Human Movement Science 42 (2015) 27–37

Contents lists available at ScienceDirect

Human Movement Science journal homepage:

Acute response to barefoot running in habitually shod males N. Fleming ⇑, J. Walters, J. Grounds, L. Fife, A. Finch Department of Kinesiology, Recreation and Sport, Indiana State University, USA

a r t i c l e

i n f o

PsycINFO classification: 3720 Keywords: Barefoot running Kinematics Electromyography Stride pattern

a b s t r a c t The aim of this study was to examine the immediate effects of barefoot (BF) running on lower limb kinematics and muscle activity in a group of habitually shod runners. Ten male runners with no prior BF or minimalist running experience performed 1-min bouts of treadmill running at 3 velocities in both shod and BF conditions. 2D video data were recorded in order to quantify ankle, knee and hip kinematics. Synchronous kinetic data were recorded from a force plate supporting the treadmill in order to quantify spatiotemporal variables. EMG data were collected from 6 lower limb muscles, quantifying recruitment patterns during discrete phases of the gait cycle. BF running resulted in significantly higher stride frequency and shorter ground contact times (P < .001). Additionally, BF running significantly reduced knee and hip range of motion but increased ankle range of motion during the absorptive phase of the stance. Alterations in ankle kinematics during BF running resulted from increased pre-activation of the medial (P < .05) and lateral (P < .01) gastrocnemius in addition to reductions in pre-activation of the tibialis anterior (P < .05). The results highlight that recruitment patterns and kinematics can change in as little as 30-s of BF running in individuals with no previous BF running experience. Ó 2015 Elsevier B.V. All rights reserved.

⇑ Corresponding author at: Arena C19, Department of Kinesiology, Recreation and Sport, Indiana State University, Terre Haute, IN 47809, USA. Tel.: +1 812 241 0493. E-mail address: neil.fl[email protected] (N. Fleming). 0167-9457/Ó 2015 Elsevier B.V. All rights reserved.


N. Fleming et al. / Human Movement Science 42 (2015) 27–37

1. Introduction The last decade has seen a remarkable increase in the interest and participation in barefoot (BF) or minimalist running. This interest was primarily driven by claims that BF running alters stride mechanics, resulting in a more forefoot strike pattern which attenuates impact forces and may ultimately reduce the risk of long-term injury (Daoud et al., 2012; Lieberman et al., 2010; Robbins & Hanna, 1987). However, recent studies examining the effects of transitioning to minimalist running have reported increased rates of injury (Ryan, Elashi, Newsham-West, & Taunton, 2014) and metatarsal stress reactions (Ridge et al., 2013), following 12 and 10 week transitions, respectively. Other studies which have examined transitioning away from cushioned running shoes have reported no difference in injury rates (McCarthy, Fleming, Donne, & Blanksby, 2014; Warne et al., 2014). Despite the lack of agreement, it is generally accepted that changing footwear or running surface increases the risk of injury (Warden, Burr, & Brukner, 2006). A more detailed examination of the acute response to a transition away from shod running is therefore warranted, in order to better understand the natural adaptive process and minimize potential risk of injury. A significant body of research has previously described the acute effects of BF or minimalist running on stance kinetics (Braunstein, Arampatzis, Eysel, & Bruggemann, 2010; De Wit, De Clercq, & Aerts, 2000; Squadrone & Gallozzi, 2009), foot strike pattern (Lieberman et al., 2010), and more recently 3D joint kinematics (Bonacci et al., 2013) and running economy (Perl, Daoud, & Lieberman, 2012). However, much of this published research used habitually BF runners (Lieberman et al., 2010; Perl et al., 2012; Squadrone & Gallozzi, 2009) and therefore does not directly apply to the general running population. Other studies using habitually shod participants have performed pre-trial familiarization periods in order to provide sufficient time for runners to alter their recruitment patterns and running mechanics (Bonacci et al., 2013; De Wit et al., 2000) or instructed the participants to run with specific foot strike patterns (Lieberman et al., 2010; Perl et al., 2012). There is therefore a lack of literature examining the naturally occurring acute response to BF running in habitually shod runners. While there is general agreement that BF running alters joint kinematics, resulting in greater plantar flexion at initial contact (Bishop, Fiolkowski, Conrad, Brunt, & Horodyski, 2006; De Wit et al., 2000) and a more forefoot strike pattern (Lieberman et al., 2010), it remains unclear if these alterations occur immediately or if they develop following a longer period of neuromuscular adaptation. Furthermore, despite general agreement in the literature regarding the kinematic changes associated with BF running, there is still some disagreement as to the neuromuscular recruitment patterns underlying those changes. Two studies have reported increased activity of the triceps surae (Divert, Mornieux, Baur, Mayer, & Belli, 2005; Olin & Gutierrez, 2013) which in part explains the increased plantar flexion reported by many authors. However, there remains lack of agreement regarding the activity of tibialis anterior in BF. While Olin and Gutierrez (2013) and von Tscharner, Goepfert, and Nigg (2003) reported significantly lower tibialis anterior activity, Divert et al. (2005) reported no differences between BF and shod conditions. Additionally, despite several authors reporting reduced knee range of motion (ROM) during BF running (Bonacci et al., 2013; De Wit et al., 2000), the contribution of knee extensor and flexor muscle activity to these changes remains unknown. Recent studies examining the effect of altering footwear and stride mechanics on impact through the shank have reported conflicting effects. Gruber, Boyer, Derrick, and Hamill (2014) reported that running with a forefoot strike pattern significantly reduced tibial shock, measured using triaxial accelerometry. However, Olin and Gutierrez (2013) observed that barefoot running significantly increased tibial shock compared to shod running, regardless of whether a forefoot or rearfoot strike pattern was adopted. Understanding both the timing and neuromuscular mechanisms underlying kinematic and kinetic alterations is of importance, in order to more safely transition habitually shod runners into minimalist or barefoot running and reduce the risk of injury. Therefore, the primary objective of this study was to examine the acute effect of BF running on lower limb recruitment, kinematics in a group of habitually shod runners. A secondary aim was to compare the effect of velocity across shod and BF conditions.

