There has been a drastic increase in the popularity of minimal footwear in recent years (18,21,24). However, research has focused primarily on the differences in muscle activation, kinetics, and kinematics among footwear conditions during impact when running, specifically between barefoot, minimal, and traditional running shoes (2,4,8,9,17,27). Increased peak vertical and peak horizontal ground reaction forces (GRF) have been found during the push-off phase of running when wearing both minimal and running shoes compared with barefoot (19). These findings suggest that thicker soles may allow for increased vertical and horizontal GRF during the concentric phase (i.e., late stance) of running. Differences in timing of leg muscle activity between barefoot and shod running (29), as well as between different nonminimal footwear conditions (7,30) have been reported. When comparing barefoot, neutral shoes and motion-control shoes during running, peak tibialis anterior (TA) activation after ground contact occurs earlier when barefoot compared with both shoe types (29). Muscle activity in various shod conditions also differs in quiet standing, as an increased magnitude of lower extremity muscle contractions occurs when wearing shoes, with further increases occurring with thicker soles (14,26). After unanticipated foot inversions during standing in different footwear conditions (i.e., barefoot, standard shoe, and shoes with 2.5- and 5-cm sole adaptations), Ramanathan et al. (20) found an increase in muscle activity of the peroneus longus (PL) when in shoes compared with barefoot. Peroneus longus activation also seemed to occur earlier in the shod conditions, which may suggest a preparatory ankle strategy from central (brain) control to prevent instability. In single-leg landing, significant differences in total dynamic postural stability (i.e., the square root of the sum of the mean squared anterior-posterior, vertical, and medial-lateral force, for 3 seconds after landing) have been shown between footwear conditions (23), with barefoot yielding increased dynamic stability compared with shod. The differences found may be attributed to the reduced foot-to-ground interaction that is associated with the presence of additional materials on the sole of a traditional shoe.
In addition to the kinetics and muscle activity during running and quiet standing, footwear may influence other activities like jumping. In an investigation examining various jump tests, LaPorta et al. (15) observed significant differences in maximum single-effort vertical jump (VJ) height (no arm swing), with barefoot and minimal conditions producing significantly higher jumps than shod. Their findings also showed that vertical GRF were greater in repetitive jumps when barefoot and in minimal footwear, contributing to increased jump height (15). These results may be explained by the cushion of the shod condition dissipating applied forces instead of directing it into the ground during propulsion. Peak power relative to weight during jumping was significantly higher when barefoot compared with shod and also when compared with the minimal (Adidas adipure Trainer; Adidas America, Inc., Portland, OR, USA) footwear condition (15). The calculated effect size (ES), however, was small (0.39), indicating that a practical difference may not have been present. The “minimal” adipure Trainer shoes also seem to be heavier and more substantial than other minimal footwear options, such as the Vibram FiveFingers (Vibram USA, Concord, MA, USA). These characteristics may explain some of the observed differences between barefoot and minimal shoe conditions and further emphasizes the need for clearly defined minimal-style shoes.
The previously mentioned results need to be considered in the context of the actual footwear worn. Although minimal footwear has been characterized as footwear with a thin and flexible sole, 0–4 mm heel-toe drop, and less than 170 g (6 ounces) in weight (3), there is no clear definition of what a “minimal” shoe is for consumers. The terms “barefoot” and “minimal” are used extensively in the marketing of footwear, making a clear definition elusive. In a 2001 survey of 6,082 runners, over 30% indicated they had tried minimal footwear, and 85% of those stated they would likely continue to use such footwear (24). Different types of shoes listed by the respondents in that study were the Vibram FiveFingers (Vibram USA), Nike Free (Nike, Inc., Beaverton, OR, USA), Saucony Kinvara (Saucony, Inc., Lexington, MA, USA), and the New Balance Minimus (New Balance, Boston, MA, USA). The Vibram FiveFingers shoe consists of 3.5-mm sole thickness and weighs up to 170 g, depending on the size and model (28), therefore meeting the criteria for minimal footwear (3). Other “minimal” footwear options from various companies can weigh close to 255 g and have a sole thickness well over 4 mm. The large degree of variability in design and weight of these shoes makes interpretation of Rothschild's study difficult (24). Nonetheless, the increase in popularity of minimal footwear within the running and recreational fitness communities may be due to the minimal footwear's purported close representation of the natural barefoot motion (5,15,25,27).
