h (Ivanenko et el., 2004b). Thesame five components were also present when walking was

combined with various voluntary motor tasks that requiredsignificantly different muscle usage (Ivanenko et al., 2005). Ifthere is a common set of oscillators for walking and running,we might anticipate that muscle activation during runningmight also be accounted for, at least in part, by the same basicmuscle activation components. If, however, separate sets ofoscillators underlie the different gaits, we might expect that

this would be associated with distinctly different timings ofmuscle activation. Although the analysis of individual EMGpatterns of human running has been performed in numerousstudies (see Prilutsky and Gregor 2001), the underlying com-mon structure of the motor output has not been previouslyinvestigated.

Therefore the aim of this study was to examine how muscleactivation depends on locomotion speed and on locomotiongait. The experimental design was to record kinematics andEMG activity from human subjects as they either walked or ranon a treadmill at each of several different speeds. We usedstatistical methods incorporating a linear decomposition of theEMG data to determine the general design of the motor output

in walking and running.

M E T H O D S

Subjects

Eight healthy subjects [6 males and 2 females, between 26 and 44yr of age, 69 10 kg (mean SD), 1.75 0.07 m] volunteered forthe experiments. All subjects were right-leg dominant. The studiesconformed to the Declaration of Helsinki, and informed consent wasobtained from all participants according to the procedures of theEthics Committee of the Santa Lucia Institute.

Experimental setup

Subjects walked or ran on a treadmill (EN-MILL 3446.527, BonteZwolle BV, Netherlands) at different controlled speeds (312 km/h).They were asked to swing their arms normally and to look straightahead. Subjects walked with their shoes on. Before the recordingsession, subjects practiced for a few minutes in walking and runningon the treadmill at different speeds. In a standard protocol, subjectswere asked to walk at 3, 5, 7, and 9 km/h and to run at 5, 7, 9, and 12km/h so that we could compare walking and running at the samespeeds (5, 7, and 9 km/h).

We also used a computer-controlled speed program that linearlyincreased and then decreased treadmill speed between 1 and 12 km/hto observe the recorded parameters during continuous speed changes(ramp speed condition, acceleration, and deceleration was set to 0.4km h1 s1). Three subjects participated in this experiment. Theywere instructed to follow the changes in speed by remaining in place

with respect to the treadmill when the belt velocity was changing. Wealso recorded during overground walking [at 5.6 1.1 (SD) km/h]and running (at 9.6 0.9 km/h). This allowed us to calculate themoments of forces (see following text) by asking subjects to step ona force plate located in the middle of a 7-m walkway.

Data recording

We recorded kinematic data bilaterally at 100 Hz by means of theVicon-612 system (Oxford, UK) with nine TV cameras spaced aroundthe walkway. Infrared reflective markers (diameter: 1.4 cm) wereattached on each side of the subject to the skin overlying the followinglandmarks: gleno-humeral joint (GH), the midpoint between theanterior and the posterior superior iliac spine (ilium, IL), greatertrochanter (GT), lateral femur epicondyle (LE), lateral malleolus

FIG. 1. General characteristics of walking and running. A: schematic rep-resentation of walking by vaulting (inverted pendulum) and running by abouncing gait (leg spring-loaded behavior during stance). B: ankle, knee, andhip moments of force and vertical ground reaction force (normalized to thesubjects weight) of the right leg during overground walking (5.4 km/h) andrunning (9.4 km/h) in 1 representative subject. C: relative stance duration and

cycle duration (

SD) in walking and running. D: foot (fifth metatarso-phalangeal joint, VM) trajectory characteristics (mean SD, normalized to thelimb length L): horizontal excursion of the VM marker [relative to greatertrochanter (GT)] and vertical VM displacements (in the laboratory referenceframe).

3427MOTOR PATTERNS IN HUMAN WALKING AND RUNNING

J Neurophysiol VOL 95 JUNE 2006 www.jn.org


4/13

(LM), heel (HE), and fifth metatarso-phalangeal joint (VM). Thespatial accuracy of the system is better than 1 mm (root mean square).

We recorded electromyographic activity by means of surface elec-trodes from 32 muscles simultaneously on the right side of the body.We used active Delsys electrodes (model DE2.1, Delsys, Boston,MA) applied to lightly abraded skin over the respective muscle belly.The signals were amplified (10,000), filtered (20450 Hz; Bagnoli16, Delsys), and sampled at 1,000 Hz. Sampling of kinematic, forceplatform, and EMG data were synchronized.

The following 32 muscles were recorded from all eight subjects:tibialis anterior (TA), flexor digitorum brevis (FDB), gastrocnemiuslateralis (LG), gastrocnemius medialis (MG), soleus (SOL), peroneuslongus (PERL), vastus lateralis (Vlat), vastus medialis (Vmed), rectusfemoris (RF), sartorius (SART), biceps femoris (long head, BF),semitendinosus (ST), adductor longus (ADD), tensor fascia latae(TFL), gluteus maximus (GM), gluteus medius (Gmed), externaloblique (OE), internal oblique (OI), latissimus dorsi (LD), iliopsoas(ILIO), rectus abdominis, superior portion (RAS), erector spinaerecorded at T1, T9, and L2 (EST1, EST9, ESL2, respectively), bicepsbrachii (BIC), triceps brachii (TRIC), deltoideus, anterior and poste-rior portions (DELTA and DELTP, respectively), trapezius, inferiorand superior portions (TRAPS and TRAPI, respectively), sternocleido

mastoideus (STER), and splenius (SPLE). Electrode placement for theerector spinae muscle was 2 cm lateral to the spinous process, forSOLabout 2 cm distal to the medial head of the gastrocnemius, forRASabout 3 cm lateral of the umbilicus (also see Winter 1991).Before the electrodes were placed, the subject was instructed abouthow to selectively activate each muscle (Kendall et al. 1993), whileEMG signals were monitored, so as to optimize the EMG signal andminimize cross-talk from adjacent muscles during isometric contrac-tions.