N. Fleming et al. / Human Movement Science 42 (2015) 27–37


2. Methods 2.1. Experimental design An a priori power analysis was conducted for expected outcomes with an alpha probability of .05 and a power of .8. The analysis indicated n = 10 would provide a statistical power of 80% (G⁄Power v3.0.10 software; Dusseldorf, Germany). Ten healthy, male recreational runners (age 24 ± 3 y, height 1.79 ± 0.07 m, body mass 75.1 ± 9.5 kg, >180 min wk 1 running) participated in the current study, which was approved by the University Health Sciences Ethics Committee. None of the participants had any experience of barefoot or minimalist running. During the initial visit, volunteers reported their prior shoe history and those who had any experience of training in racing flats or other minimalist footwear were excluded from participation. Participants were provided an opportunity to familiarize themselves with all instruments and procedures which would be carried out in visit 2. They were also tested on their ability to run comfortably at the highest treadmill velocity (V3; 4.43 m s 1). If satisfied, they signed informed consent documentation, completed a medical questionnaire and underwent an examination to rule out any contraindications to exercize. During the second visit, participants ran a series of 1-min exercise bouts at fixed velocities and running conditions. Participants initially performed a 10-min warm-up at self-selected pace in their normal running shoes. Reflective markers and EMG electrodes were then placed at discrete anatomical locations on their right leg. The participants then ran at 3 fixed velocities (V1 = 3.13, V2 = 3.80 and V3 = 4.47 m s 1, respectively) on a motorized treadmill (Proform 700 ZLT) in two running conditions (BF and shod). Both velocity and running condition were randomized for each participant; however, all 3 velocities within each condition were performed consecutively, in order to minimize variability in marker placement and reduce setup time between trials. Participants wore black compression socks during the BF trials and their normal cushioned running shoes during the shod trials. In order to facilitate digitization of foot kinematic data, a customized black cycling overshoe was placed over the right running shoe to improve reflective marker contrast. The duration of each trial was 1-min, with 5-min recovery between trials during which the participants remained seated. Data were recorded during the final 30-s of each 1-min trial. No verbal instructions or feedback on running technique was provided at any point during the trials. 2.2. Kinematic data 2D kinematic data of the lower limb were recorded during the final 30s of each trial. Reflective markers were positioned at 6 discrete locations on the right side of each participant at the following anatomical landmarks: (a) 5th metatarsal head; (b) lateral malleolus; (c) lateral calcaneus; (d) lateral femoral condyle; (e) trochanter major; (f) mandibular angle. Sagittal plane kinematics were recorded at 60 Hz with a 0.001 s shutter speed (Panasonic 3CCD, Kadoma, Japan). The camera was positioned at a fixed distance of 10 m orthogonally to the treadmill. Synchronous kinetic data were recorded at 100 Hz from an embedded force plate (Kistler, Winterthur, Switzerland,) which supported the right posterior leg of the treadmill. The measurement of vertical ground reaction forces, resultant from treadmill loading during the stance phase facilitating the identification of initial contact (IC) and toe off (TO). Marker positional data were digitized, transformed using 2D direct linear transformation, and digitally filtered at 10 Hz using APAS 2011 software (APAS v12.1, Ariel Dynamics, CA, USA). Additional data smoothing was accomplished using quintic splines as described by De Wit et al. (2000). 2.3. EMG Data Surface EMG data were collected from 6 superficial muscles of the lower limb using a Trigno wireless acquisition system (Delsys Inc., Boston, MA, USA), according to SENIAM guidelines (Hermens, Freriks, Disselhorst-Klug, & Rau, 2000). Prior to application of electrodes, recording sites were shaved, abraded and cleaned with isopropyl alcohol. Surface electrodes with an inter-electrode distance of 10 mm, 4-bar formation, and bandwidth of 20–450 Hz, were applied to the midline of the palpated


