Past Research

Summary of Selected Research Projects:

Research projects may be self contained and terminal or they may be part of a sequence of studies designed to answer fundamental questions. Our ultimate goal is to further our understanding of the interactions between physical activity, human movement performance, fitness and musculoskeletal injury risk. Research conducted to achieve this goal involves animal models if appropriate, computer modeling and movement simulation, instrumentation design, human testing.

Following are examples of some of the research conducted in the laboratory:

Muscle-Tendon Lengthening

rat1Background: Various neuromuscular disorders lead to joint stiffness and impaired joint function. Orthopaedic surgeons often intervene in these cases and surgically lengthen the muscle-tendon unit or units causing the joint stiffness. There are a variety of surgical procedures used to lengthen muscle-tendon units; however, controlled studies of the affects of these procedures on muscle-tendon performance have not been conducted.


Purpose: The purpose of this study was to investigate the use of an animal model to quantify differences in muscle-tendon structure and performance after being lengthened by one of two alternative procedures.

Methods: Ten rat gastrocnemius muscle-tendon (MT) units were lengthened ~3 mm by cutting the aponeurosis of the medial head. An additional 10 MT units were lengthened by performing a Z-cut to the Achilles tendon. A sham operation was performed on the contralateral MT unit of each rat. Following three weeks of recovery, muscle and tendon lengths and the force-ankle angle relationship were quantified. Data were compared between paired MT units within each rat and within each surgical intervention group.

Findings: There were no differences in structure and MT performance between paired comparisons between aponeurotic-lengthened and contralateral control MT units. The Z-lengthened MT units had longer tendons, and force-ankle angle profiles were shifted toward greater dorsi flexion angles compared to control muscles.

Exercise Modality, Phosphorylation of p70s6k, and Skeletal Muscle Hypertrophy

Background: Skeletal muscle strength is essential for human movement performance and quality of life. Understanding the signaling mechanisms responsible for maintaining and increasing muscle strength is fundamental for identifying strategies to combat sarcopenia and to enhance muscle strength and movement performance.

Purpose: The objective of this study was to determine if p70s6k phosphorylation, a ribosomal protein suggested to be essential for up-regulation of exercise-induced muscle protein synthesis, was dependent on the mode of resistance exercise. The hypothesis tested was: resistance exercise results in p70s6k phosphorylation, independent of the mode of resistance exercise (e.g. isometric, eccentric).

Methods: Two groups (N=5 each) of Female Sprague Dawley rats, ~12 weeks old, were tested. Rats were anesthetized and indwelling electrodes used to stimulate the right hind limb muscles via the sciatic nerve. The tibialis anterior (TA) muscle of Group 1 rats were exposed to 3 sets of 10 isometric resistance repetitions while the TA of Group 2 rats were exposed to 3 sets of 10 lengthening resistance repetitions. Contralateral TA muscles served as non-exercised controls. Rats were euthanized 6 hours post exercise, muscles harvested and muscle samples processed to determine the extent of p70s6k phosphorylation.

Findings: A single bout of TA lengthening contractions resulted in significantly (p < 0.05) higher levels of phospho-p70s6k 6 hours post exercise than the controls and isometric contractions. The differences in total p70s6k 6 hours post exercise between exercised and contralateral control were 202% + 47% S.E. and 7% + 4% S.E. for the lengthening and isometric exercise groups respectively. Results suggest that the protein synthesis pathway activated by isometric exercise may differ (i.e. a non-p70s6k activation pathway) than that activated by lengthening exercise.

Musculoskeletal Modeling & Computer Simulation Techniques

Are Fixed Limb Inertial Models Valid for Dynamic Simulations of Human Movement?

DigitalLegModelBackground: During human movement, muscle activation and limb movement result in subtle changes in muscle mass distribution. Muscle mass redistribution can affect limb inertial properties and limb dynamics, but it is not currently known to what extent. The objectives of this study were to investigate: (1) how physiological alterations of muscle and tendon length affect limb inertial characteristics, and (2) how such changes affect dynamic simulations of human movement.

Methods: A digital model of a human leg, custom software, and Software for Interactive Musculoskeletal Modeling were used to simulate mass redistribution of muscle-tendon structures within a limb segment during muscle activation and joint movement. Thigh and shank center of mass and moments of inertia for different muscle activation and joint configurations were determined and compared. Limb inertial parameters representing relaxed muscles and fully active muscles were input into a simulated straight-leg movement to evaluate the effect inertial parameter variations could have on movement simulation results.