During overground locomotion, we also recorded the ground reac-tion forces (Fx, Fy, and Fz) under the right foot at 1,000 Hz by a forceplatform (0.9 0.6 m, Kistler 9287B, Zurich, Switzerland). At theend of the recording session, we made anthropometric measurementson each subject. These included the mass and stature of the subject,the length and circumference of the main segments of the body

(Zatsiorsky et al. 1990).

Data analysis

BIOMECHANICAL ANALYSIS. The gait cycle was defined with re-spect to the right leg movement, beginning with right foot contact withthe surface (touch-down). The body was modeled as an interconnectedchain of rigid segments: IL-GT for the pelvis, GT-LE for the thigh,LE-LM for the shank, and LM-VM for the foot. For walking andrunning, the gait cycle was defined as the time between two successivefoot contacts of the right leg corresponding to the local minima of theHE marker. The timing of the lift-off was determined analogously(when the VM marker elevated by 3 cm). The touch-down and lift-offevents were also verified from the force plate recordings (when thevertical ground reaction force exceeded 7% of the body weight), andwe found that the kinematic criteria we used predicted the onset andend of stance phase with an error 2% of the gait cycle duration (seealso Borghese et al. 1996). The data were time-interpolated overindividual gait cycles on a time base with 200 points.

EMG-ANALYSIS. Raw data were numerically rectified, low-pass fil-tered with a zero-lag Butterworth filter with cutoff at 10 Hz, time-interpolated over a time base with 200 points for individual gait cyclesand averaged. Each trial for a given speed included 10 consecutivegait cycles (typically 15).

Joint moments of force

For the overground locomotion records, the moments of forces(with extensor moments being positive) at the ankle, knee, and hip

joints of the right leg were calculated using measured kinematics,force plate data, anthropometric data taken on each subject, and thetraditional Newton-Euler inverse dynamics model (Bresler andFrankel 1950). The moments were normalized to the body mass.

Factor analysis

We applied a principal component analysis (PCA) to each ofseveral data sets consisting of normalized EMG patterns over a stepcycle. Factor analysis (FA) used here has been thoroughly describedin our previous papers (Ivanenko et al. 2003, 2004b, 2005). Briefly,the steps involve calculation of the correlation matrix, extraction ofthe initial principal components (PCs), application of the varimaxrotation, calculation of factors scores (referred to as temporal activa-tion components in the text), factor loadings (weighting coefficientsacross muscles), and percent of variance accounted for (PV) by eachtemporal component in the total data set. The aim of FA is to representthe original EMG data set E (m t matrix, where m: the number ofmuscles and t 200 for all conditions because EMG data weretime-interpolated over individual gait cycles to fit a normalized200-point time base) as a linear combination of n basic temporalcomponents (n m), E W C residual, where W are weighting

coefficients or loadings (m

n matrix) and C are basic temporalcomponents (n t matrix).Some of the deeper muscles around the hip have a rather complex

architecture depending on the hip joint angle (Delp et al. 1999), andindeed we cannot rule out a possibility that there might be somecross-talk in our recordings of muscles like ILIO. Nevertheless, ouraveraged records (the major peak of activity around lift-off) areconsistent with those reported in the literature (Andersson et al. 1997;Rab 1994) and obtained using a simulated biomechanical model ofbipedal stepping (Zajac et al. 2003). A FA could theoretically becompromised by electrical cross-talk among adjacent muscle record-ings (De Luca and Merletti 1988; Nene et al. 2004). However, ifcross-talk did exist, it would most likely have affected only theweighting coefficients assigned to each component in accounting forthe activity of a muscle that was contaminated by cross-talk (Ivanenko

et al. 2004b).In the present study, unless a component explained at least as muchas 3% of the total variance, we dropped it. The higher-order compo-nents were generally variable and not significant (see RESULTS). Toassess similarities of activation components across subjects andspeeds, we used the following parameters: timing of the main peakand the width of the main peak (Ivanenko et al. 2005). The five basiccomponents were ordered according to the timing of the main peak.The width of the main peak was estimated by measuring the full-widthat half-maximum (FWHM). The half of maximum was calculated asa point corresponding to the mid-height between the peak value andthe mean value between 2 minima (1 to the left, another 1 to the rightwith respect to the main peak). In the case of boundary peaks (closeto the foot contact), we used only descending or ascending part of thecomponent profile to calculate the half-width at half-maximum and

then multiplied it by 2, under the assumption of the symmetricalprofile.

Independent component analysis and nonnegativematrix factorization

Different statistical methods have been developed to assess a lineardecomposition of the original set of data based on different assump-tions (Tresch et al. 2006). Thus factor analysis with varimax rotationconstrains the analysis to orthogonal (uncorrelated) factors. Althoughfactor analysis of EMG activity was thoroughly documented forhuman locomotion and we could thus compare our results with thosereported in the literature (Davis and Vaughan 1993; Ivanenko et al.2003, 2004b, 2005; Merkle et al. 1998; Olree and Vaughan 1995), wealso performed two other decomposition-related statistical analyses:

3428 G. CAPPELLINI, Y. P. IVANENKO, R. E. POPPELE, AND F. LACQUANITI



5/13

independent component analysis (ICA) and nonnegative matrix fac-torization (NMF).

We performed an informax ICA (Bell and Sejnowski 1995) usingthe function runica in the EEGLAB package (v4.5; www.sccn.ucsd.edu/eeglab/) (Delorme and Makeig 2004) running on Matlab v7. Therationale for this analysis is that unlike PCA (FA), ICA aims atextracting unknown hidden components from multivariate data usingonly the assumption that the unknown components are mutuallyindependent (Hart and Giszter 2004).

We also applied a NMF using the matrix multiplication algorithmdescribed by Lee and Seung (1999) that constrains the temporalcomponents and weighting coefficients to be nonnegative. The ratio-nale for application of the NMF was that our data consisted ofnonnegative values (rectified EMG activity), and the NMF constraintallows only additive not subtractive combinations (dAvella and Bizzi2005).