The purpose of this investigation was to examine the acute effects of barefoot, minimal shoes and cross-training shoes on jump performance, GRF, and EMG activity of lower extremity muscles during the VJ and standing long jump (SLJ). To the best of our knowledge, no research has examined the effects of minimal footwear on muscle activity during jumping. We hypothesized that increased jump displacement would occur for the vertical and SLJ in barefoot and minimal shoes compared with shod. Second, we hypothesized that there would be an increase in vertical GRF for both jump types and an increase in horizontal GRF for the SLJ when barefoot and minimal shoes compared with shod. Finally, we expected greater peak EMG activity in barefoot and minimal shoes compared with shod for all tested muscles during both jumps. For further data interpretation, we also studied duration of muscle activation.
Experimental Approach to the Problem
Because of the limited knowledge of the effects of footwear type on nonrunning activities, we aimed to investigate the possible kinetic and electromyographic differences between various footwear conditions, specifically during jumping movements often used to predict performance (22). We performed this investigation using repeated-measures within-subject comparison of acute vertical and SLJ performance, kinetics and EMG activity between barefoot, minimal footwear, and normally shod conditions in healthy men.
Fifteen men (179.8 ± 6.4 cm; 84.5 ± 8.5 kg; and 23.8 ± 2.3 years) participated in this investigation. Age range for participation was 18-40 years. This sample size was determined considering the jump height data of LaPorta et al, with a proposed ES of 0.88, power (1 − β) of 0.8, α < 0.05 for multiple comparisons, and an estimated minimal correlation of r = 0.7 for repeated measures. All subjects were recreationally trained (active participation in basketball, soccer, resistance training, and running) for at least 6 months before the start of the study and experienced in jump training. Subjects verbally confirmed that they were free, at the time of testing, from any injury to the lower extremities that would limit their ability to participate. Recruitment took place through word-of-mouth and posted flyers on campus. Before performing any tasks in the laboratory, subjects were informed of the testing procedures and expectations and provided written informed consent as approved by the University's institutional review board. Subjects were given an opportunity to ask for clarification of the expectations and testing protocol before providing consent and through the duration of their involvement in the study.
Subjects participated in 1 practice session and 1 testing session with a minimum of 48 hours between. Anthropometric measures of mass, height, and shoe size were recorded during the practice session. The testing session consisted of countermovement VJ and SLJ tests in the following footwear conditions: barefoot, minimal shoes (Vibram FiveFinger KSO), and cross-training shoes (MX623, New Balance). The minimal shoe condition consisted of thin sole with a <3.5 mm heel-toe drop and a mass of about 5.7 oz (161.5 g), whereas the shod condition consisted of a thick sole with a 10-mm heel-toe drop and a mass of about 12.6 oz (358 g). These characteristics indicate our footwear conditions met the requirements of Bowles et al. (3) for minimal and shod footwear specifications. Footwear conditions were supplied by the investigators and had minimal wear. After a dynamic warm up consisting of 5 minutes of stationary bicycling, 10 slow bodyweight squats, 10 jump squats, and 20 forward step walking lunges, subjects completed a brief familiarization period with the starting footwear condition, which included walking within the laboratory and practice attempts of each jump test. A brief familiarization period was also provided immediately before testing in the remaining footwear conditions, but it should be noted that any history of jumping exercises done either barefoot or in minimal footwear was not determined.
Surface EMG sensors were placed on the subjects' dominant leg following the SENIAM procedures (12). Dominant leg was determined by asking participants with which leg they would kick a ball. Peak and average muscle activity during each jump condition was measured for the following muscles: biceps femoris, medial gastrocnemius (MG), PL, semitendinosus/semimembranosus (SEM; combined activity represented by the semitendinosus placement due to the deep semimembranosus position (16)), soleus (SOL), TA, vastus lateralis (VL), and vastus medialis (VM). Before sensor application, the area was shaved to remove any hair, the skin was gently abraded with fine sandpaper to remove any other debris, and finally, the area was cleansed with alcohol. The sensors were placed in line with estimated resting pennation angle, attempting to have the same muscle fibers intersect both sensors. Sensors were 2-cm round Ag/AgCl (Ambu, Inc., Glen Burnie, MD, USA) with an inter-sensor distance of 2 cm, and the ground sensor was placed on the anterior aspect of the patella for signal noise reduction.
