The highly hydrated colloidal structure of waste-activated sludge (WAS) is the main obstacle of enhanced dewatering for sludge volume minimization. Extracellular polymeric substances (EPS) maintain the colloidal stability of bio-flocs in a three-dimensional matrix due to bindings with bivalent cations (i.e., Ca2+ and Mg2+) and hydrophobic interactions. However, few studies specifically focused on the quantitative relationships among spatial distribution of EPS, microstructure of bio-flocs and fractions of bound water (e.g. vicinal water and interstitial water). Thus, there may be still some debates on whether and what extent of the lysis or flocculation of sludge flocs is optimal for the dewaterability improvement. This study applied the gradient addition of cation exchange resin (CER) to remove EPS-complexed cations and loosen the spatial distribution of EPS. Consequently, how the spatial extension of EPS layers with relief of complex cations influenced the particle size distribution, fractal dimension, interfacial free energy and water occurrence states of WAS was systematically investigated. The quartz crystal microbalance with dissipation (QCM-D) was also applied to analyze the water-EPS interactions with and without the presence of Ca2+ and Mg2+. All the results confirmed that the dispersed EPS adhering layers led to the higher fractal dimension (Df) but the lower space filling degree of bio-flocs. Also, the 4-fold reduction in the polar/acid-based interfacial free energy could be induced by the removal of cations from EPS matrix, which indicated the significant increase in hydrophobicity. Predictably, the fractions of vicinal water and interstitial water were dominated by the polar/acid-based interfacial free energy and pore structure of microbial aggregates, respectively, which were confirmed by the strong Pearson correlation (Rp>0.80, p-value<0.04). These findings are expected to provide the improved mechanistic insights into the relationship between water occurrence states and colloidal structure of WAS, and can serve as the basis for the optimal combination of various sludge conditioning approaches towards regulating aggregation states of bio-flocs.