Findings: Muscle activation and limb movement altered limb segment center of mass and moments of inertia by less than 0.04 cm and 1.2% respectively. These variations in limb inertial properties resulted in less than 0.01% change in maximum angular velocity for a simulated straight-leg hip flexion task. These data demonstrate that, for the digital human leg model considered, assuming static quantities for segment center of masses and moments of inertia in movement simulations appears reasonable and induces minimal errors in simulated movement dynamics.

The Effects of Changing Bone and Muscle Size on Limb Inertial Properties and Limb Dynamics: A Computer Simulation

DefaultDLMBackground: The magnitude and distribution of bone and muscle mass within limbs affect limb inertial properties, maximum movement speed, and the energy required to maintain submaximal movements.

Methods: Musculoskeletal modeling and movement simulations were used to determine how changes in bone and muscle cross-sectional area (mass) affect human thigh and shank inertial properties, the maximum speed of unloaded single-leg cycling and the energy required to sustain submaximal single-leg cycling.

Findings: Depending on initial conditions, shank moments of inertia increased 61-72 kg-cm2 per kg added bone and 72-100 kg-cm2 per kg added muscle. Thigh moments of inertia increased 46-63 kg-cm2 per kg bone and 180-225 kg-cm2 per kg muscle. Maximum unloaded cycling velocity increased with increased muscle mass (~ 2.2-2.9 rpm/kg muscle), but decreased with increased cortical bone mass (~ 2.0-2.8 rpm/kg bone). The internal work associated with unloaded submaximal cycling increased with increased muscle mass (~0.42-0.48 J/kg muscle) and bone mass (~ 0.18-0.22 J/kg bone).

An Investigation of the Interactions Between Lower Limb Bone Morphology, Limb Inertial Properties and Limb Dynamics

Background: Bone mass and size clearly affect the safety and survival of wild animals as well as human beings, however, little is known about the interactions between bone size and movement dynamics.

Methods: A modeling approach was used to investigate the hypothesis that increased bone cortical area causes increased limb moments of inertia, decreased lower limb movement maximum velocities, and increased energy requirements to sustain submaximum lower limb locomotion movements. Custom software and digital data of a human leg were used to simulate femur, tibia, and fibula cortical bone area increases of 0%, 22%, 50%, and 80%. Limb segment masses, center of mass locations, and moments of inertia in the sagittal plane were calculated for each bone condition. Movement simulations of unloaded running and cycling motions were performed. Linear regression analyses were used to determine the magnitude of the effect cortical area has on limb moment of inertia, velocity, and the internal work required to move the limbs at a given velocity.

Findings: The thigh and shank moment of inertia increased linearly up to 1.5% and 6.9%, respectively for an 80% increase in cortical area resulting in 1.3% and 2.0% decreases in maximum unloaded cycling and running velocities, respectively, and in 3.0% and 2.9% increases in internal work for the cycling and running motions, respectively. These results support the hypothesis and though small changes in movement speed and energy demands were observed, such changes may have played an important role in animal survival as bones evolved and became less robust.

Athletic Performance

Rowing – Dry Land Training with Real-Time Feedback

Background: A dry-land rowing system was developed to provide the coach and/or athlete with quantitative information about the athlete’s kinetics and kinematics while the athlete trains.

Methods: Two iterations of this system were developed using a Concept II rowing ergometer. The first system was instrumented with a force transducer and potentiometer, four electrogoniometers attached to the athlete’s ankle, knee, hip, and elbow, and a data acquisition computer. The force transducer is used to quantify the athlete’s pulling force. The potentiometer signal is used to locate the position of the handle. The electrogoniometers provide signals proportional to joint angles. A link segment model of the human body is used to locate joint centers based on limb lengths and joint angles. The computer is used to collect and process all the transducer signals, perform the link segment calculations, and provide feedback to the coach or athlete in the form of a stick figure animation overlaid with kinematic and kinetic information.

Row2A second iteration of the dry-land training rowing system was developed to improve the hardware and software interfaces for the user. The system consists of a Concept II rowing ergometer instrumented with a load cell and a series of potentiometers, a data acquisition computer, and custom software. Kinematic and kinetic rowing data are displayed in the form of a two-dimensional stick figure animation overlaid with kinematic and kinetic profiles. The software allows data to be saved and later replayed.