Subjects performed as many practice attempts as needed to feel comfortable jumping with the tested conditions within the confines of the experimental setup. Once comfortable, subjects completed 3 attempts at each tested jump: countermovement VJ and countermovement SLJ presented in a randomized order to reduce potential order effects. Subjects used a self-selected depth and arm swing for both jump types. A 1-minute rest period was provided between each attempt, during which surface EMG sensors were checked for adhesion.
Following previously described procedures (1), VJ displacement was measured using a stand-alone Vertec jump trainer (Sports Imports, Hilliard, OH, USA), which was placed adjacent to the force platform. The best trial was defined by jump height to the nearest 1.27 cm, calculated as height measured by the Vertec jump trainer subtracting out 1-hand reach height from a static plantar-flexed position (1). Following the SLJ procedures of Robbins (22), horizontal displacement was measured using premarked long jump mats (Power Systems, Knoxville, TN, USA). From a standing, 2-footed position, subjects jumped as far forward as possible using a countermovement with arm swing. Horizontal displacement was measured from the distance of the marked start line to the point of heel contact on landing as monitored visually by 2 of 3 present investigators. Stance width at takeoff was measured during the practice attempts and marked to ensure consistency for all jumps across footwear conditions. After the completion of all jumps in each condition, subjects completed an in-house survey providing qualitative feedback about their preference for comfort, confidence, and perceived jumping performance as well as providing their weekly use (in hours) of each condition.
Electromyographic signals were recorded using Myopac Jr. (RUN Technologies; Mission Viejo, CA, USA) with 8 dual-lead channels. The system had a common mode rejection of 90 dB and an input impedance of 1.0 MΩ. Gain was set at 1000. Data were collected at 2000 Hz (Datapac 5; RUN Technologies; Mission Viejo, CA, USA) and channeled through a 12-bit analog-to-digital converter (DAS1200Jr; Measurement Computing, Middleboro, MA, USA). During offline analysis using the Visual3D biomechanical software suite (C-Motion, Germantown, MD, USA), raw surface EMG signals were band-pass filtered using a fourth-order Butterworth digital filter (10–450 Hz cutoff). The band-pass signals were full-wave rectified and then smoothed using a root mean square (RMS) filter with a moving window of 125 milliseconds (10). Activation onsets were determined from the time between the start of the countermovement (described below) and the time when each respective muscle's EMG signal reached a threshold that was 2 SDs greater than that of a separate 10-second static standing trial (31). For the static standing trial, we asked our subjects to stand still with arms crossed over the chest.
Take-off GRF were acquired with a 3-dimensional force platform (BP600900; AMTI, Watertown, MA, USA). The force platform was interfaced to a PC through 12-bit analog-digital converter (DAS1200Jr; Measurement Computing, Middleboro, MA, USA). Data were sampled at 2,000 Hz using Datapac 5 (RUN Technologies, Mission Viejo, CA, USA) and temporally synchronized with the EMG data. The analog GRF data were smoothed using a fourth-order low-pass Butterworth filter with a cutoff frequency of 50 Hz. The start of the countermovement phase was defined as the time at which vertical force was 30 N less than that of static bodyweight to represent initial unloading of our subjects. Impulse was computed by taking the time integral of the vertical GRF (area under the vertical GRF curve), and we identified the event in which impulse crossed zero to locate the body's lowest point during the countermovement. This point represents the end of the countermovement phase and the start of the propulsive phase (movement of the body changed its vertical direction) (6). The end of the propulsive phase was determined when the vertical force decreased to a threshold of 10 N.
Repeated-measures analyses of variance were used to compare muscle activity and jump kinetics in each of the 3 footwear conditions. Dependent t-tests were used for post hoc analysis, using the Least Significant Difference (LSD) method for multiple comparisons. Standardized ESs were also calculated for pairwise comparisons with corrections for correlations of repeated measures, using the magnitude interpretation of Hopkins (13). Statistical significance was set a priori at p ≤ 0.05.
Jump Displacement and Kinetics
No significant differences in displacement were found for the VJ (p = 0.808; Figure 1) and SLJ (p = 0.766; Figure 2) between the 3 footwear conditions. Similarly, no statistical differences were present in peak vertical ground reaction force between the 3 footwear conditions (Table 1) for the VJ (p = 0.257) and SLJ (p = 0.669). Peak horizontal ground reaction forces (GRFy) also showed no statistical differences between footwear conditions for the VJ (p = 0.097) and the SLJ (p = 0.301). However, the ES for the VJ revealed a moderate-to-large difference for peak anterior-posterior ground reaction force (GRFy) between barefoot and shod (ES = 0.82; Table 1).