All methods were applied to the same data set in each task and theresults across methods and tasks (walking vs. running) were com-pared. We reduced the dimensionality of the data to the same finalnumber of temporal components as in the FA for both the ICA andNMF analyses. Using the correlation matrix for factor analysis (seepreceding text) automatically implies normalization of each EMG

waveform to its SD value. To compare the results of FA with those ofICA and NMF, the amplitude of EMG waveforms was normalized bytheir mean SD values prior to the application of ICA and NMF. Thisis reasonable because we were mainly interested in rhythmic pattern-ing elements in the control of locomotion (temporal components)rather than in muscle synergies (weighting coefficients) and becausethe amplitude of EMG activity was roughly similar when walking andrunning at the same speed (see RESULTS).

Statistics

A factor analysis was performed using Statistica v6.0 (StatSoft).Statistical analyses (Students t-tests) were used to compare thehalf-widths of the component peaks (FWHM) in walking and running.Statistics on correlation coefficients was performed on the normallydistributed, Z-transformed values.

R E S U L T S

General characteristics of walking and running

Running and walking gaits are usually adopted for differentspeeds of locomotion, with a preferred transition occurring at7 km/h for most human subjects (Nilsson et al. 1985). Somemajor differences between the two gaits are illustrated in Fig.1A. During walking, the leg tends to behave like a rigid strut,and the joints remain relatively extended throughout the stancephase (Lee and Farley 1998). In contrast, during running, the

major leg joints undergo substantial flexion and extensionduring stance as the leg behaves in a more spring-like manner.Both running and walking can occur, however, over a wide

range of speeds (Minetti et al. 1994). We examined both gaitsat 5, 7, and 9 km/h and walking alone at 3 km/h and runningalone at 12 km/h. Although increases in locomotion speed inboth gaits were accomplished by increasing both the cadenceand stride length, the duration of the stance phase was always50% of the gait cycle for walking and 50% for running(Fig. 1C). This difference corresponds to a change from adouble support phase during walking to a single support duringrunning, which is then accompanied by a greater groundreaction force occurring shortly after foot contact at the begin-ning of the cycle (Fig. 1B).

Figure 1D shows the normalized values (over all trials andsubjects) of horizontal (VMx) and vertical (VMy) excursion ofthe foot plotted as function of speed. These global gait param-eters exhibited the well-known monotonic relationship withincreasing speed and different behaviors between walking andrunning (Nilsson et al. 1985).

Muscle activity patterns

The overall effect of speed on the average EMG activity wasconsistent with that reported in the literature (den Otter et al.2004; Hogue 1969; Ivanenko et al. 2002; Murray et al. 1984;Nilsson et al. 1985), and it may be seen in Fig. 2. The intensityof distal leg activity is more similar across speeds, whereas theintensity of proximal leg and trunk activity increases muchmore with speed. Except for a few muscles like Bic, there is noobvious distinction in the intensity of muscle activation at thetransition from walking to running (dotted vertical lines in Fig.2A). However, there are differences in the patterns of activationat the transition, particularly in the distal leg muscles. The peak

of LG activation shifts to an earlier phase in the cycle. Insteadthe changes in Vlat and ST activation are less obvious (Fig.2B). The single cycle records at the gait transition points showthat the difference in the activation timing of the calf musclesbetween walking and running occurs abruptly within a singlestep cycle (Fig. 2B). Some high-intensity muscle activationprior to the walk-to-run transition (see LG activity) is consis-tent with the idea that the activation of major leg muscles mightact as a trigger for gait transition and explain why locomotionat nonpreferred speeds requires excessive muscle activation(Prilutsky and Gregor 2001).

The average EMGs for each muscle we recorded are illus-trated in Fig. 3. They show the same basic result illustrated byFig. 2 but with more detail. During walking, distal leg activity

had about the same modulation intensity and overall pattern atall speeds although mean activity increased with speed. Prox-imal leg activity became more robust at the higher speeds inwalking, and individual patterns of activity were often differentat different speeds (see also Ivanenko et al. 2002; Winter andYack 1987). The patterns of activity in trunk and abdominalmuscles also tended to change dramatically as a function ofspeed. Cervical muscle activity, which was quite low at thelowest speeds, became significant at the higher speeds. There isan increase in the EMG activity of all muscles during running.In general muscles are most active around the time of footcontact.

The posterior calf muscles tend to function as one unit.

During walking, the anterior muscles (TA) become active priorto lift-off and remain active throughout the swing phase andinto the first 10% of the next cycle. The posterior calf muscles(MG, LG, SOL, PERL) are all active at 40% of the cyclewhen the anterior muscles are relatively silent. With running,this pattern reverses so that the anterior muscles increase theiractivity during the mid part of the cycle, and the posteriormuscles are active instead around the time of foot contact.This, in fact, is the most marked change in muscle coordinationwe observed in the transition between walking and running.These changes are associated with changes in leg and footmovements. In walking, the foot strikes the ground with therear part of the heel and with a marked plantarflexion in theankle. In running, on the other hand, initial contact is generally




6/13

made with a more anterior part of the foot. Individual differ-ences exist in the way the foot is placed on the ground (Chanand Rudins 1994; Rodgers 1988), which must also lead todifferences in the activity pattern of muscles controllingthe ankle.

EMG components: comparison of analytical methods

In our previous study (Ivanenko et al. 2005), we comparedthree different forms of factor analysis to find common com-ponents in the EMG patterns across muscles. Because each ofthese statistical approaches (FA, ICA, and NMF, see METHODS)places different restrictions on the outcomes, they might alsoextract different sets of activation components. We found,however, that the different algorithms converged to a similarsolution about the temporal structure of the EMG activationpatterns, both during normal walking and during walkingcombined with a voluntary movement.

We repeated this comparison here to determine whether thesame components were also identified for EMG activity re-corded during running. As before, we ordered the ICA and

NMF components so that their waveforms matched the corre-sponding FA components (Fig. 4). The results were similar tothose we reported previously (Ivanenko et al. 2005; see alsoTresch et al. 2006) in that the three methods gave essentiallythe same result for both walking at 5 km/h (average waveformcorrelation with FA component r 0.92 0.05 for ICA and0.89 0.09 for NMF) and for running at 12 km/h ( r 0.94

0.03 for ICA and 0.87

0.11 for NMF). In addition, theloadings or weighting coefficients on particular muscles werealso similar in the three methods. Finally, the total varianceexplained by basic temporal components was also similar.For instance, in Fig. 4, the percent of variance explained byall components was 87, 90, and 86% during walking and90, 91, and 89% during running using FA, ICA and NMF,respectively.

