ELMER, STEVEN J.1; BARRATT, PAUL R.2; KORFF, THOMAS2; MARTIN, JAMES C.1
Cycle ergometry is commonly used to quantify muscular work and power and to elicit perturbations to metabolic homeostasis for a broad range of physiological investigations (31). During cycling, power delivered to the ergometer pedals is produced by ankle, knee, and hip joint actions (joint-specific powers) and by actions of the upper body that transfer power across the hip (8,21,30). Further, joint-specific powers averaged over the extension and flexion phases of the pedal cycle (e.g., knee extension power) can provide a more detailed description of the joint actions used during cycling. Several researchers (8,16,27,30) have previously quantified joint-specific powers during submaximal cycling (SUBcyc, 80-340 W), and their data generally indicate that SUBcyc is performed with a dominant reliance on knee extension actions with smaller contributions from knee flexion and hip extension actions. Conversely, during maximal cycling (MAXcyc, 1080 W), Martin and Brown (21) recently reported that hip extension power is twice as great as knee extension power and that knee flexion power is similar to knee extension power. Taken together, these results suggest a decreased reliance on knee extension power and an increased reliance on knee flexion and hip extension powers when transitioning from SUBcyc to MAXcyc. However, joint-specific powers produced at higher net cycling power outputs that are above physiological steady state but below maximum muscular power (e.g., 350-1000 W) have not been reported.
In addition to the increased reliance on knee flexion and hip extension powers, Martin and Brown (21) reported that cyclists exploited redundant degrees of freedom during the cycling action to perform ankle, knee, and hip joint extension actions for more than half of the pedal cycle. The ratio of the time for extension to the time for flexion is defined as duty cycle and a value of 1.0 represents equal time for extension and flexion. The increased duty cycle values (duty cycle > 1.0) reported by Martin and Brown (21) allowed the powerful ankle, knee, and hip joint extension actions to act for a greater portion of the pedal cycle and thus to increase net cycling power (pedal power) averaged over complete pedal cycles. Further, increased duty cycle values may be an important strategy for increasing net cycling power because duty cycle values >1.0 have been reported to increase muscular power during isolated muscle actions (4), animal locomotion (6,7,19,20), and single-leg cycling (23). To our knowledge, duty cycle values have not been reported for SUBcyc.
Understanding the changes in relative joint-specific powers and duty cycle values as a function of net cycling power may be useful for researchers as well as clinicians, coaches, and athletes. Specifically, researchers who use cycling protocols to elicit perturbations to physiological homeostasis (e.g., V˙O2max test) may need to know whether the relative contributions from ankle, knee, and hip joint actions change when net cycling power increases. Such changes, if they do occur, could influence the physiological responses separate from net cycling per se (18). In addition to laboratory-based cycling protocols, competitive cycling is inherently a non-steady-state activity performed with intermittent high net cycling power (e.g., 350-1000 W) (26). Insight into how these higher net cycling powers are produced may be useful when evaluating elite-level performance and implementing training and rehabilitation programs. Although it may seem reasonable to make conclusions based on SUBcyc and MAXcyc joint-specific power data from previous investigations, such a comparison is limited in two ways. First, pedaling rate in these previous investigations differed: 40-110 rpm for SUBcyc (8,16,27,30) and 120 rpm for MAXcyc (21). Consequently, different pedaling rates used across these investigations may be responsible for some of the observed differences. Second, these previous data were collected from different populations ranging from elite junior cyclists to recreationally active cyclists, and these groups may have had different metabolic and muscular capacities that influenced their cycling biomechanics.
To our knowledge, no previous investigators have quantified the progression of joint-specific powers produced over a broad range of net cycling powers (e.g., 250, 400, 550, 700, and 850 W and MAXcyc) by one group of individuals while using a constant pedaling rate. Therefore, our primary purpose for conducting this investigation was to determine the extent to which ankle, knee, and hip joint actions produced power over a range of net cycling powers. On the basis of previous reports, we hypothesized that relative knee extension power would decrease and relative knee flexion and hip extension powers would increase as net cycling power increased. A secondary purpose of this study was to quantify whole-leg and ankle, knee, and hip joint duty cycle values over a range of net cycling powers. We hypothesized that duty cycle values would increase as net cycling power increased.
