The aim of this study was to determine the effects of a newly developed reciprocal gait orthosis (RGO) with a variable constraint hip mechanism (VCHM) on the kinematics and kinetics of normal gait. The VCHM was compared with the isocentric reciprocating gait orthosis (IRGO) for walking after paraplegia. Both the VCHM and the IRGO were evaluated with able-bodied volunteers with the hip reciprocating mechanisms coupled and uncoupled. The VCHM was further evaluated with context-dependent coupling based on a finite-state control algorithm utilizing information from brace-mounted sensors. Walking performance for each brace condition was also compared to normal walking without an orthosis. Without the hip controller, the VCHM affected the kinematics of the hip joint in a similar manner as the IRGO, regardless of whether the hip reciprocator was coupled or uncoupled. With the controller active, hip kinematics with the VCHM were closer to normal gait than with the IRGO or any other condition tested (Intraclass correlation coefficient, ICC=0.96). The effects of the braces on the knee and ankle angles were not as prominent as their effects on the hip angles. In terms of kinetics, the VCHM with controller active allowed the generation of joint moments that were closer to normal (ICC=0.80) than the IRGO with hips coupled (ICC= 0.68). There was no statistically significant difference between the various conditions tested in terms of step-length and no statistically significant difference in the preferred walking speed between the IRGO and normal walking, whether or not the hips were coupled. However, there was a 25% reduction in walking speed with the VCHM when compared to normal, and the relative magnitudes of the EMG activity of three muscles (tibialis anterior, quadriceps, and hamstrings) were also higher with the VCHM than with either the IRGO or normal gait, likely due to the additional weight of the mechanism. Overall, the VCHM with controller active provided smooth control of the hip joints via context-dependent coupling and allowed for increased hip flexion relative to the IRGO. The results suggest that the VCHM with controlled joint coupling may eventually be a valuable component of a hybrid system combining functional electrical stimulation (FES) with orthotics.