Factor analysis

Given the nearly equivalent results of the previous section,we again used the same FA approach as we did previously(Ivanenko et al. 2003, 2004b, 2005) to describe common

FIG. 2. Slow changes in treadmill belt speed (ramp speed condition) in 1 representative subject. A: electromyographic (EMG) activity, instantaneous treadmillbelt speed, limb angle, and gait cycle duration. Limb angle was defined as GT-LM elevation angle [the angle between the greater trochanter (GT)-lateral

malleoulus (LM) segment projected on the sagittal plane and the vertical]. B: EMG activity of selected muscles [gastrocnemius lateralis (LG), tibialis anterior(TA), semitendinosus (ST), and vastus lateralis (Vlat)] computed for single cycles before and after the transition cycle. Vertical dashed lines denote the transitionfrom walking and running (defined by the presence/absence of the double support phase). The transition typically occurred at the higher speed during gaitacceleration as opposed to gait deceleration, consistent with previous observations (Thorstensson and Roberthson 1987). Note that except for a few muscles likeBic (associated with an abrupt change in the forearm elevation during running), there is no obvious distinction in the intensity of muscle activation at the transitionfrom walking to running. However, there are differences at the transition in the pattern of muscle activation.




7/13

components in the EMG patterns across muscles. We analyzedEMG patterns separately for each subject and for each speed(Fig. 5). The basic result was that five temporal componentsaccounted for between 83 and 99% of the total EMG waveformvariance for all subjects across all conditions. The most sig-nificant component of order higher than five explained only3.1 1.1% of the total variance and was highly variable acrosssubjects and conditions (component 6, Fig. 5).

The five components determined from the data recordedduring walking were essentially the same as those we described

previously (Ivanenko et al. 2004b), but they extend the range ofthose results to now include walking at 7 and 9 km/h. Fivetemporal components were also consistently found for allspeeds during running (Fig. 5B). They each corresponded to acomponent derived from the walking data with the single clearexception of component 2. This component was substantiallyphase-shifted from a peak at45% of the cycle during walkingto between 20 and 30% of the cycle during running.

As before, we found small but systematic shifts in the timingof the main peaks of the components to an earlier phase with

FIG. 3. Ensemble averaged activity patterns recorded from 8 subjects for 32 muscles. EMG records were filtered with a low-pass cut-off of 10 Hz. Top: stickdiagrams. Data were plotted vs. a normalized cycle. As the relative duration of stance varied across subjects, a hatched region indicates an amount of variabilityin the stance phase duration across subjects.




8/13

increasing speed (Fig. 6A). We showed earlier that these shiftstended to scale with the duration of the stance phase as thatchanged with speed (Ivanenko et al. 2004b). At the higherspeeds used here (3 km/h), the duration of stance did notdecrease as much with speed as it did at lower walking speeds(13 km/h), so the component phase shifts were correspond-ingly less. There were also comparable phase shifts with speedduring running. In fact, the phases of component 1, 3, 4, and 5were essentially the same as we found at comparable speeds forwalking (Fig. 6A). This might be slightly less clear in the caseof component 5 because there may be a difference in therunning and walking phases at 9 km/h.

The phase of component 2 seems to behave differentlyduring walking and running. During walking the shift withspeed is comparable to what we observed with the other fourcomponents. During running, however, not only is there adiscrete phase shift with respect to the corresponding walkingcomponent, but the shift with speed seems to be greater.However, when we consider the phases relative to lift-off (endof stance) instead of relative to heel strike or touch-down, thephase of component 2 is consistently 80% of the cycle (20%prior to lift-off) during both walking and running (Fig. 6B). Incontrast, with lift-off as the reference, the phases of the otherfour components are no longer comparable during walking andrunning. In fact the phase of component 4 during walking

aligns with component 3 during running and component 1during walking aligns with the phase of component 5 duringrunning. Moreover components 5 and 3 during walking andcomponents 1 and 4 during running were not aligned withcomparable components. Thus all but component 2 may bebetter represented by the reference to touch-down as we havealso used in our previous studies.

The half-widths of the component peaks (FWHM) wereessentially the same (13% of the cycle duration) for eachcomponent across speeds (Fig. 6C). There was some tendencythough for the running components 1 and 2 to be slightly wider(1517%) at the lower speeds relative to the walking compo-nents (P 0.06 for 5 km/h and P 0.05 for 7 km/h, pairedt-tests).

The total amount of waveform variance accounted for (PV)by each component was quite similar for walking and running,with components 4 and 5 typically accounting for less variancethan the other three during both walking and running (Fig. 6D).Moreover, some differences that depended on walking speed,such as an increase in PV by component 3 and decrease bycomponent 4 with increasing speed (Ivanenko et al. 2004b)seemed to hold also during running.

The relative strength of the effect of each activation com-ponent on a given EMG pattern is given by the componentloading or weighting coefficient (Fig. 7B). We estimatedany changes in weighting coefficients of the five temporal

components between walking and running at the same speed (7km/h) by correlating the set of weighting coefficients forvarious groups of muscles. We made separate comparisons forthe leg and lower trunk muscles and the upper trunk andshoulder muscles for each corresponding component in run-ning and walking (Fig. 7B). There was a fairly high consistencyof relatively high correlation coefficients (rLEG 0.68 0.17)indicating that similar muscle groups in the limb and lowertrunk were activated with common timing during walking andrunning. Interestingly the best correspondence was for compo-nent 2, indicating that the loading of component 2 on legmuscles was basically the same in walking and running in spiteof the shift in the timing of component 2.

The consistency was much less, however, for the upper trunkand shoulder muscles (rTRUNK 0.28 0.39), where thebiggest differences were seen with components 1 and 4. Thesecomponents therefore participate in the activation of differentgroups of muscles in walking and running. It is obvious in factthe upper body and arm motion in particular tend to be muchdifferent in walking and running.