Experimental procedures used in this investigation were reviewed and approved by the University of Utah Institutional Review Board. The protocol and procedures were explained verbally, and all participants provided written informed consent before testing. Eleven trained male cyclists aged 19-44 yr (mass = 75 ± 7 kg, height = 1.78 ± 0.07 m) performed two familiarization sessions that consisted of at least two SUBcyc trials (5 min, 240 W at 90 rpm) and four MAXcyc trials (3 s) on an isokinetic cycle ergometer (described below). We adjusted the ergometer seat height to match each participant's accustomed cycling position, and participants wore cleated cycling shoes that locked onto the pedal surface (Speedplay, Inc., San Diego, CA).
On the experimental day, each participant reported to the laboratory where body mass, height, thigh length (greater trochanter to lateral femoral condyle), shank length (lateral femoral condyle to lateral malleolus), foot length (heel to toe), and kinematic foot length (pedal spindle to lateral malleolus) were recorded. After a 5-min cycling warm-up (100-150 W, 90 rpm), participants performed SUBcyc trials at 250, 400, 550, 700, and 850 W. All SUBcyc trials were performed on an isokinetic cycle ergometer (described below) at 90 rpm, and participants remained seated throughout each trial. A power meter (Schoberer Rad Messtechnik, Jülich, Germany) displayed the instantaneous power (sampled at 10 Hz) that the participant was producing, which allowed the individual to cycle at the prescribed net cycling target power. During SUBcyc, participants were given time to stabilize at their target power and then maintained that power for an additional 3 s while data were collected. Participants also performed two seated 3-s MAXcyc trials at 90 and 120 rpm. The order of cycling trials was presented using a Latin square design, and participants rested for 2 min between trials.
A Monark (Vansbro, Sweden) cycle ergometer frame and flywheel were used to construct the isokinetic ergometer. The flywheel was driven in the forward direction by a 3750-W direct-current motor (Baldor Electric Co. model CDP3605; Fort Smith, AR) via pulleys and a belt. The motor was controlled by a speed controller equipped with regenerative braking capability (Minarik model RG5500U; Glendale, CA). The right pedal (for a picture, see Elmer et al. (14) and Elmer and Martin (15)) of the isokinetic ergometer was equipped with two three-component piezoelectric force transducers (Kistler 9251; Kistler USA, Amherst, NY), and the right pedal and crank were equipped with digital position encoders (US Digital model S5S-1024; Vancouver, WA).
Kinematic and kinetic data.
Two-dimensional kinematic and kinetic data were obtained using the methods described by Martin and Brown (21) and Martin et al. (22). Briefly, pedal forces, pedal and crank positions, and the position of an instrumented spatial linkage system (ISL, described below) were recorded at 240 Hz using Bioware software 3.0 (Kistler USA). Normal and tangential pedal forces, pedal position, crank position, and ISL position data were filtered using a fourth-order zero-lag low-pass Butterworth filter with a cutoff frequency of 8 Hz. Pedal power was calculated as the dot product of pedal force and linear pedal velocity. Positions of the right greater trochanter and iliac crest were determined by collecting a static trial of each participant attached to the ISL, and the relative position was assumed to remain constant (25). During the cycling protocols, iliac crest and pedal and crank position coordinates were recorded, which allowed sagittal plane limb segment positions to be determined geometrically. Linear and angular velocities and accelerations of the limb segments were determined by finite differentiation of position data. Segmental masses, moments of inertia, and location of centers of mass were estimated using the regression equations reported by de Leva (10). Sagittal plane joint reaction forces and net joint moments at the ankle, knee, and hip were determined by using inverse dynamics techniques (13). Ankle, knee, and hip joint-specific powers were calculated as the product of net joint moments and joint angular velocities. Power transferred across the hip joint was calculated as the dot product of the hip joint reaction force and linear velocity. Joint-specific powers were averaged over all of the complete pedal cycles within the 3-s measurement interval. In addition, joint-specific powers were averaged over the extension and flexion phases, which were defined by joint angular velocity directions (21). Negative joint-specific powers (i.e., power absorption) occurred when the net joint moment and joint angular velocity terms were in opposite directions. Joint-specific powers for each action (e.g., knee extension power) were normalized to the mean pedal power for complete pedal cycles to facilitate comparisons across SUBcyc and MAXcyc trials. Because most power is produced during the extension phase, power values averaged over the extension phase can be larger than those averaged over complete pedal cycles, and consequently, the sum of these relative joint-specific power values can exceed 100%. Duty cycle values for the whole leg were based on the magnitude of the position vector from the hip joint to the pedal spindle, with extension defined as an increasing magnitude and flexion as a decreasing magnitude (21). Ankle, knee, and hip joint duty cycle values were calculated as the ratio of the time for extension to the time for flexion (21). Note that joint duty cycle values can sometimes be <1.0, whereas whole-leg duty cycle values are close to or >1.0 because of linear movement of the hip joint center.
