Fitness Research update
Knee behaviour in squatting

A partnership between Australian Fitness Network, the University of the Sunshine Coast and the Australian Institute of Fitness, Fitness Research studies the populations, communities and environments related to the fitness industry, with the mission of improving the health of Australians through an improved body of fitness knowledge.

Research paper: Knee Behaviour in Squatting
Research team: Dr Mark McKean; Professor Brendan Burkett, University Sunshine Coast
Published: Journal of Australian Strength & Conditioning, Volume 20 Number 2 June 2012
Read more:


Introduction: From motor control research the central nervous system (CNS) selects muscle torques at each joint to develop movement. The order in which these movements are initiated has resulted in the leading joint hypothesis (LJH) which states there is one leading joint that initiates or controls movement of the entire limb. The leading joint accelerates or decelerates in order to control the movement pattern, while the other joints simply regulate muscle torque and movement for the task. In the squat movement, the LJH suggests one of the joints involved would control the movement and the other joints would simply follow. However, there is little, if any, evidence in the published literature to suggest that any of the lower extremity joints would behave in a leading joint way

The purpose of this study was to examine if the leading joint hypothesis exists for the squat movement.

Methods: Twenty-nine subjects – a cross sectional group with sub-elite and strength training backgrounds (16 males and 13 females), at least 12 months squatting experience and free of musculoskeletal injury – participated in the study.
Data pertaining to the lower limbs and torso was captured as subjects completed four sets of eight repetitions of a below parallel squat using two different loads: bodyweight (BW), and an additional 50 per cent of the individual’s bodyweight (BW+); and two different widths of stance: narrow stance (equal to anterior superior iliac spine (ASIS)-width), and wide stance (equal to twice ASIS width).

The load and stance order was randomised and a two-minute rest was implemented between sets. Foot alignment in narrow-stance was toes straight ahead and wide-stance no more than 30 degrees away from midline. Time for descent and ascent was normalised: the top of the squat 0 per cent, and bottom 100 per cent. Hip angles were defined as the anterior angle between lines connecting knee joint centre with hip joint centre, with midline of trunk. Knee angles were defined as the posterior angle between a line connecting hip joint centre with knee joint centre and ankle joint centre. The shank angle was the most anterior angle reached, using a global reference system with vertical being the reference of zero degrees.

Knee-width was reported as the distance between knee joint centres, and knee forward position reported as the distance of the patella relative to the vertical line of toes. Data was analysed for three consecutive repetitions in the middle of each of the four sets, with subjects being blind to the actual repetitions used.

Results: Men and women achieve similar hip angles when performing below parallel squats with hip angles being within 3 degrees for all variations. Similarly, shank angles were also less than 1.4 degrees different between genders for all squat variations. The maximum knee angle difference between genders for each variation ranged from 10.8 degrees to 16.4 degrees. Synchronisation of the hip knee and shank was shown to be consistent in all squat variations and loads. Shank angular velocity suggests a mechanism to control balance by altering the shank and knee position to maintain the synchronisation at hip and knee. Knee-width changed by 5 to 10cm throughout the squat, moving through both a narrower and wider distance than that of the width of stance – showing that the knees do not remain aligned with the toes or at a constant distance apart (Figure 1). The early movement of the knees anteriorly (aligned with toes before halfway down – Figure 2) supports the LJH and the results from this study suggest that in order to maintain good synchronisation of the hip and knee joints, the knee must initiate the downward movement and move forward to a point where this synchronisation can be maintained.

Take home message: It appears that early knee movement is important in initiating the downward movement of the body to squat. Knee positions also appear to be less static than previously suggested, and it may be important to allow your client to adjust their knees both anteriorly and mediolaterally in order to achieve a good pattern of movement. The teaching cue ‘break at the hips’ may alter this pattern and cause less than ideal synchronisation of the hips and knees. This ‘knees first’ approach to good squatting also shows that the knees will move forwards, but that each person will have a different range of anterior movement of the knee, with some clients moving further forward from the vertical alignment of the toes than others.

Summary: The authors suggest that rather than having a black and white approach to medial and anterior knee movement, trainers should view their clients in the sagittal plane as well as frontal plane. If the squat pattern shows similar timing for the hip and knee then the knees may be reaching the desired position. However, if the timing of the hip and knees appears to be ‘disorganised’, then coach the movement of the knees to move first and allow the knees to adjust slightly, both medially and laterally, in order to gain better synchronisation.