D I S C U S S I O N

We examined muscle activation in human subjects over arange of speeds during both walking and running. Consistentwith our previous result on the effects of walking speed(Ivanenko et al. 2004b), we found that speed changes during

FIG. 4. Comparison of activation components during walking and running at the same speed (7 km/h), derived from averaged muscle activity patterns (seeFig. 3) using different statistical methods [factor analysis (FA), independent component analysis (ICA) and nonnegative matrix factorization (NMF)]. Theamplitude of EMG waveforms was normalized by the mean SD before application of ICA and NMF. Correlations between components obtained from ICA and

NMF and corresponding varimax factors (FA) are indicated.




9/13

both walking and running are associated with increases in the

intensity of muscle activation and only minor changes in theirrelative timings. The same five activation components ac-counted for muscle activity during either walking or runningover the entire speed range tested. Moreover, in spite ofsignificant differences in the overall characteristics of the twogaits, we found that major differences in muscle activationtiming occurred only during stance, while walking and runningtiming was basically the same during the swing phase.

Speed versus gait effects

Changes in locomotion speed are accompanied by changesin cycle length and cycle duration as well as in the fraction ofthe cycle devoted to stance or swing. Although such changes

are also accompanied by changes in muscle activation patterns,

most of these scale with the cycle duration so that changes inactivation intensity seem to dominate (Andersson et al. 1997;den Otter et al. 2004Ivanenko et al. 2004b; Murray et al. 1984;Prilutsky and Gregor 2001; Winter and Jack 1987). Gaitchanges on the other hand, such as the transition from walkingto running, are often considered to involve a different muscleusage and not just more intense muscle activation. However,this is not what we found. The major change in muscleactivation during running compared with walking was a phaseshift attributable to a single activation component duringstance, whereas the muscles activated by the shifted compo-nent remained the same. Increasing running speeds, from 7 to12 km/h, was mostly accompanied by a similar activation of

FIG. 5. Temporal activation components by subject. A: varimax factors derived from muscle activity patterns by factor analysis for each of 8 subjects (thinlines). Bold traces show the average of the components across 8 subjects. Five common components across conditions are designated in a chronological order(with respect to the timing of the main peak). The next higher-order component (that explained on average only 3% of variance) was also plotted. It displayedvery high variability across subjects and conditions and, therefore, was dropped from the analysis. Stance phase was somewhat different across subjects asindicated by hatched area of the stance-swing bar.




10/13

leg muscles and somewhat more intense activation of trunk and

shoulder muscles.Muscle activation patterns recorded at 3 km/h were basically

the same as those recorded at higher walking speeds with onlyfew exceptions (e.g., TFL, ADD, Fig. 3). Instead, most of the32 muscles we recorded exhibited more intense activation athigher speeds, and the activation timing was fully accountedfor by five activation components (Ivanenko et al. 2004b). Weexpected though that running, which places different biome-chanical demands on the neuromuscular system, would exhibitdifferent muscle activation patterns. However, many muscleactivation patterns were quite similar to those observed duringwalking. In fact, the five components that represent thosepatterns were each associated with the activation timing of

nearly the same set of leg and lower trunk muscles duringrunning and walking. The average correlation across weightingcoefficients was 0.68 (range: 0.380.82). Thus these compo-nents seem to play essentially the same role in both conditionsindependently of speed [however, see Ivanenko et al. (2004b)and Fig. 6B, which shows a decrease in PV for component 4and an increase for component 3 with increases in speed].

Gait changes

The activation of ankle extensors at an earlier phase of thelocomotion cycle was the major EMG change we found to beassociated with running compared with walking, and it reflectsa difference in the support function of the leg in the two gaits.

One hypothesis proposed to account for the preferred transitionto running with increasing speed predicts this difference inactivation timing. It notes that at higher walking speeds, theankle extensors are loaded toward the end of stance in anunfavorable portion of their force-velocity curve, so their forceproduction is limited even as levels of activation increase.Shifting the activation to an earlier phase of stance shifts the

activation to a more favorable force production range (Neptuneand Sasaki 2005). This can also be seen as a shift in the groundreaction force (Fz) to near the beginning of the cycle (Fig. 1B).

This shift of muscle activation timing from walking torunning and the greater peak activation intensity during walk-ing are both clearly evident in the EMG recordings from theposterior calf muscles (Fig. 3, MG, LG, PERL, and SOL). The

FIG. 6. Characteristics of activation temporal components. A: the timing(SD) of the peaks of 5 common FA components across speeds and tasks. The5 basic components are each characterized by a relatively narrow peak ofactivation at a particular phase of the cycle. Zero timing refers to the heelstrike. Note the abrupt change in the timing of component 2 between walkingand running. B: the timing of the same peaks referenced to the lift-off event.Gait cycle and 0 timing refers to the lift-off in this plot. Note the invariance ofthe timing of component 2 in walking and running when referenced to thelift-off event. C: the width (SD) of the main peaks of components estimatedas the full-width at half-maximum (FWHM). D: percent of variance explainedby each component in the data sets obtained from all individual muscles.

FIG. 7. Activation temporal components derived from averaged muscleactivity patterns (see Fig. 3). A: superimposed activation components inwalking and running. Note a prominent phase shift of the timing of the mainpeak of the second component in running. B: weighting coefficients (across 32muscles) of the 5 basic temporal components during walking and running atthe same speed (7 km/h). The correlation of weighting patterns betweenwalking and running is shown at the right of each record both for the leg/lowertrunk [tibialis anterior (TA) rectus abdominis, superior portion (RAS)] andupper trunk/shoulder [latissimus dorsi (LD) sternocleido mastoideus(STER)] muscles separately.




11/13

timing shift is reflected in the activation components as a shiftin the timing of component 2, which is weighted primarily onthose same muscles. That is, component 2 accounts for theactivation timing of basically the same muscles in both gaits.The preceding explanation about an earlier phase of ankleextensor activation implied a control based on optimal muscleperformance using feedback information about muscle force

production, length, and velocity. However, muscle perfor-mance may not contribute directly to the control of timing. Thesubjects in our study were able to use either gait voluntary at5, 7, and 9 km/h, and their muscle activation exhibited a similardiscrete shift in timing between walking and running at eachspeed. It is also evident however, that component 2 timingshifted to progressively earlier phases in the cycle as runningspeed increased, indicating a possible role for muscle perfor-mance, as suggested by this hypothesis.