To test our hypothesis that relative knee extension power would decrease and knee flexion and hip extension power would increase as net cycling power increased, we performed linear regression analyses on 1) mean relative joint-specific powers across mean net cycling powers and 2) individual relative joint-specific powers across individual net cycling powers. Note that 120-rpm MAXcyc was not included in the regression analyses because paired t-tests were used to compare pedal and joint-specific powers between 90- and 120-rpm MAXcyc. A multivariate ANOVA (MANOVA) with repeated measures was performed with net cycling power as the independent variable and whole-leg and ankle, knee, and hip joint duty cycles as dependent variables. If the MANOVA revealed a significant effect of net cycling power on duty cycle values, then additional one-way repeated-measures ANOVA procedures were performed on each dependent variable. If the ANOVA was significant, then subsequent post hoc analyses (Fischer least significant difference) were used to identify which net cycling powers were performed with different duty cycles values. Crank angles associated with flexion and extension phases for the whole leg and ankle, knee, and hip joints are reported for descriptive purposes. All data are presented as mean ± SE, and α was set to 0.05.
Mean powers delivered to the right pedal were approximately one-half (116 ± 4, 200 ± 4, 271 ± 5, 351 ± 5, and 415 ± 5 W) of the prescribed net cycling target powers (250, 400, 550, 700, and 850 W, respectively) for SUBcyc trials, suggesting that total power from both legs was close to the target power (Table 1). Pedal power increased during the extension and flexion phases as net cycling power increased (Table 1, Fig. 1). Ankle joint-specific power increased primarily during the extension phase as power was negative (power absorbed) during the flexion phase (Table 1, Fig. 1). Knee joint-specific power increased during both the extension and flexion phases as net cycling power increased (Table 1, Fig. 1). Hip joint-specific and hip transfer powers increased mostly during the extension phase (Table 1, Fig. 1).
Linear regression analyses indicated that relative knee extension power decreased, whereas relative knee flexion power increased as net cycling power increased (r2 = 0.88, F1,5 = 28.94, P = 0.01; r2 = 0.98, F1,5 = 181.19, P < 0.001; for group mean values, respectively; Table 1, Figs. 2 and 3). The mean r2 values across all participants for relative knee extension and knee flexion powers were 0.59 ± 0.09 and 0.63 ± 0.12, respectively. Relative hip extension power did not change as net cycling power increased. Mean pedal power increased by 9% between the 90- and 120-rpm MAXcyc trials (516 ± 23 vs 563 ± 19 W, t10 = −5.62, P < 0.01; Table 1). Hip extension power increased by 19% between the 90- and 120-rpm MAXcyc trials (356 ± 21 vs 423 ± 24 W, t10 = −4.90, P < 0.01; Table 1), whereas knee extension and knee flexion powers did not differ.