N. Fleming et al. / Human Movement Science 42 (2015) 27–37

muscle belly, parallel to muscle fiber orientation. EMG data were recorded from the rectus femoris (RF), vastus lateralis (VL), biceps femoris (BiF), medial gastrocnemius (GM), lateral gastrocnemius (GL) and tibialis anterior (TA), at a frequency of 2000 Hz and transferred via wireless telemetry to a standalone laptop for subsequent processing and analysis. Synchronization of EMG, kinematic and kinetic data was performed using a customized trigger module (Delsys Inc., Boston, MA, USA) which emitted a 5V square wave of 1-s duration to the Trigno system, AD channel on the Kistler force plate and LED light which was captured in the kinematic data at the onset of all data collection phases. Finally, accelerometry data were recorded at 200 Hz from a triaxial accelerometer (Delsys Inc., Boston, MA, USA) attached to the tibial crest, midway between the tibial tuberosity and medial malleolus, in order to estimate tibial shock associated with impact forces and compare results to previously published literature on habitually shod runners (Gruber et al., 2014; Olin & Gutierrez, 2013). The accelerometer was fixed directly to the skin overlying the tibia using dermatological tape. 2.4. Data reduction All data were transferred to Matlab (Mathworks, Version 7.14, Mathworks, MA, USA) and processed using customized algorithms. Spatiotemporal variables were initially quantified for 10 consecutive stride cycles from vertical ground reaction forces recorded from the force platform. IC was defined as the time point at which vertical ground reaction force exceeded 20 N above baseline. TO was defined as the time point at which vertical ground reaction force dropped to within 20 N of baseline. This procedure has previously been used to quantify the onset of stride cycle during treadmill running (Bosco & Rusko, 1983; McCallion, Donne, Fleming, & Blanksby, 2014). Identification of IC and TO facilitated the subsequent calculation of stride frequency, stride duration, absolute (ms) and normalized (% of stride cycle) ground contact time (GCT). Stride duration was calculated as the time duration (ms) from IC to the next ipsilateral IC. GCT was defined as the time duration (ms) from IC to TO. Raw EMG data were initially band-pass filtered between 50 and 200 Hz, zero offset removed, and root mean squared at a resolution of 20 ms for 10 consecutive stride cycles. Data were subsequently temporally normalized via cubic spline fitting, to account for stride to stride variations in duration. Finally, rmsEMG was quantified and averaged over 10 consecutive cycles for the following phases; (1) pre-activation phase (final 15% of the stride cycle); (2) absorptive phase (from IC to peak knee flexion); (3) propulsive phase from peak knee flexion to TO; as described by Divert et al. (2005). All data were normalized and expressed as a percentage of the maximal EMG detected during the stance phase in any of the six trials (Winter & Yack, 1987). Joint angles from the ankle, knee and hip kinematic data were also temporally normalized via cubic spline fitting and averaged over 10 consecutive stance phases for each trial. Kinematic variables of interest were ankle, knee and hip angle at IC, TO and ROM during the absorptive phase of the stance. In addition, time to peak knee flexion was quantified in order to establish the duration of absorptive phase. Finally, peak and average stance phase accelerometer data in the vertical axis were averaged over 10 consecutive cycles for each trial, as described by Olin and Gutierrez (2013). 2.5. Statistical analysis Normality for all data sets was assessed using Kolmogorov–Smirnov tests and statistical analysis was performed using 2 factor (condition  velocity) repeated measures ANOVA, Tukey post-hoc tests quantified differences within interactions (P < .05 inferring statistical significance). Non-normal data sets (time to peak knee flexion, hip angle at TO, pre-activation in RF, VL, TA, absorptive phase in TA and BiF) were assessed using Mann–Whitney U tests. 3. Results Significant alterations in both lower limb joint kinematics and spatiotemporal variables were observed comparing BF and shod running. Overall, stride frequency was significantly higher (P < .001) and stride duration significant shorter (P < .001) in BF running (see Table 1). Increased stride frequency and shorter stride duration at fixed velocities inferred that participants had shorter stride


N. Fleming et al. / Human Movement Science 42 (2015) 27–37

Table 1 Group mean (SD) sagittal plane joint kinematic and spatiotemporal data. Asterisks infer significant difference between running conditions (⁄P < .05; ⁄⁄P < .01; ⁄⁄⁄P < .001). Plus signs infer significant difference to V1 (+P < .05; ++P < .01; +++P < .001). Alpha signs infer significant difference to V2 (aP < .05; aaP < .01; aaaP < .001). V1









Ankle Angle at IC (°) Angle at TO (°) ROM (°)

4 (5)⁄⁄ 14 (5)⁄ 23 (4)⁄

1 (6) 10 (7) 20 (5)

6 (4)⁄⁄ 18 (6)⁄,++ 26 (4)⁄⁄,+

3 (5) 15 (6)++ 22 (5)