The duration of the countermovement phase was not significantly different between footwear conditions for the VJ (p = 0.629) and SLJ tests (p = 0.746), nor was propulsive phase durations for the VJ (p = 0.181) and SLJ (p = 0.951). Moderate-to-large ES values, however, were present for the VJ between barefoot and shod, and minimal shoes and shod (ES = 0.85 and 0.66, respectively; Table 1). Vertical impulse showed no statistical differences for the VJ or SLJ in the countermovement (p = 0.061 and 0.897, respectively), or propulsive phases (p = 0.192 and 0.295, respectively).
No significant differences in peak RMS were present for any of the 8 muscles examined during the countermovement phase for both the VJ and SLJ tests. Significant differences were found in the omnibus test for peak propulsive RMS for SOL (p = 0.026) and VM (p = 0.047) during the VJ test (Table 2). Post hoc dependent t-tests revealed significant differences between barefoot and minimal shoes (p = 0.030) and minimal shoes and shod (p = 0.031) conditions for SOL, whereas no differences were found after between conditions for the VM. Moderate-to-large ES values (Table 2) were present in the SLJ between barefoot and minimal shoes (ES = 0.81) for the MG and between minimal shoes and shod for the PL (ES = 0.98). Large ES values were also found in the VJ between barefoot and minimal shoes (p = 0.68; ES = 1.24) and minimal shoes and shod (p = 0.053; ES = 1.36) for the VM.
No significant differences in average countermovement RMS were found during the SLJ between conditions for any of the 8 examined muscles (Table 3). Significant differences were found in the omnibus test for average countermovement RMS amplitude of the SEM during the VJ (p = 0.042), and dependent t-tests revealed a significant pairwise difference between barefoot and shod (p = 0.039). Additionally, a moderate ES value (p = 0.053; ES = 0.88) was seen for SEM activity between minimal shoes and shod. Moderate-to-large ES values were also present for TA between barefoot and shod and minimal shoes and shod (ES = 0.59 and 0.75, respectively). For average propulsive RMS, no significant differences were found between any of the 8 muscles for either jump test.
Duration of muscle activation showed no significant differences across conditions for any muscle during the countermovement phase for both jump tests (Table 4), although moderate-to-large ES values were present for the VL during the VJ between barefoot and shod and between minimal shoes and shod (ES = 0.63 and 1.2, respectively). Moderate ES values were also found for SOL during the SLJ between barefoot and minimal shoes (ES = 0.59), and between minimal shoes and shod (ES = 0.70). No significant differences or meaningful effects were found for the duration of RMS activation for the propulsive phase during either jump test.
Qualitative feedback from our subjects revealed that jump performance was perceived to be best when shod during the VJ (53% of the responses compared with 27% barefoot and 20% minimal shoes). However, during the SLJ, perceived performance was equal between conditions (33% of the responses went to each condition). Concerning comfort during each jump, subjects indicated that comfort was superior when shod (60% of responses compared with 27% barefoot and 13% minimal shoes) for both the VJ and SLJ.
The purpose of this investigation was to examine the effects of different footwear conditions on jump performance, GRF, and EMG activity of lower extremity muscles during the vertical and SLJ. Contrary to our first hypothesis, VJ and SLJ displacements were not different between footwear conditions. LaPorta et al. (15) found that barefoot and minimal shoes yielded higher VJs than shod (44.5 ± 4.46 cm barefoot; 43.47 ± 5.5 cm minimal shoes; 41.47 ± 14.45 cm shod). The dissimilar findings between our studies may be explained by our subjects' performance of all jump attempts in each tested footwear condition on the same day. LaPorta et al. (15) allotted a specific day for each footwear condition, perhaps allowing their subjects to become more familiar with the conditions. Additionally, the added use of the arm swing in this study may have been influential, as the addition of the arm swing to a countermovement significantly increases the displacement of a jump (11). It is possible that arm swing eliminated any differences in jump performance between footwear conditions assuming it was inconsistent between attempts. However, athletes and practitioners frequently use an arm swing while jumping in training and performance situations, and therefore, allowing our subjects to freely use their arms during jumps may be a more realistic movement pattern. The lack of differences in SLJ displacement may also be explained by the inclusion of arm swing, but our investigation provides the first comparison of SLJ performance between footwear and suggests no effects of footwear. Furthermore, potential differences in shoe mass (particularly between barefoot and shod; 358 g per shoe) as well as traction and stability (between all conditions) may have played a role in this finding; however, it may be unlikely.