Findings: This system allows the coach and athlete to quickly review rowing mechanics, to evaluate the effects that technique changes have on the power produced by the athlete, and to identify how technique changes with fatigue.

Rowing – On-Water Performance

Purpose: It was hypothesized that a crew’s rowing performance was predictable based on their total propulsive power, synchrony (a real-time comparison of rower propulsive force magnitudes) and total drag contribution (a measure of the rowers’ effect on shell drag forces during the recovery), quantities calculated from individual rower’s force-time profiles and recovery kinematics.

Methods: A rowing pair was equipped with transducers to gather shell velocity, propulsive blade force, oar angular position and seat displacement. Eight subjects (4 port, 4 starboard) participated in two rounds of data collection. The first round pairings were random, while the second round pairings were assigned based on Round 1 results. Regression analysis and ANCOVA were used to test the validity of assumptions inherent in the predictive model and, if applicable, explore a linear model predicting rowing performance based on total propulsive power, synchrony and total drag contribution.

Findings: Total propulsive power, synchrony and total drag contribution were correlated and further were affected by pairing, violating assumptions inherent in the linear model. The original hypothesis was not supported based on these violations. Important findings include (1) performance cannot be predicted using the simple linear model proposed, (2) rowers’ force-time profiles are repeatable between trials, with some but not all rowers adapting their force-time profile dependent on their pair partner, presumably in an effort to increase the level of synchrony between the two, and 3) subtle biomechanical factors may play a critical role in performance.

Soccer Throw-In

throwexpPurpose: The objective of this study was to analyze the mechanics involved in executing a standard soccer throw-in to determine the anthropometric, strength, and coordination factors that contribute most to a long throw. Two hypotheses were tested: 1) athletes producing longer throws are characterized by having longer segment lengths (torso, upper arm, forearm) and/or joint strengths (hip, shoulder, and elbow), and 2) long throwers utilize the same limb kinematic and kinetic sequencing patterns as short throwers.

Methods: Twenty collegiate level soccer players participated in this study. Their maximum throw distance was recorded from three trials. Segment lengths, and back, shoulder, and elbow maximum isometric strengths were quantified. The correlations between throw distance and an athlete’s segment lengths and joint strengths were determined. Three-dimensional upper extremity kinematics and joint kinetics were quantified for six athletes (3 long and 3 short throwers).

Findings: Correlation coefficients between throw distance and segment lengths were all less than 0.45. Correlation coefficients between throw distance and hip, shoulder, and elbow strength were 0.55, 0.70, and 0.65, respectively. Kinematic and kinetic profiles of long and short throwers were not clearly different. Kinematic and kinetic profiles for long throwers were not always consistent with theoretical predictions. Results suggest that 1) joint strength but not segment lengths correlate with throw performance, and 2) most athletes can improve their throw distance by altering their movement strategy.

Backpack Load Carriage

BackpackBackground: Backpacks provide an effective means of load carriage and hip belts are an important design feature of many backpacks, serving to reduce the loads carried by the shoulders. Custom molded hip belts are a recent design innovation intended to distribute loads over a greater contact area, thus improving backpack user comfort.

Purpose: The purpose of this study was to evaluate the efficacy of custom molded hip belts for reducing skin pressure during backpack usage.

Methods: Pressure levels developed under a standard and custom molded hip belt were quantified while subjects toted a mountaineering pack under controlled gait conditions. Six subjects (3 male, 3 female) stood, walked, ascended and descended stairs while carrying a gender-specific backpack loaded with 20.4 kg (45 lbs) and equipped either with a standard hip belt or a custom molded hip belt. Pressures from 29 locations under the right side of the hip belt and shoulder strap were recorded for at least three gait cycles of each gait and hip belt condition. Paired T-test analyses were used to test for differences in the peak pressure, mean peak pressure, and the variation in mean peak pressure between the two hip belts during different gait conditions.