Other proposals to explain the transition to running alsoinvolve parameters that are expected to change continuouslywith speed such as the pendulum-related limitations for thefrequency of leg oscillations, metabolic energy expenditure

(Margaria 1976), or an increased sense of effort due to theexaggerated swing-related activation of ankle knee and hipflexors (Prilutsky and Gregor 2001) or have some critical valuelike a critical velocity of ankle flexion (Hreljiac 1995), criticalforces (Farley and Taylor 1991; Raynor et al. 2002), or acritical angle between the thighs (Minetti et al. 1994; see alsoBiewener et al. 2004; Brisswalter and Mottet 1996; Cavagna etal. 1976; Kram et al. 1997; Neptune and Sasaki 2005; Nilssonet al. 1985; Thorstensson and Roberthson 1987). Althoughsuch speed-related parameters may account for a bifurcation inbehavior, they do not easily account for a discrete change,which can occur over a large range of speeds.

Our analysis suggests another possibility, namely that themotor program activates ankle extensors at a fixed interval

preceding lift-off. We found that this timing, corresponding tocomponent 2 activation at 20% of the cycle prior to lift-off,was basically the same over the entire speed range for bothwalking and running (Fig. 6B). Thus the propulsive forcegenerated as a result of this activation timing occurred at theend of stance for both gaits. Although this timing may producea more efficient force production during running due to thebiomechanical factors discussed in the preceding text, it is alsoa simple strategy requiring only that the timing of one com-ponent be coordinated with lift-off, which may in turn simplydepend on sensory information about the transition from stanceto swing. The effect, however, is to coordinate the swing andstance phases in the two gaits.

According to this hypothesis, the timing of components 1, 3,4, and 5 are referenced to the beginning of the cycle at heelstrike or touch-down, whereas component 2 occurs at a phasethat is appropriate for launching the swing phase. This impliesthat the invariant relative timing of the five muscle activationcomponents we have noted for walking does not represent abasic unit for all locomotion. As we found for running, therelative timing of component 2 timing can be disassociatedfrom the timing of the four components because their phasesare referenced to different events in the locomotion cycle. Therelative timing of other components might also be gait depen-dent in some way so that other forms of locomotion gait (e.g.,hopping, skipping) could also have different relative compo-nent timings that are invariant only for a particular gait pattern.

This might be expected if component timing (or CPG phase)can be directly controlled by sensory input or descendingcommands (Pearson 2004).

This may also relate to experimental work in frogs suggest-ing that a small collection of unit burst generators of synergis-tic muscle activity can be triggered by sensory input to producecorrections and other adjustments of spinally organized trajec-tories (Hart and Giszter 2004; Kargo and Giszter 2000). Inhumans, several researchers have provided support for a seg-mental movement generation and superposition of short burstsor fixed timing elements to control upper limb point to pointand cyclic motions (e.g., Krebs et al. 1999). This is not unlikeour previous observation about the coordination of voluntarymotor acts with walking (Ivanenko et al. 2005). In that case,the five activation components associated with walking werealso present in the combined motor behavior, and a newcomponent with task-related timing was superimposed. Thus tocoordinate the stance and swing phases in both walkingand running, four of the activation components remain invari-antly timed with respect to the beginning of the stance phase.

The other component (2) is inserted to correspond with thephase of the lift-off and therefore shifts its phase duringrunning (Fig. 8).

This interpretation implies that the same basic motor pro-gram consisting of five distinct muscle activation phases char-acterizes both walking and running at various speeds. It also

FIG. 8. Hypothetical motor programs for walking (A) and running (B) interms of the characteristic timing of muscle activations. In both programs, 4activation components (1, 3, 4, and 5) occur relative to the touch-down (TD)phase of the cycle and component 2 occurs relative to the lift-off (LO) phase,at about 20% of the cycle preceding LO. As 1 gait mode is switched to theother, the timing of LO relative to TD changes along with the relative timingof component 2 with respect to the other 4 components.




12/13

implies that peripheral factors like the transition to the swingphase may have essential modulatory roles affecting the rela-tive timing of activation. It is not clear, however, to what extentthese influences might contribute to the basic program toprovide an appropriate and invariant kinematic pattern (Lac-quaniti et al. 1999).

Conclusions

In summary, gait changes and speed changes each havespecific effects on muscle activation during locomotion. Nev-ertheless, the diverse biomechanical demands of running andwalking and of locomoting at different speeds are met bysurprisingly similar general template of muscle activation. Thewalking and running motor program in humans may be con-sidered as a sequence of five temporal activation components(Fig. 8). One gait mode is switched to a distinctly different gaitmode by an interplay between the timings of activation com-ponents relative to the stance and swing phase of the locomo-tion.

G R A N T S

This work was supported by the Italian Ministry of Health, Italian Ministryof University and Research, and Italian Space Agency.

R E F E R E N C E S

Andersson EA, Nilsson J, and Thorstensson A. Intramuscular EMG from thehip flexor muscles during human locomotion. Acta Physiol Scand 161:361370, 1997.

dAvella A and Bizzi E. Shared and specific muscle synergies in natural motorbehaviors. Proc Natl Acad Sci USA 102: 30763081, 2005.

Bell AJ and Sejnowski TJ. An information-maximization approach to blindseparation and blind deconvolution. Neural Comput 7: 11291159, 1995.

Biewener AA, Farley CT, Roberts TJ, and Temaner M. Muscle mechanicaladvantage of human walking and running: implications for energy cost.

J Appl Physiol 97: 22662274, 2004.Bizzi E, DAvella A, Saltiel P, and Tresch M. Modular organization of spinal

motor systems. Neuroscientist8: 437442, 2002.Bizzi E, Mussa-Ivaldi FA, and Giszter S. Computations underlying the

execution of movement: a biological perspective. Science 19; 253: 287291,1991.

Borghese NA, Bianchi L, and Lacquaniti F. Kinematic determinants ofhuman locomotion. J Physiol 494: 863879, 1996.

