Knee rigidity due to aging or disease is associated with falls. A causal relationship between instability and knee rigidity has not been established. Here, we examined whether insufficient knee movement due to knee rigidity could underlie poor balance control in patients. We addressed this by examining the effect of artificially "locking" the knees on balance control in 18 healthy subjects, tested with and without individually fitted knee casts on both legs. Subjects were exposed to sudden rotations of a support surface in six different directions. The primary outcome measure was body centre of mass (COM) movement, and secondary outcome measures included biomechanical responses of the legs, pelvis and trunk. Knee casts caused increased backward COM movement for backward perturbations and decreased vertical COM movement for forward perturbations, and caused little change in lateral COM movement. At the ankles, dorsiflexion was reduced for backward perturbations. With knee casts, there was less uphill hip flexion and more downhill hip flexion. A major difference with knee casts was a reversed pelvis pitch movement and an increased forward trunk motion. These alterations in pitch movement strategies and COM displacements were similar to those we have observed previously in patients with knee rigidity, specifically those with spinocerebellar ataxia (SCA). Pelvis roll and uphill arm abduction were also increased with the casts. This roll movement strategy and minor changes in lateral COM movement were not similar to observations in patients. We conclude that artificial knee rigidity increases instability, as reflected by greater posterior COM displacement following support surface tilts. Healthy controls with knee casts used a pitch movement strategy similar to that of SCA patients to offset their lack of knee movement in regaining balance following multidirectional perturbations. This similarity suggests that reduced knee movements due to knee rigidity may contribute to sagittal plane postural instability in SCA patients and possibly in other patient groups. However in the roll plane, healthy controls rapidly compensate by adjusting arm movements and hip flexion to offset the effects of knee rigidity.