Our second hypothesis was not supported, as no significant differences were found in peak vertical and horizontal (anterior-posterior) GRF between footwear conditions for both jumps. Although running and jumping require different mechanics such as single vs. double limb support during push-off, both require sufficient vertical force to become airborne (6). This investigation followed previous guidelines regarding minimal and shod specifications (3), specifically related to sole height, heel drop, and mass. It is unclear whether LaPorta et al. followed the same criteria. Although LaPorta et al. previously failed to find significant differences between footwear conditions for peak vertical GRF, they did find differences in VJ displacement (discussed above). We initially attributed their lack of reported differences in vertical GRF to using a “minimal” footwear condition that may not actually be minimal. Because of this, we expected there to be a difference in vertical GRF as well as jump displacement between conditions in our investigation. However, our expectations of finding differences in vertical GRF and jump displacement were not supported. In running, Squadrone and Gallozzi (27) found no difference in peak vertical GRF during the push-off phase (i.e., propulsive phase) during running stance between barefoot, minimal shoes, and shod. Their results occurred even with a more similar minimal shoe condition to that used in this investigation and the use of a 10-day familiarization period to the minimal condition. These findings suggest that our lack of a longer familiarization period may not have played a major role in our results. Conversely, Paquette et al. found larger late-stance peak vertical and peak horizontal GRF during running in minimal shoes and running shoes compared with barefoot. No familiarization period was used in their study, further suggesting that the lack of a familiarization period may not have played a role in our findings. It is possible that changes in joint mechanics at the hip, knee, and ankle occur after an acute switch in footwear condition. The presence of reduced involvement of the ankle seems to produce lower concentric (propulsive) ankle power when running barefoot compared with minimal shoes and shod, thus reducing vertical GRF (19). This occurrence may act to reduce vertical GRF during propulsion in jumping as well, but it should be noted that this was untested in our investigation. Although we did not find performance or kinetic differences between footwear conditions, we did find shorter propulsive phase durations for the VJ in the shod condition, with a 49.0-millisecond reduction compared with barefoot and 52.0-millisecond reduction compared with minimal shoe condition. This may be the result of constrained foot-joint motion when shod which would shorten the time of force application. In fact, propulsive vertical impulse was slightly lower when shod compared with barefoot and minimal shoes in this study, which may potentially be due to the shortened time of force application. However, it should again be noted that jump performance was unaffected by footwear condition.
For muscle activation during the countermovement phase, our third hypothesis was also not supported. Our results suggest that during the countermovement phase of jumping, peak activation of the muscles tested herein is not affected by changing footwear condition. However, average activation of the SEM was significantly higher when shod, and a moderate-to-large effect was present for PL, and TA during the VJ. This may be a response to a reduced proprioceptive sensitivity that is associated with shod conditions, as previously noted during dynamic stability after single-leg jump landings (23). In this study, barefoot and minimal shoe conditions may have allowed for greater sensory input due to the lack of or a thinner sole compared with the shod condition. This may suggest that certain muscles are more sensitive to different sole thicknesses during the countermovement phase of jumping.
For muscle activation during the propulsive phase, our third hypothesis was also not supported by the majority of muscles tested, although peak propulsive activation of the MG and PL during the SLJ showed higher values (moderate ESs) when barefoot and minimal shoes compared shod. In walking, Scott et al. saw a similar occurrence, as MG and PL peak activation during propulsion was significantly higher when barefoot compared with shod. It was suggested that the increased stability of the foot-ankle complex provided by the shod condition might have influenced these findings (26). We believe that the shod condition used in this study may have provided increased foot stiffness (i.e., restricted foot motion) compared with barefoot and minimal shoes, thus increasing stability of the foot-ankle complex during the propulsive phase. The SOL showed significantly greater peak activation when barefoot and shod compared minimal shoes during the VJ, suggesting that acute changes in peak propulsive muscle activation may not be consistent. Cheung and Ng (7), previously suggested that running in shoes with noticeable cushion significantly delays the onset of VM activation and that delay continues to increase with distance. Our results show similar findings, as moderate-to-large effects were present between barefoot and shod (ES = 0.63) and minimal shoes and shod (ES = 1.2), as evident by a delayed response (average countermovement duration of activation) in VM activation occurring when shod compared with barefoot (50 milliseconds) and minimal shoes (90 milliseconds) during the VJ.