5 (6)⁄⁄ 19 (6)⁄,+++ 26 (5)⁄⁄,+

2 (5) 15 (7)+++ 21 (4)+

Knee Angle at IC (°) Angle at TO (°) ROM (°) Time to peak (%)

19 23 23 38

17 22 28 41

21 22 22 36

18 22 28 40

22 23 23 38

Hip Angle at IC (°) Angle at TO (°) ROM (°)

32 (4) 2 (3) 2 (1)⁄

32 (7) 2 (6) 4 (1)

34 (4)++ 5 (3)+ 1 (1)⁄

34 (7)++ 5 (6)+ 3 (1)

36 (6)+++,a 7 (3)+++ 1 (1)⁄

37 (7)+++,aa 5 (6)++ 3 (1)

87 (5)⁄⁄ 689 (34)⁄⁄ 220 (19) 32 (2)

85 (5) 707 (38) 225 (19) 32 (2)

91 (5)⁄⁄⁄,+++ 656 (34)⁄⁄⁄,+++ 197 (15)⁄,+++ 30 (2) +++

88 (6)+ 688 (46) +++ 206 (18) +++ 30 (2) +++

96 (6)⁄⁄⁄,+++,aa 626 (33)⁄⁄⁄,+++,aa 178 (18)⁄,+++,aaa 28 (2)+++,aa

92 (6)+++,aa 651 (44) +++,aa 188 (15)+++,aaa 29 (2)+++,aa

Spatiotemporal Stride freq. (stride min Stride duration (ms) Absolute GCT (ms) Normalised GCT (%)



(5)⁄ (4) (4)⁄⁄⁄ (4)

(6) (2) (3) (5)

(5)⁄ (4) (3)⁄⁄⁄ (3)⁄

(6) (7) (2) (6)

(5)⁄,+ (3) (5)⁄⁄⁄ (3)⁄

19 23 28 41

(5)+ (7) (2) (6)

lengths during BF running. In addition, GCT was significantly shorter in BF (P < .001); however, when GCT was normalized to stride duration, no significant difference between conditions was observed (see Table 1). As expected, velocity had a significant effect on stride frequency (P < .001), stride duration (P < .001) and both absolute and normalized GCT (P < .01). With regards to kinematic variables, BF running resulted in significantly greater plantar flexion (P < .01 at all velocities) and knee flexion (P < .05 at all velocities) at IC, greater ankle ROM during the absorptive phase (P < .05 at V1; P < .01 at V2 and V3), reduced knee (P < .001 at all velocities) and hip ROM (P < .05 at all velocities) during the absorptive phase, and greater plantar flexion at TO (P < .05 at all velocities); see Table 2. Velocity significantly increased plantar flexion and hip extension at TO, and significantly increased hip and knee flexion at IC (see Table 1). Significant differences in recruitment patterns were observed between conditions for TA, GM, and GL, respectively. During the pre-activation phase in BF running, TA activity was significantly lower (P < .05, see Table 2). Post-hoc tests established that the difference in TA pre-activation was only significant at V1 (8 ± 4 vs. 12 ± 10%, P < .05) and V2 (11 ± 6 vs. 15 ± 12%, P < .05). Differences in TA preactivation were not significant at V3. In contrast, GM and GL pre-activation were significantly greater during BF running (P < .01 and P < .05, respectively; see Table 2). GM pre-activation during BF running was significantly greater at V1 (19 ± 12 vs. 12 ± 8%, P < .05) and V3 (22 ± 11 vs. 16 ± 6%, P < .05); however, the difference was not significant at V2 (21 ± 10 vs. 17 ± 10%, P = .16). GL pre-activation during BF running was significantly greater than shod at all velocities (10 ± 7 vs. 6 ± 3% at V1, P < .05; 12 ± 6 vs. 8 ± 3% at V2, P < .05; 17 ± 10 vs. 11 ± 5% at V3, P < .01). No significant conditional effects were observed for any muscles during the absorptive or propulsive phases of the stance. Velocity had a significant effect on RF, VL, BiF and GL with all muscles exhibiting greater activity with increasing velocities (see Table 2). These effects were observed in the pre-activation, absorptive and propulsive phases of the stance. Velocity did not have an effect on GM or TA activity during any of the phases. Comparison of accelerometer data from the tibia revealed significantly higher peak tibial shock during BF running at V1 (4.3 vs. 3.6 G; P < .05), V2 (5.0 vs. 4.4 G; P < .05) and V3 (5.7 vs. 5.1 G; P < .05), respectively. Mean tibial shock was consistently higher in BF running (0.37 vs. 0.17 G at V1; 0.40 vs. 0.26 G at V2; 0.48 vs. 0.29 G at V3); however, differences between condition were not significant (P = .102). Velocity significantly increased both peak and mean tibial shock in both BF (P < .01) and shod running (P < .01).