The MANOVA revealed a significant effect of net cycling power on duty cycle values (Wilks λ = 0.20, F24,200 = 4.95, P < 0.01). The follow-up ANOVA procedures revealed that the effect of net cycling power on whole-leg and ankle, knee, and hip joint duty cycle values was significant (F6,60 = 11.92, P < 0.01; F6,60 = 7.55, P < 0.01; F6,60 = 8.39, P < 0.01; F6,60 = 12.11, P < 0.01; respectively). Post hoc t-tests indicated that whole-leg and knee and hip joint duty cycle values during 250-W SUBcyc differed from those for MAXcyc (P < 0.01, Table 2). Ankle joint duty cycle values during 250-W SUBcyc differed from those during 550-, 700-, and 850-W SUBcyc and MAXcyc (P < 0.05, Table 2). Crank angles associated with flexion and extension phases for the whole leg and for the knee and hip joints were generally similar during SUBcyc but seemed to differ for MAXcyc (Fig. 4). Ankle plantarflexion phases seemed to increase with increased net cycling power (250 W to MAXcyc, Fig. 4). Compared with SUBcyc, extension phases for the whole leg and ankle, knee, and hip joints during MAXcyc generally started earlier in the pedal cycle and continued later into the cycle (Fig. 4).
The primary purpose of this investigation was to test the hypothesis that relative knee extension power would decrease and knee flexion and hip extension powers would increase as net cycling power increased. Our data indicated that the cyclists in this investigation used relatively less knee extension and more knee flexion power as net cycling power increased. In contrast, relative hip extension power did not change with increased net cycling power. This may be partially due to the fact that hip extension power was the most powerful action across all net cycling powers, which differs from previous reports of SUBcyc biomechanics. These data partially support our hypothesis and demonstrate that knee joint actions used to produce power during SUBcyc are relatively different from those joint actions used during MAXcyc. An additional finding was that cyclists spent more time in the extension phase (increased duty cycle) during MAXcyc suggesting that increased duty cycle values likely serve as a means to increase maximum power production.
Absolute joint-specific power data indicated that knee extension and flexion powers substantially increased as a function of increased net cycling power. However, relative knee extension power decreased, whereas relative knee flexion power increased with increased net cycling power. Thus, these data suggest that knee flexor function becomes relatively more important during high-intensity cycling. Relative knee extension and flexion powers produced during 250-W SUBcyc (57% and 18% of the power delivered to the right pedal, respectively) are in general agreement with those powers reported by Ericson (16) during 200-W SUBcyc (∼72% and 24% of the power delivered to one pedal, respectively). Similarly, relative knee extension and flexion powers produced during 90- and 120-rpm MAXcyc (40%-41% and 38%-40% of the power delivered to the right pedal, respectively) agree with a recent work by Martin and Brown (21) (40% and 37% of the power delivered to the right pedal). Together, our results support, expand, and unify previous SUBcyc and MAXcyc findings. We should also add that when we performed linear regression analyses on individual relative joint-specific powers across individual net cycling powers, relative knee extension and flexion power production was more variable as demonstrated by the smaller mean of the individual r2 values (0.59 and 0.63 for relative knee extension and flexion powers, respectively). Thus, on average, this group of cyclists used relatively less knee extension and more knee flexion power as net cycling power increased, but individual differences should be considered as well.
Cyclists in our study used relatively more knee extension and less knee flexion power at lower net cycling powers (e.g., 250 W). Interestingly, Korff et al. (18) reported that gross efficiency decreased when cyclists were instructed to actively "pull up" during leg flexion. Thus, cyclists may intuitively rely more on knee extension actions during SUBcyc to maximize efficiency. During MAXcyc, where metabolic efficiency is not the limiting factor, increased knee flexion power likely increases the overall power delivered to the environment. In addition to increased knee flexion power during MAXcyc, there was a substantial increase in hip transfer power (absolute and relative values) demonstrating that participants relied heavily on actions of the torso and upper body to transfer power across the pelvis to the leg. Increased hip transfer power and torso movement would likely be counterproductive during SUBcyc because McDaniel et al. (24) demonstrated that excess torso movement during SUBcyc reduces metabolic efficiency. Taken together, these results suggest a trade-off between maximizing efficiency and power production (29) during SUBcyc and MAXcyc, respectively. Investigators who use cycling to elicit perturbations to physiological homeostasis should consider that when net cycling power increases, the relative contribution from knee extension decreases, whereas the contribution from knee flexion and torso actions increases. Consequently, the physiological responses may vary because of the relative reliance on different muscle groups (e.g., quadriceps vs hamstrings). In addition, athletes who perform cycling tasks above physiological steady state (e.g., intermittent high-intensity cycling) may find it beneficial to incorporate specific training of knee flexor muscles into their training program. Alternatively, intermittent high-intensity cycling may provide a useful rehabilitation modality for hamstring muscle issues.