The responses to the qualitative survey are interesting considering the kinetic and EMG results of this investigation. As a group, our subjects felt that they performed the best and felt the most comfortable when shod, although shod did not yield superior jump performance within the resolution of our measurements. Additionally, the increased comfort noted by our subjects combined with the increased activation of certain muscles (PL, SEM, and TA) during both jumps is in agreement with previous findings (29), suggesting that increased muscle activation may be correlated with the reported comfort of the shoe. This may prove valuable when initially selecting a shoe before performing a VJ or SLJ, as initial comfort might not be the best indication of improved acute performance or muscular function. It should be noted that since our subjects were likely habitually shod, comfort might be higher in this condition. No questions specifically considered landing from a jump, and thus, landing might be of concern to those who are habitually shod. A potential limitation of our study may have been not collecting information related to subject history and experience jumping while using any of the specific footwear conditions used in our protocol. It is possible that some subjects were current or previous wearers of 1 or more of the tested footwear conditions, potentially influencing their results. Furthermore, our shod condition consisted of a very firm forefoot, which may have eliminated any dissipation of applied forces to the ground that we expected to see from the added material on the shoe sole. Kinematic analyses may be useful to provide further information to support our findings and to confirm whether changes in joint mechanics at the hip, knee, or ankle occur from an acute change to one of the tested footwear conditions. Additionally, musculoskeletal modeling techniques may provide support for the acute changes in intermuscular coordination patterns that were found in select leg muscles, thus providing insight to possible benefits of long-term exposure to any of the conditions tested herein. This may also provide valuable information regarding possible health implications in addition to possible performance changes.
Although other studies have found differences in VJ performance between footwear conditions in men, our results suggest that no acute differences in jump displacement occurred between footwear conditions. Some differences in acute muscle activation and timing seem to be present during the VJ, and thus, potential training effects should be examined in future studies to suggest these muscle activation changes are more consistent than random. Because of the large degree of variability within minimal and traditional shoes, conclusions on the efficacy of the different types for nonrunning performance may remain elusive. Coaches and practitioners interested in incorporating barefoot and minimal shoes to their training practices would be wise to allow for a familiarization period to the condition to ensure high comfort levels, performance, and consistent muscle activation.
The authors specially thank Daniel Melcher and Lisa Smith for their contributions to data collection for this investigation. The National Strength and Condition Association's Graduate Research Grant—Master's Level was awarded to the first author. For this grant, the amount of $5,000 was awarded in June 2013 and partially funded this project.
1. Barnes JL, Schilling BK, Falvo MJ, Weiss LW, Creasy AK, Fry AC. Relationship of jumping and agility performance in female volleyball athletes. J Strength Cond Res 21: 1192–1196, 2007.
2. Bishop M, Fiolkowsi P, Conrad B, Brunt D. Athletic footwear, leg stiffness, and running kinematics. J Athl Train 387, 2006.
3. Bowles C, Ambegaonkar JP, Cortes N, Caswell S. Footwear for distance runners: The minimalism trend. Int J Athl Ther Train 17: 14–18, 2012.
4. Braunstein B, Arampatzis A, Eysel P, Bruggemann GP. Footwear affects the gearing at the ankle and knee joints during running. J Biomech 43: 2120–2125, 2010.
5. Bruggemann GP, Potthast W, Braunstein B, Niehoff A. Effect of increased mechanical stimuli on foot muscles functional capacity. ISB XXth Congress—ASB 29th Annual Meeting 553, 2005.
6. Chapman A. Biomechanical Analysis of Fundamental Human Movements. Champaign, IL: Human Kinetics, 2008.
7. Cheung RT, Ng GY. Motion control shoe affects temporal activity of quadriceps in runners. Br J Sports Med 43: 943–947, 2009.
8. De Witt B, De Clercq D. Timing of lower extremity motions during barefoot and shod running at three velocities./Cinetique des mouvements des membres inferieurs lors de la course a trois vitesses, pieds nus ou avec chaussures. J Appl Biomech 16: 169–179, 2000.
9. Divert C, Mornieux G, Baur H, Mayer F, Belli A. Mechanical comparison of barefoot and shod running. Int J Sports Med 26: 593–598, 2005.