Findings: Comparing the quantitative and qualitative results helps discern what may cause discomfort for the backpackers. The Osprey backpack with molded hipbelt was given the highest rating for overall comfort by every subject group while tied with the Osprey backpack with standard hipbelt for the lowest body fat composition group. The subjects also rated the Osprey backpack with molded hipbelt as the condition where many, but not all, of the individual quantities were felt the least (a better rating), suggesting some quantities may contribute to overall comfort more than others. The females gave the Osprey backpack with molded hipbelt a better rating for every quantity; however, the males rated the Osprey backpack with molded hipbelt better for skin pressure and muscle exertion and the Osprey backpack with standard hipbelt better for lower back temperature and relative motion, but the same average score for overall comfort. This suggests that relative to comfort men may be more sensitive to skin pressure and muscle exertion compared to lower back temperature and pack motion, or perhaps there were other factors not measured here that made the Osprey backpack with molded hipbelt more comfortable.

Quantitative Ultrasound Imaging of Muscle-Tendon Units

usang1Ultrasound is being used to non-invasively study the behavior of muscle and tendon as they act within the body under various conditions.

Two ultrasound images of the gastrocnemius muscle, a muscle on the back of the lower leg, are shown to the right. The top image is from a relaxed muscle and the bottom image is from a contracted muscle. The two images illustrate how the orientation of the fascicles within the muscle change with muscle activation.

Fundamental Study of Quantitative Ultrasound

Background: Numerous studies have used ultrasonography to non-invasively quantify muscle-tendon (MT) behavior in-vivo, however the limitations of quantitative ultrasonography have not been thoroughly characterized and therefore conclusions drawn from such studies are questionable.

Methods: Common assumptions inherent in quantitative MT ultrasonography were tested and ultrasound limitations characterized to provide critical information needed to design valid ultrasound studies of MT units. The accuracy and resolution of a commercial ultrasound system used to quantify muscle-tendon deformation and strain were determined by comparing known marker location data with marker location data obtained from analysis of ultrasound images of a phantom, isolated muscle, tendon and fat, and the Achilles tendon of human cadaver specimens during simulated muscle loading.

CadaverLegResults: Results indicate that ultrasound systems have the resolution to make meaningful geometric and deformation measurements of tendon, but speed of sound (SOS) variation, image clarity, and probe location can create large measurement errors. Differences between the actual structure SOS and the SOS used by ultrasound systems (1540 m/s) to create an image can lead to thickness and cross-sectional area (CSA) errors. Structure thickness obtained from digitized ultrasound images were underestimated by 4.8% for tendon, 2.1% for muscle and overestimated by 7.5% for fat. Similar errors would result in CSA measurements. Under ideal conditions ultrasound systems can resolve adult human Achilles tendon strain to less than 0.2%. However in practice poor image quality created by rapid movement and/or structure orientation can reduce this resolution by making it difficult to accurately locate structure boundaries. Achilles tendon strains in a cadaver system with simulated muscle forces were typically determined to be accurate to within 0.2%.

Conclusions: Ultrasound provides a valuable tool for non-invasively studying MT in-vivo, but has limitations that must be appreciated when designing studies and interpreting results. Tendons of sufficient length should be selected based on the ultrasound system resolution to ensure desired strain differences can be detected. CSA data should be interpreted cautiously since large errors may result from errors in tendon boundary identification and differences between the actual and assumed SOS. Image quality and a fixed probe location relative to the plane of muscle-tendon translation are critical considerations to ensure accurate quantitative ultrasonography.

Lateral Force Transmission

Background: Force generated within a muscle-tendon unit (either through muscle activation or stretch) is commonly thought to be transmitted serially from one structure to the next, referred to as myotendinous force transmission. However, there exist alternate pathways that employ intramuscular and intermuscular connective tissues to transmit force laterally, referred to as lateral force transmission (LFT). The extensiveness of LFT and the mechanisms responsible for it are poorly understood.

Methods: Lateral force transmission was investigated in normal and partially compromised passive skeletal muscle systems to determine the fraction of total system force that can be transferred laterally and to investigate possible mechanisms contributing to LFT. Chicken peroneus longus (PL)/middle gastrocnemius (MG)/fascia complexes were isolated taking care to maintain their normal connection to each other. They were attached to a testing fixture that allowed removal and reattachment of the distal end of the muscles. Tensile tests were conducted under three levels of tenotomy (i.e. both muscles attached, only PL attached, only MG attached) and three levels of fasciotomy (100%, 66% and 33% intact).

Findings: LFT between muscles was sizeable and showed directionality (32% of the force applied distally to the MG muscle was transferred laterally to the PL muscle while only 16% of the force applied distally to the PL was transferred laterally to the MG). There was poor correlation between LFT and the ratio of muscle elastic moduli. The contact area between muscles and the ratio of fascia to muscle elastic moduli were the greatest contributors to LFT.

