All humans are unstable in the mediolateral direction, and this instability is made worse when walking on uneven surfaces. To compensate for the instability, several strategies are utilized by the neuromuscular system, one being muscular co-contraction. An overwhelming majority of studies utilize EMG data to quantify co-contraction. However, muscle force and EMG are not always directly correlated. Furthermore, some studies have shown a lack of correlation between quasi-stiffness and muscle co-contraction using an EMG-only method. Therefore, the first objective of this thesis was to examine the muscular co-contraction strategies utilized on mediolaterally uneven surfaces and to compare muscle co-contraction measures obtained from muscle moments using a dynamic optimization procedure (EMG-Track), with co-contraction measures derived from EMG data (EMG-Based). Our additional objective was to determine correlations between our EMG-Track method with subtalar joint quasi-stiffness using muscle pairs and muscle group moments (i.e., pronators and supinators of the subtalar joint). On average EMG-Track was better able to detect differences in co-contraction across mediolaterally uneven surfaces and speeds. Muscle group co-contraction was more consistently correlated with quasi-stiffness than muscle pair co-contraction. Furthermore, for muscle group co-contraction, we observed positive correlations before heel strike, and negative correlations following heel strike. These conclusions have important implications for other populations such as older adults, where co-contraction interpretations can alter therapeutic interventions.