Although absolute hip extension power increased by fourfold across the range of net cycling powers tested and was the most powerful action, relative hip extension power did not change with increased net cycling power. Thus, hip extension is an important action during both SUBcyc and MAXcyc. Relative hip extension power produced during 250-W SUBcyc (81% of the power delivered to the right pedal) contrasts with those values reported by Ericson (16) during 200-W SUBcyc (∼50% of the power delivered to one pedal). Conversely, relative hip extension powers produced during 90- and 120-rpm MAXcyc (70%-76% of the power delivered to the right pedal, respectively) agree with the recent work by Martin and Brown (21) (78% of the power delivered to the right pedal). Our results may have implications for researchers, clinicians, and athletes. Researchers often use cycling exercise to induce fatigue (e.g., time trial) and subsequently quantify fatigue via pre- to postexercise changes in knee extensor muscle function (e.g., force-generating capacity) (1-3). Because hip extension is a major power-producing action during SUBcyc, it may be just as important to also consider pre- to postexercise changes in hip extensor muscle function when investigating the neuromuscular mechanisms associated with fatigue. In clinical settings, cycling may serve as a unique exercise for rehabilitating hip extensor function while simultaneously eliciting cardiovascular benefits. Finally, cyclists could possibly benefit from incorporating hip extensor training into training programs.
Our results relating to hip extension power were unanticipated because previous findings from separate groups (16,21) suggest an increased reliance on hip extension power when transitioning from SUBcyc to MAXcyc. These differences may reflect individual variations and/or methodological issues. Coyle (9) suggested that more experienced cyclists distribute work across more muscular actions to reduce localized stress. Thus, participants in our study may have been more adept at this aspect of cycling technique than those in previous studies (e.g., trained cyclists vs recreationally active cyclists). Alternatively, our choice of isokinetic cycle ergometry in which participants' modulated power with feedback from a power meter may have influenced pedaling technique compared with isotonic or isopower ergometry (16) in which the individual modulates pedaling rate. Both possible explanations deserve further study, and we are currently planning to determine the effects of cycling experience as well as feedback and ergometry methods on joint-specific power. Interestingly, when pedaling rate was increased from 90 to 120 rpm during MAXcyc, hip extension power was significantly greater, whereas knee extension and flexion powers did not differ. We have previously observed that maximum hip extension power occurs at a higher pedaling rate than knee extension and flexion powers (unpublished observation). Thus, it is possible that hip extensors have a unique joint power-velocity relationship. Individual muscle architectural and morphological characteristics should be considered as well. The extent to which these factors influence joint-specific power production requires further investigation, and these topics are an area of interest for our laboratory that may have an application in clinical and sport performance settings.
Duty cycle values for the whole leg and for the knee and hip joints were generally similar during SUBcyc but increased during MAXcyc. In contrast, ankle joint duty cycle values increased across the range of SUBcyc powers (from 0.77 at 250 W to 1.20 at 850 W) and increased further during MAXcyc (1.25). These results emphasize that cycling is not a one-degree-of-freedom action and demonstrate the effects of motor control on the cycling action. Our findings support works by several previous groups that have observed duty cycle values >1.0 during locomotor tasks including pigeon flight (7), mallard swimming (6), scallop swimming (19,20), and human cycling (21,23). Together, these findings indicate that increasing duty cycle values represent an important strategy for maximizing muscular power during a variety of voluntary activities. In addition to providing insight into basic aspects of control of multijoint actions, these results may be useful for applied sport scientists and cyclists. That is, our finding that cyclists naturally take advantage of prolonged extension periods primarily during MAXcyc may partially explain why commercially available crank sets (e.g., Rotor Pedaling System) that alter the pedal trajectory (i.e., duty cycle values) have been effective in improving MAXcyc but not SUBcyc performance (28).