N. Fleming et al. / Human Movement Science 42 (2015) 27–37

Table 2 Group mean (SD) muscle activity data. Data are presented as a percentage of maximal stance phase activity recorded during any trial. Asterisks infer significant difference between running conditions (⁄P < .05; ⁄⁄P < .01; ⁄⁄⁄P < .001). Plus signs infer significant difference to V1 (+P < .05; ++P < .01; +++P < .001). Alpha signs infer significant difference to V2 (aP < .05; aaP < .01; aaaP < .001). V1 BF

V2 Shod


Pre-activation RF 8 (7) VL 10 (8) BiF 16 (7) GM 19 (12)⁄ GL 10 (7)⁄ TA 8 (4)⁄

8 (6) 9 (7) 12 (7) 12 (8) 6 (3) 12 (9)

12 12 20 21 12 11

Absorptive phase RF 20 (14) VL 18 (10) BiF 11 (5) GM 29 (12) GL 19 (11) TA 18 (14)

14 15 12 29 15 11

28 25 15 30 20 21

Propulsive phase RF 5 (5) VL 3 (1) BiF 13 (8) GM 14 (7) GL 6 (4) TA 4 (4)

3 (2) 2 (1) 11 (10) 16 (8) 7 (6) 3 (3)

(9) (12) (6) (14) (6) (10)

V3 Shod


(10)+ (7) (9) (10) (6)⁄ (6)⁄

11 (10) 12 (9) 16 (6) 17 (10) 8 (3) 15 (10)

11 14 24 22 17 13

(7)++ (8) (9)+++,a (11)⁄ (10)⁄⁄,+++,a (8)

10 13 21 16 11 16

(7) (10) (8)+++,a (6) (5)++ (9)

(14) (15) (9) (13) (13) (13)

21 23 17 31 18 19

34 27 20 33 26 20

(13)++ (14) (11)++ (14) (12)++,a (9)

23 23 22 31 22 15

(13)+ (14) (11)+++ (10) (10)+ (12)

7 (6) 5 (4) 15 (9) 15 (7) 8 (5) 4 (3)

(12) (15) (10) (14) (8) (17)

5 (4) 4 (3) 14 (10) 17 (7) 9 (7) 6 (4)


11 (8)++,a 7 (4)++ 18 (11)+ 17 (9) 10 (5)++,a 6 (3)

8 (5)+ 7 (6)++ 17 (9)++ 20 (10) 11 (8)+ 10 (7)

4. Discussion The main findings of this study suggest that as little as 30-s of BF running is sufficient to alter the spatiotemporal variables of the stride, joint kinematics and muscle recruitment patterns of the lower limb in a group of individuals with no previous BF or minimalist running experience. The alterations included increased stride frequency, reduced GCT, increased ankle ROM and reduced knee and hip ROM during the absorptive phase. Several of the kinematic adaptations are directly explainable through the significant alterations in muscle activity. The increased stride frequency observed during BF running has been observed by many researchers (Bonacci et al., 2013; De Wit et al., 2000; McCallion et al., 2014) and most likely played a role in reducing the runner’s vertical displacement (Farley & Gonzalez, 1996). Ankle and knee kinematics observed during BF running are also in agreement with previous research reporting increased plantar flexion and knee flexion at IC for BF conditions (Bonacci et al., 2013; De Wit et al., 2000). Reduced knee and hip ROM during the absorptive phase highlights increased leg stiffness during BF running, which is also in agreement with previous literature (De Wit et al., 2000; Perl et al., 2012). It is also likely that increased stride frequency played a direct role in altering knee and hip kinematics, resulting in this increased leg stiffness (Farley & Gonzalez, 1996). In contrast to the hip and knee ROM, BF running significantly increased ankle ROM during the absorptive phase of the stance. These kinematic changes result in greater joint loading at the ankle and reduced loading at the proximal joints (Braunstein et al., 2010). It therefore appears that in less than 1-min of BF running, an individual will increase loading at the ankle via an increase in plantar flexion at IC, reducing the demand for joint loading at the knee and hip. 4.1. Changes at ankle joint BF running has been shown to increase plantar flexion at IC and ankle ROM during the absorptive phase of the stance (De Wit et al., 2000; Lieberman et al., 2010; Perl et al., 2012). These alterations occur in order to produce a more anterior strike pattern and attenuate initial impact peak and high loading rate associated with a rearfoot strike pattern (Giandolini et al., 2013; Lohman, Sackiriyas, &

N. Fleming et al. / Human Movement Science 42 (2015) 27–37


Fig. 1. Group mean stance phase kinematics at V2. Asterisks indicates significant difference between conditions at IC or TO (⁄P < .05; ⁄⁄P < .01; ⁄⁄⁄P < .001); alpha signs indicate significant difference between conditions in ROM (aP < .05; aaP < .01; aaa P < .001); plus signs indicate significant difference between conditions in time to peak flexion (+P < .05).

Fig. 2. Group mean EMG ensembles at V2. Dashed lines indicate 1 SD above or below the mean. BF data are represented in gray; shod data are represented in black. Vertical dashed lines at 30% indicate the end of stance phase (TO).

