Date of Award


Document Type



Edith Cowan University

Degree Name

Doctor of Philosophy


School of Exercise and Health Sciences


Faculty of Health, Engineering and Science

First Supervisor

Associate Professor Anthony J. Blazevich

Second Supervisor

Dr Chris R. Abbiss


Neuromuscular fatigue is an inevitable process at play during prolonged exercise, and may be caused by multiple alterations within the central nervous system and peripheral musculature. As fatigue develops, the neuromuscular system must adapt to these changes by making compensatory movement pattern adjustments so as to use motor pathways that are less fatigued in an effort to maintain task performance; motor variability is thus increased. The primary purpose of the four studies contained within this doctoral thesis was to detail the progression of exercise-induced neuromuscular fatigue, and to improve our understanding of the muscle activation and joint kinematic alterations that occur as fatigue accumulates. Within this context, cycling was used as the exercise model, and the relationship between physiological and biomechanical aspects of high-intensity, moderate duration (<10 >min) cycling were specifically examined. The first two studies of this thesis were aimed at understanding the progression of neuromuscular fatigue as well as the associated motor control and biomechanical (i.e. muscle activation and kinematic) changes that occur during exhaustive cycling. Specifically, the time course and relative contributions of central and peripheral fatigue mechanisms, and the associated changes in muscle activation and both lower (i.e. hip, knee and ankle joint) and upper (i.e. trunk) limb kinematics were examined during a high-intensity cycling time to exhaustion (TTE) test. This was performed at 90% maximal aerobic power (Pmax) with nine well-trained cyclists. Temporal relationships between joint kinematics and changes in markers of central and peripheral fatigue were also examined. Peripheral fatigue (i.e. impaired contractile function: reduced peak twitch torque, −39.9%; twitch contraction time, −10.7%; and the average rates of twitch torque development −34.7% and relaxation −36.7% at task failure i.e., T100) developed early in the exercise bout from 60% of the time to task failure (p < 0.05). However, a central facilitation, measured as an increase in peak vastus medialis (38.9%) and gluteus maximus electromyogram (87.2%) amplitudes at T100, rather than central fatigue, occurred towards the end of the exercise task (p < 0.05). Thus, neuromuscular fatigue development was associated with an increase in the magnitude of lower limb muscle activity, which may have represented an attempt to increase muscle force to maintain the required power output of the cycling task. Increases in trunk flexion were observed from 60% of the time to task failure (p < 0.05), and were therefore notable at or after the point of significant peripheral fatigue. Conversely, increases in trunk medio-lateral sway (lateral flexion), hip abduction/adduction and knee valgus/varus were observed only from 80% of the time to task failure (p < 0.05), which paralleled the increase in central motor drive. The results of this study therefore indicate that significant trunk kinematic changes in the sagittal plane occurred at or after the point of significant peripheral fatigue development, whereas, significant changes at the trunk, hip and knee joints in the coronal plane occurred later in the exercise task and paralleled the facilitation of central motor drive during the cycling task. In the third study, the effects of real-time, kinematic feedback provision for trunk flexion (TTETflex), trunk medio-lateral sway (TTETsway) and hip abduction/adduction (TTEHabd/add) during a high-intensity TTE cycling test (90% Pmax) in nine well-trained cyclists were examined. The times taken to reach task failure were compared to a TTE test completed with no feedback. The times taken to reach task failure were not significantly different when provided with trunk flexion (TTETflex) and hip abduction/adduction (TTEHabd/add) feedback compared to the non-feedback condition (p > 0.05). There was, however, a significant decrease in the time to task failure during the TTETsway test (p < 0.05). Not all participants could maintain trunk and/or hip movement within a set movement pattern criteria; and three participants were therefore excluded from the kinematic analyses for both the TTETflex and TTETsway tests (n = 6) as were two participants from the TTEHabd/add test (n = 7). For participants who correctly used the kinematic feedback, no differences in the times taken to reach failure were observed in between the feedback (TTETflex, TTETsway and TTEHabd/add) and nonfeedback test conditions (p > 0.05). Despite being given feedback, changes in joint kinematics were similar across all test conditions; significant alterations were observed at the trunk and knee joints in the sagittal plane and at the hip and knee joints in the coronal plane (p < 0.05). Given trunk flexion feedback (TTETflex), significant increases in left hip flexion and trunk medio-lateral sway ROM were observed (p < 0.05), whereas given trunk medio-lateral sway feedback (TTETsway), increases in right hip flexion ROM also occurred (p < 0.05). These results indicate that, regardless of whether or not well-trained cyclists are able to control the level of kinematic variability when fatigued, acute exposure to real-time kinematic feedback to limit trunk or hip movement during high-intensity cycling may influence cycling kinematics (i.e. technique) and, in some cases (e.g. trunk medio-lateral sway), may reduce performance. The final study examined the relationship between joint kinematics, measured in non-fatigued and fatigued high-intensity cycling, and the cyclists’ physiological profiles (i.e., physiological attributes indicative of successful cycling ability, including both maximal oxygen consumption and peak power output relative to body mass, maximal heart rate, both power output and heart rate at the first and second ventilatory thresholds and cycling economy at 100 W) and the time taken to reach task failure. Submaximal physiological attributes were correlated with hip (abduction/adduction angle and ROM), knee (flexion angle) and ankle (flexion ROM) kinematics measured in a non-fatigued state at the start of the trial (r > 0.40; p < 0.05). However, both physiological attributes associated with maximal exercise capacity and cycling economy were correlated with trunk (flexion angle) and ankle (flexion angle and ROM) kinematics measured in a fatigued state at the end of the test (r > 0.40; p < 0.05). Trunk flexion and medio-lateral sway ROM in a non-fatigued state, and trunk flexion angle in a fatigued state, were associated with the time to task failure (r > 0.50; p < 0.05). Thus, the degree of trunk flexion and medio-lateral sway may be important kinematic variables that are indicative of cycling performance. These findings reveal an interdependence between cycling kinematics and both the physiological attributes indicative of successful cycling performance and the time taken to reach task failure during high-intensity, constant-load cycling. In conclusion, the findings presented in this thesis indicate that the temporal patterns of central and peripheral neuromuscular fatigue differ (Study 1; Chapter 3). Task failure during high intensity cycling appears to be associated with the development of peripheral fatigue despite the presence of an increase in central motor drive. Subsequent to the development of neuromuscular fatigue, muscle activation and joint kinematic alterations can be observed, which may represent compensatory mechanisms employed by the neuromuscular system to continue task performance (Studies 1 and 2; Chapters 3 and 4). Joint kinematic alterations in the sagittal plane were associated with the development of peripheral fatigue whereas coronal plane adjustments occurred in parallel with central facilitation, and/or when a more substantial level of peripheral fatigue accumulated. Such compensatory kinematic strategies are also associated with an athlete’s physiological attributes and their cycling performance (i.e., time to task failure) (Study 4; Chapter 6). Importantly, imposing specific joint kinematic restrictions (trunk flexion, trunk medio-lateral sway and hip abduction/adduction) during exhaustive cycling, influenced cycling kinematics (i.e. technique) and, in some cases (e.g. trunk medio-lateral sway), reduced the time taken to reach task failure for well-trained cyclists (Study 3; Chapter 5). Such findings enhance our understanding of how the neuromuscular system copes with fatigue development, and should assist coaches and/or occupational health practitioners to better understand the fatigue process and neuromuscular strategies utilised during exercise tasks with similar characteristics to that used in the current studies.