Greater power production associated with increased duty cycle likely occurred because the extension phases for the whole leg and ankle, knee, and hip joints generally started earlier in the pedal cycle and continued later into the cycle. Further, concomitant effects of increased muscle excitation and reduced muscle shortening velocity may have also contributed to greater power production (4,23). For example, Martin et al. (23) reported that increased power production with increased duty cycle values was associated with an increased time and reduced pedal speed (marker for muscle shortening velocity) for the whole-leg extension phase. Finally, alterations in duty cycle values may also reflect the differences in the task goal of producing and controlling a targeted power during SUBcyc as opposed to maximizing power during MAXcyc.
In this investigation, we used inverse dynamics techniques to quantify joint-specific power production during SUBcyc and MAXcyc. Although these methods are often used to describe the biomechanics associated with producing muscular power (5,11,12,16,17), they cannot be used to specifically quantify individual muscle forces (32). The use of forward dynamic simulation techniques, which allow for the quantification of individual muscle forces, may offer additional insight into muscular function associated with SUBcyc and MAXcyc. Similarly, electromyography measures could identify which muscles (e.g., vastus lateralis) are actively recruited during knee extension and flexion and hip extension actions.
In summary, cyclists in this investigation used relatively less knee extension and more knee flexion power (on average) as net cycling power increased. Hip extension was the most powerful action across all net cycling powers, but its relative contribution did not change with increased net cycling power. Duty cycle values increased significantly during MAXcyc suggesting that the time spent during extension is important for maximizing power. These are the first data to document joint-specific power production across such a broad range of net cycling powers and to highlight distinct differences between SUBcyc and MAXcyc. Finally, these results may allow researchers, clinicians, and coaches and athletes to take an even greater advantage of cycling as a research model and as a training and rehabilitation modality.
This work was supported by funding from the Engineering and Physical Sciences Research Council Doctoral Training Grant Scheme.
The authors thank the participants who took part in this study for their enthusiastic efforts in performing the cycling trials and the undergraduate students in the Neuromuscular Function Laboratory for their assistance with the data collection.
The results of the present investigation do not constitute endorsement by the American College of Sports Medicine.
1. Amann M, Dempsey JA. Locomotor muscle fatigue modifies central motor drive in healthy humans and imposes a limitation to exercise performance. J Physiol
2. Amann M, Eldridge MW, Lovering AT, Stickland MK, Pegelow DF, Dempsey JA. Arterial oxygenation influences central motor output and exercise performance via effects on peripheral locomotor muscle fatigue in humans. J Physiol
. 2006;575(Pt 3):937-52.
3. Amann M, Proctor LT, Sebranek JJ, Pegelow DF, Dempsey JA. Opioid-mediated muscle afferents inhibit central motor drive and limit peripheral muscle fatigue development in humans. J Physiol
. 2009;587(Pt 1):271-83.
4. Askew GN, Marsh RL. The effects of length trajectory on the mechanical power output of mouse skeletal muscles. J Exp Biol
. 1997;200(Pt 24):3119-31.
5. Bezodis IN, Kerwin DG, Salo AI. Lower-limb mechanics during the support phase of maximum-velocity sprint running. Med Sci Sports Exerc
6. Biewener AA, Corning WR. Dynamics of mallard (Anas platyrynchos
) gastrocnemius function during swimming versus terrestrial locomotion. J Exp Biol
. 2001;204(Pt 10):1745-56.
7. Biewener AA, Corning WR, Tobalske BW. In vivo
pectoralis muscle force-length behavior during level flight in pigeons (Columba livia
). J Exp Biol
. 1998;201(Pt 24):3293-307.