Swen, 2011; Warne et al., 2014). The findings of the current study highlight that changes in ankle kinematics appear rapidly following initial exposure. Increasing ankle ROM during the absorptive phase of the stance was primarily accomplished via increased plantar flexion at IC, since no change in peak dorsiflexion was observed; see Fig. 1. This increased plantar flexion at IC was brought about via enhanced pre-activation of GM and GL and attenuated TA activity in the pre-activation phase; see Figs. 2 and 3. Increased pre-activation of the triceps surae complex during BF running has previously been reported by Divert et al. (2005); however, no difference in TA pre-activation was reported in that study. The current results suggest significant alterations in pre-activation of both plantar- and dorsi-flexors facilitate the changes in ankle kinematics. Giandolini et al. (2013) also observed significant increases in GL and reductions in TA pre-activation when runners consciously shifted from their normal stride pattern to a more mid-foot strike pattern. Olin and Gutierrez (2013) reported increased mean activation of the


N. Fleming et al. / Human Movement Science 42 (2015) 27–37

Fig. 3. Group mean (SD) EMG data for GM, GL and TA during the pre-activation phase of the stride cycle. Asterisks indicates significant difference between conditions (⁄P < .05; ⁄⁄P < .01); plus signs indicate significant difference between velocities (+P < .05; +++P < .001).

GM and GL during the stance phase; however, the current data does not support those findings. No differences in GM or GL activity were detected during either the absorptive or propulsive phase of the stance in the current study. However, TA activity during the absorptive phase was consistently higher across all velocities in BF. This is most likely due to increased myotatic reflex response occurring in BF running, immediately following impact (Gottlieb & Agarwal, 1979). The tibial shock data is lower than previously reported by Olin and Gutierrez (2013), but similar to data reported by Gruber et al. (2014). This is probably due to minor differences in sensor location on the tibial crest between studies. Overall, the current data is in agreement with Olin and Gutierrez (2013), who also reported significantly higher peak tibial shock during BF running. It is also worth noting that while velocity significantly increased both mean and peak tibial shock data, kinematic and neuromuscular differences in BF condition were for the most part consistent across velocity. It would therefore appear that a lower running velocity is more appropriate for transitioning away from cushioned running shoes, since it appears effective at altering neuromuscular response but does so with less impact on the tibia. 4.2. Changes at knee and hip joints Previous research has reported that BF or minimalist running increases knee flexion at IC and reduces knee ROM during the absorptive phase (De Wit et al., 2000; Perl et al., 2012). It has been suggested that these changes in knee flexion may explain the improved running economy which researchers have observed in habitually BF runners (Perl et al., 2012) and more recently following a 4-week minimalist transition away from cushioned running shoes (Warne & Warrington, 2014). Increasing joint stiffness through co-activation of muscles around the knee is an important biomechanical factor affecting running economy (Kyrolainen, Belli, & Komi, 2001; Saunders, Pyne, Telford, & Hawley, 2004) and by increasing joint stiffness, BF running may improve economy. The EMG data from the current study suggests that increased knee joint stiffness during BF running is brought about via increased activation of the knee extensors (RF and VL) during the absorptive phase of the stride. Both muscles exhibited consistently higher activity during this phase at all velocities (see Table 2); however, these differences were not statistically significant. Interestingly, time to peak knee flexion was significantly shorter during BF running, suggesting that runners also spend less time in the absorptive or ‘‘braking’’ phase and more time in the propulsive phase of the stance. This may explain the improvements in economy reported following a transition away from shod running (Warne & Warrington, 2014), since more time is spent propelling rather than braking. 4.3. Barefoot vs. minimalist running While the current study did not assess minimalist running, it is worth considering the results in the context of previously published work in this area. The current data contradict the findings of Willy and