10. Fauth ML, Petushek EJ, Feldmann CR, Hsu BE, Garceau LR, Lutsch BN, Ebben WP. Reliability of surface electromyography during maximal voluntary isometric contractions, jump landings, and cutting. J Strength Cond Res 24: 1131–1137, 2010.
11. Harman EA, Rosenstein MT, Frykman PN, Rosenstein RM. The effects of arms and countermovement on vertical jumping. Med Sci Sports Exerc 22: 825–833, 1990.
12. Hermens H, Freriks B, Merletti R, Stegeman D, Blok J, Rau G, Disselhorst-Klug C, Hagg G. SENIAM 8 European Recommendations for Surface ElectroMyoGraphy. Enschede, The Netherlands; Roessingh Research and Development, 1999.
13. Hopkins WG. A new view of statistics. 1997. Available at: http://sportsci.org/resource/stats/index.html
. Accessed December 2013.
14. Komi PV, Gollhofer A, Schmidtbleicher D, Frick U. Interaction between man and shoe in running: Considerations for a more comprehensive measurement approach. Int J Sports Med 8: 196–202, 1987.
15. LaPorta JW, Brown LE, Coburn JW, Galpin AJ, Tufano JJ, Cazas VL, Tan JG. Effects of different footwear on vertical jump and landing parameters. J Strength Cond Res 27: 733–737, 2013.
16. McAllister MJ, Schilling B,K, Hammond KG, Weiss LW, Farney TM. Muscle activation during various hamstring exercises. J Strength Cond Res 27: 181–187, 2013.
17. McCallion C, Donne B, Fleming N, Blanksby B. Acute differences in foot strike and spatiotemporal variables for shod, barefoot or minimalist male runners. J Sports Sci Med 13: 280–286, 2014.
18. McDougall C. Born to Run: A Hidden Tribe, Superathletes, and the Greatest Race the World Has Never Seen. New York, NY: Vintage Books, 2009.
19. Paquette MR, Zhang S, Baumgartner LD. Acute effects of barefoot, minimal shoes, and running shoes on lower limb mechanics in rear and forefoot strike runners. Footwear Sci 5: 9–18, 2013.
20. Ramanathan AK, Parish EJ, Arnold GP, Drew TS, Wang W, Abboud RJ. The influence of shoe sole's varying thickness on lower limb muscle activity. Foot Ankle Surg 17: 218–223, 2011.
21. Rixe JA, Gallo RA, Silvis ML. The barefoot debate: Can minimalist shoes reduce running-related injuries? Curr Sports Med Rep 11: 160–165, 2012.
22. Robbins DW. Relationships between National Football League combine performance measures. J Strength Cond Res 26: 226–231, 2012.
23. Rose W, Bowser B, McGrath R, Salerno J, Wallace J, Davis I. Effects of footwear on balance. Proceedings of the American Society of Biomechanics, Long Beach, CA, 2011.
24. Rothschild C. Primitive running: A survey analysis of runners' interest, participation, and implementation. J Strength Cond Res 26: 2021–2026, 2012.
25. Rothschild C. Running barefoot or in minimalist shoes: Evidence or conjecture? Strength Cond J 34: 8–17, 2012.
26. Scott LA, Murley GS, Wickham JB. The influence of footwear on the electromyographic activity of selected lower limb muscles during walking. J Electromyogr Kinesiol 22: 1010–1016, 2012.
27. Squadrone R, Gallozzi C. Biomechanical and physiological comparison of barefoot and two shod conditions in experienced barefoot runners. J Sports Med Phys Fitness 49: 6–13, 2009.
28. Vibram Five Fingers. 2014. Available at: http://www.vibramfivefingers.com/index.htm
. Accessed February 2014.
29. von Tscharner V, Goepfert B, Nigg BM. Changes in EMG signals for the muscle tibialis anterior while running barefoot or with shoes resolved by non-linearly scaled wavelets. J Biomech 36: 1169–1176, 2003.
30. Wakeling JM, Pascual SA, Nigg BM. Altering muscle activity in the lower extremities by running with different shoes. Med Sci Sports Exerc 34: 1529–1532, 2002.
31. Wulf G, Dufek JS, Lozano L, Pettigrew C. Increased jump height and reduced EMG activity with an external focus. Hum Mov Sci 29: 440–448, 2010.