Bresler E and Frankel JB. The forces and moments in the leg during levelwalking. J Appl Mech 72: 2736, 1950.

Brisswalter J and Mottet D. Energy cost and stride duration variability atpreferred transition gait speed between walking and running. Can J ApplPhysiol 21: 471480, 1996.

Cavagna GA, Thys H, and Zamboni A. The sources of external work in levelwalking and running. J Physiol 262: 639657, 1976.

Chan CW and Rudins A. Foot biomechanics during walking and running.

Mayo Clin Proc 69: 448461, 1994.Collins JJ. Gait transitions. In: The Handbook of Brain Theory and NeuralNetworks (2nd ed.), edited by Arbib MA. Cambridge: The MIT Press, 2003,p. 463466.

Davis BL and Vaughan CL. Phasic behavior of EMG signals during gait: useof multivariate statistics. J EMG Kinesiol 3: 5160, 1993.

de Leon R, Hodgson JA, Roy RR, and Edgerton VR. Extensor- andflexor-like modulation within motor pools of the rat hindlimb duringtreadmill locomotion and swimming. Brain Res 654: 241250, 1994.

Delorme A and Makeig S. EEGLAB: an open source toolbox for analysis ofsingle-trial EEG dynamics including independent component analysis.

J Neurosci Methods 134: 921, 2004.Delp SL, Hess WE, Hungerford DS, and Jones LC. Variation of rotation

moment arms with hip flexion. J Biomech 32: 493501, 1999.De Luca CJ and Merletti R. Surface myoelectric signal cross-talk among

muscles of the leg. Electroencephalogr Clin Neurophysiol 69: 568575,1988.

den Otter AR, Geurts AC, Mulder T, and Duysens J. Speed related changes

in muscle activity from normal to very slow walking speeds. Gait Posture

19: 270 278, 2004.Farley CT and Taylor CR. A mechanical trigger for the trot-gallop transition

in horses. Science 253: 306308, 1991.Full RJ and Koditschek DE. Templates and anchors: neuromechanical

hypotheses of legged locomotion on land. J Exp Biol 202: 332532, 1999.Gillis GB and Biewener AA. Hindlimb muscle function in relation to speed

and gait: in vivo patterns of strain and activation in a hip and knee extensorof the rat (Rattus norvegicus). J Exp Biol 204: 27172731, 2001.

Giszter SF, Moxon KA, Rybak IA, and Chapin JK. Neurobiological andneurorobotic approaches to control architectures for a humanoid motorsystem. Robot Auton Syst 37: 219235, 2001.

Golubitsky M, Stewart I, Buono PL, and Collins JJ. Symmetry in locomo-tor central pattern generators and animal gaits. Nature 401: 693695, 1999.

Grillner S. Control of locomotion in bipeds, tetrapods, and fish. In: Handbookof Physiology. The Nervous System. Motor Control. Bethesda, MD: Am.Physiol. Soc., 1981, sect. 1, vol. 2, part 1, p. 11791236.

Grillner S and Zangger P. On the central generation of locomotion in the lowspinal cat. Exp Brain Res 34: 241261, 1979.

Hart CB and Giszter SF. Modular premotor drives and unit bursts asprimitives for frog motor behaviors. J Neurosci 24: 526982, 2004.

Heglund NC, Cavagna GA, and Taylor CR. Energetics and mechanics ofterrestrial locomotion. III. Energy changes of the centre of mass as a

function of speed and body size in birds and mammals. J Exp Biol 97:4156, 1982.

Hogue RE. Upper-extremity muscular activity at different cadences andinclines during normal gait. Phys Ther 49: 963972, 1969.

Hreljac A. Preferred and energetically optimal gait transition speeds in humanlocomotion. Med Sci Sports Exerc 25: 11581162, 1993.

Hreljac A. Determinants of the gait transition speed during human locomo-tion: kinematic factors. J Biomech 28: 669677, 1995.

Hultborn H. State-dependent modulation of sensory feedback. J Physiol 533:513, 2001.

Ivanenko YP, Cappellini G, Dominici N, Poppele RE, and Lacquaniti F.

Coordination of locomotion with voluntary movements in humans. J Neu-rosci 25: 72387253, 2005.

Ivanenko YP, Dominici N, Cappellini G, Dan B, Cheron G, and Lacqua-

niti F. Development of pendulum mechanism and kinematic coordinationfrom the first unsupported steps in toddlers. J Exp Biol 207: 37973810,

2004a.Ivanenko YP, Grasso R, Macellari V, and Lacquaniti F. Control of foot

trajectory in human locomotion: role of ground contact forces in simulatedreduced gravity. J Neurophysiol 87: 30703089, 2002.

Ivanenko YP, Grasso R, Zago M, Molinari M, Scivoletto G, Castellano V,Macellari V, and Lacquaniti F. Temporal components of the motorpatterns expressed by the human spinal cord reflect foot kinematics. J Neu-rophysiol 90: 35553565, 2003.

Ivanenko YP, Poppele RE, and Lacquaniti F. Five basic muscle activationpatterns account for muscle activity during human locomotion. J Physiol556: 267282, 2004b.

Ivanenko YP, Poppele RE, and Lacquaniti F. Spinal cord maps of spatio-temporal alpha-motoneuron activation in humans walking at differentspeeds. J Neurophysiol 95: 602618, 2006.

Jordan IM. Brain stem and spinal cord mechanisms for the initiation oflocomotion. In: Neurobiological Basis of Human Locomotion, edited by

Shimamura M, Grillner S, and Edgerton VR. Tokyo: Japan ScientificSocieties Press, 1991, p. 320.Kargo WJ and Giszter SF. Rapid correction of aimed movements by

summation of force-field primitives. J Neurosci 20: 409426, 2000.Kendall FP, McCreary EK, and Provance PG. Muscles. Testing and

Function. Baltimore, MD: Williams and Wilkins, 1993.Kram R, Domingo A, and Ferris DP. Effect of reduced gravity on the

preferred walk-run transition speed. J Exp Biol 200: 821826, 1997.Krebs HI, Aisen ML, Volpe BT, and Hogan N. Quantization of continuous

arm movements in humans with brain injury. Proc Natl Acad Sci USA 96:46454649, 1999.