N. Fleming et al. / Human Movement Science 42 (2015) 27–37


Davis (2014), who examined the acute effects of minimalist running in a group of habitually shod runners. After 1 min of exposure to minimalist footwear (Nike Free 3.0), they reported significantly reduced ankle plantar flexion and increased knee flexion at IC, despite no change in stride rate or stride length (Willy & Davis, 2014). In contrast, participants in the current study increased plantar flexion and knee flexion at IC, in addition to increasing stride rate and reducing stride duration, within 30 s of exposure to BF running. It would therefore appear that BF running may elicit neuromuscular alterations more rapidly than minimalist running; an important finding to consider when transitioning away from cushioned running shoes. Bonacci et al. (2013) reported significant kinematic and kinetic differences between BF and minimalist running (Nike Free 3.0), while McCallion et al. (2014) recently reported that minimalist running (Vibram FiveFingers) is more similar to shod than BF running, when comparing spatiotemporal variables in habitually shod males. While Squadrone and Gallozzi (2009) reported that minimalist footwear (Vibram FiveFingers) was effective at imitating BF running, their cohort were all habitually BF runners and therefore their findings do not reflect acute responses. Nonetheless, there appears to be growing evidence that BF and minimalist running are not the same (Bonacci et al., 2013; McCallion et al., 2014). The current data suggests that initial response to BF running is more pronounced than the previously reported initial response to minimalist footwear (Willy & Davis, 2014). This would suggest that a transition to BF running may be more effective at eliciting neuromuscular and kinematic adaptations than a transition to minimalist footwear. 4.4. Limitations Before drawing definitive conclusions from the current study, several methodological limitations must be considered. Firstly, all tests were performed on a motorized treadmill. Treadmill and overground gait are not identical; therefore the current findings may not directly apply to regular overground running (Leitch, Stebbins, Paolini, & Zavatsky, 2011; Riley et al., 2008). However, treadmill running has advantages for data collection as it allows averaging over consecutive step cycles and the participants are unaware of when data is collected. This is not the case in over-ground running studies which use fixed calibration volumes and imbedded force plates for kinematic and kinetic data acquisition (Bonacci et al., 2013; De Wit et al., 2000). Another limitation of the current study was the collection of 2D joint kinematic data at a sampling rate of 60 Hz. Both sagittal and frontal planes kinematics, recorded at higher sampling rates would be preferable for assessing joint kinematics during running. Unfortunately, higher frequency 3D recording equipment was unavailable for the current study. However, the low sampling rate combined with averaging over 10 consecutive stride cycles greatly smoothed the data for each participant and most likely reduced the resolution for detecting statistically significant changes across condition, compared to higher frequency recordings. It is therefore likely that the kinematic changes would be also observed using higher sampling frequencies. Finally, participants wore thin compression socks to reduce risk of abrasive injury to the feet during BF running. The possibility that sensory feedback from the foot is attenuated by this additional layer cannot be completely ruled out. However, data from previous studies (McCallion et al., 2014) would suggest that differences between shod and BF condition would have likely been of a larger magnitude, had runners not worn this protective layer. 5. Conclusion The main finding of the current study is that habitually shod runners alter their lower limb muscle recruitment patterns and joint kinematics in as little as 30-s of BF running exposure, with no prior familiarization. These differences for the most part appear independent of running velocity. The findings highlight the rapid adjustments that can be made to running pattern and joint kinematics, brought about by acute awareness of altered impact forces by tactile receptors in the foot and proprioceptive organs in the shank. It appears that even runners with no previous BF running experience can rapidly adjust their motor output in response changes underfoot. However, it should be noted that these rapid adjustments in muscle recruitment and kinematics did not appear to reduce stress on the lower limb, since tibial shock was significantly higher during the BF running across all velocities.