Dynamic Creep

CreepChartBackground: Warm-up exercises are often advocated prior to strenuous exercise, but the warm-up duration and effect on muscle-tendon behavior are not well defined.

Methods: The Gastrocnemius-Achilles tendon complexes of 18 subjects were studied to quantify the dynamic creep response of the Achilles tendon in-vivo and the warm-up dose required for the Achilles tendon to achieve steady-state behavior. A custom testing chamber was used to determine each subject’s maximum voluntary contraction (MVC) during an isometric ankle plantar flexion effort. The subject’s right knee and ankle were immobilized for one hour. Subjects then performed over seven minutes of cyclic isometric ankle plantar flexion efforts equal to 25% to 35% of their MVC at a frequency of 0.75 Hz. Ankle plantar flexion effort and images from dual ultrasound probes located over the gastrocnemius muscle-Achilles tendon junction and the calcaneus-Achilles tendon junction were acquired for eight seconds at the start of each sequential minute of the activity. Ultrasound images were analyzed to quantify the average relative Achilles tendon strain at 25% MVC force (ε25%MVC) for each minute.

Findings: The ε25%MVC increased from 0.3% at the start of activity to 3.3% after seven minutes, giving a total dynamic creep of ~3.0%. The ε25%MVC increased by more than 0.56% per minute for the first five minutes and increased by less than 0.13% per minute thereafter. Therefore, following a period of inactivity, a low intensity warm-up lasting at least six minutes or producing 270 loading cycles is required for an Achilles tendon to reach a relatively steady-state behavior.

Achilles Tendon Stiffness and Hysteresis During Cyclic Loading

StiffnessChartBackground: Tendon stiffness and hysteresis can change during the onset of physical activity, but the magnitudes of these changes and their temporal responses are not well defined.


Methods: The gastrocnemius-Achilles tendon complexes of six subjects were studied in-vivo to quantify changes in Achilles tendon stiffness and hysteresis during the onset of light physical activity. Subjects performed more than 8 minutes of cyclic isometric ankle plantar flexion efforts equal to 25-35% of their MVC at a frequency of 0.75 Hz in a custom testing chamber. Achilles tendon force and length were quantified using a force transduction system and a dual probe ultrasound system, respectively. Achilles tendon stiffness and hysteresis were calculated from cyclic force-deformation responses.

Findings: The average Achilles tendon stiffness decreased from 1228+330 N/mm to 225+82 N/mm during the first five minutes (225 cycles) of loading, and reached a relatively steady-state thereafter. The average percent hysteresis per cycle decreased from 27% to 7% during the first 225 loading cycles. Following a period of inactivity, Achilles tendon stiffness and percent hysteresis can decrease by more than 80% and 70%, respectively, during the first 225 loading cycles of a low intensity activity. These results should be considered in the design of warm-up and pre-conditioning protocols.

Gastrocnemius/Soleus Muscle Force Contributions to Ankle Plantar Flexion Torque

GasSolContributionBackground: The soleus (sol) and gastrocnemius (gast) muscles provide the primary forces responsible for ankle plantar flexion. Understanding the relative and absolute contribution each muscle can make to plantar flexion torque (PFT) as a function of joint angles is fundamental to understanding the role these muscles play in various movements. This study was conducted to quantify the contribution that the sol and gast muscles can make to ankle PFT as a function of ankle and knee angles.

Methods: Collegiate level endurance athletes (6 males (21.5±0.8 years) and 6 females (22.5±2.1 years)) performed isometric maximum ankle plantar flexion efforts (IMAPFE) while positioned in a modified incline bench equipped with an ankle torque transduction system that allowed testing of various ankle-knee angle combinations. Subjects performed three IMAPFE at varying ankle angles (~65, 75, 90, 105, and 120 deg included angle) with their knee in a fully flexed position to minimize gast contributions and therefore isolate the sol muscle’s contribution to PFT. Subjects then performed three IMAPFE at varying knee angles (60, 90, 120, 150, 180 deg included angle) with their ankle fixed at the angle that the subject produced their greatest PFT during the previous testing. Gast muscle contributions to PFT were calculated as the difference in the average PFT determined from this testing and the average maximum PFT from the previous testing.