8. Broker JP, Gregor RJ. Mechanical energy management in cycling: source relations and energy expenditure. Med Sci Sports Exerc
9. Coyle EF. Integration of the physiological factors determining endurance performance ability. Exerc Sport Sci Rev
10. de Leva P. Adjustments to Zatsiorsky-Seluyanov's segment inertia parameters. J Biomech
11. DeVita P, Helseth J, Hortobagyi T. Muscles do more positive than negative work in human locomotion. J Exp Biol
. 2007;210(Pt 19):3361-73.
12. Dutto DJ, Hoyt DF, Clayton HM, Cogger EA, Wickler SJ. Joint work and power for both the forelimb and hindlimb during trotting in the horse. J Exp Biol
. 2006;209(Pt 20):3990-9.
13. Elftman H. Forces and energy changes in the leg during walking. Am J Physiol
14. Elmer SJ, Madigan ML, LaStayo PC, Martin JC. Joint-specific power absorption during eccentric cycling. Clin Biomech (Bristol, Avon)
15. Elmer SJ, Martin JC. Joint-specific power loss after eccentric exercise. Med Sci Sports Exerc
16. Ericson MO. Mechanical muscular power output and work during ergometer cycling at different work loads and speeds. Eur J Appl Physiol Occup Physiol
17. Korff T, Hunter EL, Martin JC. Muscular and non-muscular contributions to maximum power cycling in children and adults: implications for developmental motor control. J Exp Biol
. 2009;212(Pt 5):599-603.
18. Korff T, Romer LM, Mayhew I, Martin JC. Effect of pedaling technique on mechanical effectiveness and efficiency in cyclists. Med Sci Sports Exerc
19. Marsh RL, Olson JM. Power output of scallop adductor muscle during contractions replicating the in vivo
mechanical cycle. J Exp Biol
20. Marsh RL, Olson JM, Guzik SK. Mechanical performance of scallop adductor muscle during swimming. Nature
21. Martin JC, Brown NA. Joint-specific power production and fatigue during maximal cycling. J Biomech
22. Martin JC, Elmer SJ, Horscroft RD, Brown NA, Schultz BB. A low-cost instrumented spatial linkage accurately determines ASIS position during cycle ergometry. J Appl Biomech
23. Martin JC, Lamb SM, Brown NA. Pedal trajectory alters maximal single-leg cycling power. Med Sci Sports Exerc
24. McDaniel J, Subudhi A, Martin JC. Torso stabilization reduces the metabolic cost of producing cycling power. Can J Appl Physiol
25. Neptune RR, Hull ML. Accuracy assessment of methods for determining hip movement in seated cycling. J Biomech
26. Quod MJ, Martin DT, Martin JC, Laursen PB. The power profile predicts road cycling MMP. Int J Sports Med
27. Reiser RF 2nd, Peterson ML, Broker JP. Understanding recumbent cycling: instrumentation design and biomechanical analysis. Biomed Sci Instrum
28. Rodríguez-Marroyo JA, García-López J, Chamari K, Córdova A, Hue O, Villa JG. The rotor pedaling system improves anaerobic but not aerobic cycling performance in professional cyclists. Eur J Appl Physiol
29. Rome LC, Lindstedt SL. Mechanical and metabolic adaptation of the muscular system. In: Dantzler WH, editor. Handbook of Physiology, Section 1: II Comparative Phys
. Bethseda (MD): American Physiological Society; 1997. p. 1587-652.
30. van Ingen Schenau GJ, van Woensel WW, Boots PJ, Snackers RW, de Groot G. Determination and interpretation of mechanical power in human movement: application to ergometer cycling. Eur J Appl Physiol Occup Physiol
31. Wilmore JH, Costill DL. Physiology of Sport and Exercise
. 4th ed. Champaign (IL): Human Kinetics; 1994. p. 12-3.
32. Zajac FE, Neptune RR, Kautz SA. Biomechanics and muscle coordination of human walking. Part I: introduction to concepts, power transfer, dynamics and simulations. Gait Posture