N. Fleming et al. / Human Movement Science 42 (2015) 27–37

It therefore remains to be seen if a transition to running barefoot is truly desirable for improved performance or reduced risk of injury. Further research examining long-term injury rates is therefore required before a transition to barefoot running can be recommended. Acknowledgements No funding was received to complete this work. The authors declare no conflicts of interest – financial, professional, or otherwise. References Bishop, M., Fiolkowski, P., Conrad, B., Brunt, D., & Horodyski, M. (2006). Athletic footwear, leg stiffness, and running kinematics. Journal of Athletic Training, 41, 387–392. Bonacci, J., Saunders, P. U., Hicks, A., Rantalainen, T., Vicenzino, B. G., & Spratford, W. (2013). Running in a minimalist and lightweight shoe is not the same as running barefoot: a biomechanical study. British Journal of Sports Medicine, 47, 387–392. Bosco, C., & Rusko, H. (1983). The effect of prolonged skeletal muscle stretch-shortening cycle on recoil of elastic energy and on energy expenditure. Acta Physiologica Scandinavica, 119, 219–224. Braunstein, B., Arampatzis, A., Eysel, P., & Bruggemann, G. (2010). Footwear affects the gearing at the ankle and knee joints during running. Journal of Biomechanics, 43, 2120–2125. Daoud, A. I., Geissler, G. J., Wang, F., Saretsky, J., Daoud, Y. A., & Lieberman, D. E. (2012). Foot strike and injury rates in endurance runners: a retrospective study. Medicine & Science in Sports & Exercise, 44, 1325–1334. De Wit, B., De Clercq, D., & Aerts, P. (2000). Biomechanical analysis of stance phase during barefoot and shod running. Journal of Biomechanics, 33, 269–278. Divert, C., Mornieux, G., Baur, H., Mayer, F., & Belli, A. (2005). Mechanical comparison of barefoot and shod running. International Journal of Sports Medicine, 26, 593–598. Farley, C. T., & Gonzalez, O. (1996). Leg stiffness and stride frequency in human running. Journal of Biomechanics, 29, 181–186. Giandolini, M., Arnal, P. J., Millet, G. Y., Peyrot, N., Samozino, P., Dubois, B., et al (2013). Impact reduction during running: efficiency of simple acute interventions in recreational runners. European Journal of Applied Physiology, 113, 599–609. Gottlieb, G. L., & Agarwal, G. C. (1979). Response to sudden torques about ankle in man: myotatic reflex. Journal of Neurophysiology, 42, 91–106. Gruber, A., Boyer, K., Derrick, T., & Hamill, J. (2014). Impact shock frequency components and attenuation in rearfoot and forefoot running. Journal of Sport and Health Science, 3, 113–121. Hermens, H., Freriks, B., Disselhorst-Klug, C., & Rau, G. (2000). Development of recommendations for SEMG sensors and sensor placement procedures. Journal of Electromyography & Kinesiology, 10, 361–374. Kyrolainen, H., Belli, A., & Komi, P. V. (2001). Biomechanical factors affecting running economy. Medicine & Science in Sports & Exercise, 33, 1330–1337. Leitch, J., Stebbins, J., Paolini, G., & Zavatsky, A. B. (2011). Identifying gait events without a force plate during running: a comparison of methods. Gait & Posture, 33, 130–132. Lieberman, D. E., Venkadesan, M., Werbel, W. A., Daoud, A., D’Andrea, S., Davis, I., et al (2010). Foot strike patterns and collision forces in habitually barefoot versus shod runners. Nature, 463, 531–535. Lohman, E. B., 3rd, Sackiriyas, K. S., & Swen, R. W. (2011). A comparison of the spatiotemporal parameters, kinematics, and biomechanics between shod, unshod, and minimally supported running as compared to walking. Physical Therapy in Sport, 12, 151–163. McCallion, C., Donne, B., Fleming, N., & Blanksby, B. (2014). Acute differences in foot strike and spatiotemporal variables for shod, barefoot or minimalist male runners. Journal of Sports Science & Medicine, 13, 280–286. McCarthy, C., Fleming, N., Donne, B., & Blanksby, B. (2014). 12 weeks of simulated barefoot running changes foot-strike patterns in female runners. International Journal of Sports Medicine, 35, 443–450. Olin, E. D., & Gutierrez, G. M. (2013). EMG and tibial shock upon the first attempt at barefoot running. Human Movement Science, 32, 343–352. Perl, D. P., Daoud, A. I., & Lieberman, D. E. (2012). Effects of footwear and strike type on running economy. Medicine & Science in Sports & Exercise, 44, 1335–1343. Ridge, S. T., Johnson, A. W., Mitchell, U. H., Hunter, I., Robinson, E., Rich, B. S., et al (2013). Foot bone marrow edema after a 10-wk transition to minimalist running shoes. Medicine & Science in Sports & Exercise, 45, 1363–1368. Riley, P. O., Dicharry, J., Franz, J., Della Croce, U., Wilder, R. P., & Kerrigan, D. C. (2008). A kinematics and kinetic comparison of overground and treadmill running. Medicine & Science in Sports & Exercise, 40, 1093–1100. Robbins, S. E., & Hanna, A. M. (1987). Running-related injury prevention through barefoot adaptations. Medicine & Science in Sports & Exercise, 19, 148–156. Ryan, M., Elashi, M., Newsham-West, R., & Taunton, J. (2014). Examining injury risk and pain perception in runners using minimalist footwear. British Journal of Sports Medicine, 48, 1257–1262. Saunders, P. U., Pyne, D., Telford, R., & Hawley, J. (2004). Factors affecting running economy in trained distance runners. Sports Medicine, 34, 465–485. Squadrone, R., & Gallozzi, C. (2009). Biomechanical and physiological comparison of barefoot and two shod conditions in experienced barefoot runners. Journal of Sports Medicine & Physical Fitness, 49, 6–13. von Tscharner, V., Goepfert, B., & Nigg, B. M. (2003). Changes in EMG signals for the muscle tibialis anterior while running barefoot or with shoes resolved by non-linearly scaled wavelets. Journal of Biomechanics, 36, 1169–1176. Warden, S. J., Burr, D. B., & Brukner, P. D. (2006). Stress fractures: pathophysiology, epidemiology, and risk factors. Current Osteoporosis Reports, 4, 103–109.

N. Fleming et al. / Human Movement Science 42 (2015) 27–37


Warne, J. P., Kilduff, S. M., Gregan, B. C., Nevill, A. M., Moran, K. A., & Warrington, G. D. (2014). A 4-week instructed minimalist running transition and gait-retraining changes plantar pressure and force. Scandinavian Journal of Medicine & Science in Sports, 24, 964–973. Warne, J. P., & Warrington, G. D. (2014). Four-week habituation to simulated barefoot running improves running economy when compared with shod running. Scandinavian Journal of Medicine & Science in Sports, 24, 563–568. Willy, R. W., & Davis, I. S. (2014). Kinematic and kinetic comparison of running in standard and minimalist shoes. Medicine & Science in Sports & Exercice, 46, 318–323. Winter, D., & Yack, H. J. (1987). EMG profiles during normal human walking: stride-to-stride and inter-subject variability. Electroencephalography & Clinical Neurophysiology, 67, 402–411.

Acute response to barefoot running in habitually shod males.

The aim of this study was to examine the immediate effects of barefoot (BF) running on lower limb kinematics and muscle activity in a group of habitua...
710KB Sizes 0 Downloads 9 Views