Findings: With the knee fully flexed, subjects generated their largest PFT in dorsiflexion (60- 75 deg angle). Average maximum PFT was 50 N-m (range 27-70 N-m). PFT decreased with plantar flexion to about 15% of maximum PFT at a 120 deg ankle angle. As a function of knee angle, PFT was greatest at 90 deg (avg=96 N-m, range 58-144 N-m). The gast and sol contributed about equally to PFT for knee angles 60-180 deg (avg gast contribution was 37-54%). The sol and gast contributions to PFT depend on the combination of ankle and knee angles. The average sol PFT is largest when the ankle is dorsiflexed (50 Nm) and decreases to less than 8 Nm in extreme plantar flexion. With the ankle dorsiflexed, the gast can generate its greatest PFT with the knee at 90 degrees. The relative contribution of the gast for this joint configuration and maximum effort is approximately 54%. Gast contribution increases if the ankle angle increases.

Ground Reaction Force Estimation from Hip Acceleration Data

GRF_PA_ModelBackground: To address a variety questions pertaining to physical activity, simple methods are needed to quantify, outside a laboratory setting, the forces acting on the human body during daily activities. The purpose of this study was to develop a statistically based model to estimate peak vertical ground reaction force (pVGRF) during youth gait.

Methods: Repeated measures mixed effects regression models to estimate pVGRF from Biotrainer activity monitor (AM) acceleration in youth (girls 10-12, boys 12-14 years) while walking and running were developed.

Findings: Log transformed pVGRF had a statistically significant relationship with AM acceleration, centered mass, sex (girl), type of locomotion (run), and locomotion type-acceleration interaction controlling for subject as a random effect. A generalized regression model without subject specific random effects was also developed. The average absolute difference between the actual and predicted pVGRF was 5.2% (1.6% standard deviation) and 9% (4.2% standard deviation) using the mixed and generalized models, respectively.

Injury Prevention

Achilles Tendon

AT_strainBackground: The main objective of this study was to test the hypothesis that: physically active adults who are susceptible to Achilles tendon overuse injuries experience higher Achilles tendon strains during an isometric maximum voluntary contraction of the ankle plantar flexors compared to people not susceptible to such an injury.

Methods: Nineteen subjects participated in this study (14 uninjured, 5 injured). Subjects performed controlled isometric ankle plantar flexion efforts in a custom testing chamber while force and muscle-tendon image data were collected simultaneously using a force transducer and a Hitachi EUB 6500 Ultrasound System (Hitachi Corporation) with dual EUP L53 linear probes. Force data were used to determine forces in the gastrocnemius-soleus-Achilles complex (GSATC). Ultrasound images were digitized to determine muscle and Achilles tendon deformation during loading. Achilles tendon strain during maximum isometric ankle plantar flexion was calculated and compared between the uninjured control group and the asymptomatic leg of the Achilles tendon injured group.

Findings: Strain in the Achilles tendon during maximum isometric ankle plantar flexion was higher in the injured group compared to the uninjured group (P=0.127). Tendon strain during isometric ankle plantar flexion may be a good predictor of a person’s risk for sustaining an Achilles tendon injury and may be useful quantity to track during injury prevention or rehabilitation interventions.

Anterior Cruciate Ligament

StopJumpACLBackground: Anterior cruciate ligament (ACL) injuries are one of the most common and potentially debilitating sports injuries. Approximately 70% of ACL injuries occur without contact and are believed to be preventable. Jump stop movements are associated with many non-contact ACL injuries. It was hypothesized that an athlete performing a jump stop movement can reduce their peak tibial shear force (PTSF), a measure of ACL loading, without compromising performance, by modifying their knee flexion angle, shank angle, and foot contact location during landing.

Methods: PTSF was calculated for fourteen female basketball players performing jump stops using their normal mechanics and mechanics modified to increase their knee flexion angle, decrease their shank angle relative to vertical and land more on their toes during landing. Every subject tested experienced drastic reductions in their PTSF (average reduction = 56.4%) using modified movement mechanics. The athletes maintained or improved their jump height with the modified movement mechanics (an average increase in jump height of 2.5 cm).

Findings: The hypothesis was supported: modifications to jump stop movement mechanics greatly reduced PTSF and therefore ACL loading without compromising performance. The results from this study identify crucial biomechanical quantities that athletes can easily modify to reduce ACL loading and therefore should be targeted in any physical activity training programs designed to reduce non-contact ACL injuries.