Lacquaniti F, Grasso R, and Zago M. Motor patterns in walking. NewsPhysiol Sci 14: 168174, 1999.

Lee CR and Farley CT. Determinants of the center of mass trajectory inhuman walking and running. J Exp Biol 201: 29352944, 1998.

Lee DD and Seung HS. Learning the parts of objects by non-negative matrixfactorization. Nature 401: 788 791, 1999.




13/13

Makeig S, Debener S, Onton J, and Delorme A. Mining event-related braindynamics. Trends Cogn Sci 8: 204210, 2004.

Margaria R. Biomechanics and Energetics of Muscular Exercise. Oxford:Clarendon, 1976.

Merkle LA, Layne CS, Bloomberg JJ, and Zhang JJ. Using factor analysisto identify neuromuscular synergies during treadmill walking. J Neurosci

Methods 82: 207214, 1998.Minetti AE, Ardigo LP, and Saibene F. The transition between walking and

running in humans: metabolic and mechanical aspects at different gradients.Acta Physiol Scand 150: 315323, 1994.Mori S, Matsuyama K, Miyashita E, Nakajima K, and Asanome M. Basic

neurophysiology of primate locomotion. Folia Primatol 66: 192203, 1996.Mori S, Sakamoto T, Ohta Y, Takakusaki K, and Matsuyama K. Site-

specific postural and locomotor changes evoked in awake, freely movingintact cats by stimulating the brain stem. Brain Res 505: 6674, 1989.

Murray MP, Mollinger LA, Gardner GM, and Sepic SB. Kinematic andEMG patterns during slow, free, and fast walking. J Orthop Res 2: 272280,1984.

Nene A, Byrne C, and Hermens H. Is rectus femoris really a part ofquadriceps? Assessment of rectus femoris function during gait in able-bodied adults. Gait Posture 20: 113, 2004.

Neptune RR and Sasaki K. Ankle plantar flexor force production is animportant determinant of the preferred walk-to-run transition speed. J Exp

Biol 208: 799808, 2005.Nilsson J, Thorstensson A, and Halbertsma J. Changes in leg movements

and muscle activity with speed of locomotion and mode of progression inhumans. Acta Physiol Scand 123: 457475, 1985.

Olree KS and Vaughan CL. Fundamental patterns of bilateral muscle activityin human locomotion. Biol Cybern 73: 409414, 1995.

Orlovsky GN, Deliagina TG, and Grillner S. Neural Control of Locomotion.From mollusc to man. Oxford, UK: Oxford Univ. Press, 1999.

Patla AE, Calvert TW, and Stein RB. Model of a pattern generator forlocomotion in mammals. Am J Physiol 248: R484494, 1985.

Pearson KG. Generating the walking gait: role of sensory feedback. ProgBrain Res 143: 123129, 2004.

Pribe C, Grossberg S, and Cohen MA. Neural control of interlimb oscilla-tions. II. Biped and quadruped gaits and bifurcations. Biol Cybern 77:141152, 1997.

Prilutsky BI and Gregor RJ. Swing- and support-related muscle actionsdifferentially trigger human walk-run and run-walk transitions. J Exp Biol204: 22772287, 2001.

Rab GT. Muscle. In: Human Walking (2nd ed.), edited by Rose J and GambleJG. Baltimore, MD: Willaims and Wilkins, 1994, p. 101102.

Raibert MH. Legged Robots That Balance. Cambridge, MA: MIT Press,

1986.Raynor AJ, Yi CJ, Abernethy B, and Jong QJ. Are transitions in human gait

determined by mechanical, kinetic or energetic factors? Hum Mov Sci 21:785805, 2002.

Rodgers MM. Dynamic biomechanics of the normal foot and ankle duringwalking and running. Phys Ther 68: 18221830, 1988.

Shik ML. Action of the brainstem locomotor region on spinal stepping

generators via propriospinal pathways. In: Spinal Cord Reconstruction,edited by Kao CC, Bunge RP and Reier PJ. New York: Raven, 1983, p.421434.

Shik ML, Severin FV, and Orlovsky GN. Control of walking and running bymeans of electrical stimulation of the midbrain. Biophysics 11: 756765,1966.

Smith JL, Chung SH, and Zernicke RF. Gait-related motor patterns andhindlimb kinetics for the cat trot and gallop. Exp Brain Res 94: 308322,

1993.Thorstensson A and Roberthson H. Adaptations to changing speed in human

locomotion: speed of transition between walking and running. Acta PhysiolScand 131: 211214, 1987.

Ting LH and Macpherson JM. A limited set of muscle synergies for forcecontrol during a postural task. J Neurophysiol 93: 609613, 2005.

Tresch MC and Bizzi E. Responses to spinal microstimulation in the chron-ically spinalized rat and their relationship to spinal systems activated by low

threshold cutaneous stimulation. Exp Brain Res 129: 401416, 1999.Tresch MC, Cheung VCK, and dAvella A. Matrix factorization algorithms

for the identification of muscle synergies: evaluation on simulated andexperimental data sets. J Neurophysiol 95: 21992212, 2006.

Winter DA. The Biomechanics and Motor Control of Human Gait: Normal,Elderly and Pathological. Waterloo, Ontario: Waterloo Biomechanics,1991.

Winter DA and Yack HJ. EMG profiles during normal human walking:stride-to-stride and inter-subject variability. Electroencephalogr Clin Neu-rophysiol 67: 402411, 1987.

Zajac FE, Neptune RR, and Kautz SA. Biomechanics and muscle coordi-nation of human walking. II. Lessons from dynamical simulations andclinical implications. Gait Posture 17: 117, 2003.

Zatsiorsky V, Seluyanov V, and Chugunova L. In vivo body segmentinertial parameters determination using a gamma-scanner method. In: Bio-mechanics of Human Movement: Applications in Rehabilitation, Sports and

Ergonomics, edited by Berme N and Cappozzo A. Worthington, OH: Bertec,1990, p. 186202.


J Neurophysiol VOL 95 JUNE 2006 www jn org

motor patterns in human walking and running

Documents