In hydraulic fracturing, the transport and deposition patterns of proppant particles are crucial for high recovery production of oil and gas hydrocarbons. The conductivity of proppant pack depends closely on the effective stress, proppant strength, proppant embedment, and proppant areal concentration. In this study, a series of laboratory experiments was conducted to study the role of effective stress, proppant particle size, and particle size heterogeneity on the fracture conductivity versus proppant areal concentration curve. In the previous research from our lab, a numerical-modeling approach, combining the discrete-element method (DEM) with single-/multiphase Lattice Boltzmann (LB) simulation, was introduced to advance understanding of the interaction between reservoir depletion, proppant particle compaction, and single-/multiphase flows in a proppant-supported hydraulic fracture. Our numerical modeling results showed that the conductivity curve of a proppant pack having various particle sizes cannot be simply obtained based on linear interpolation of the conductivity curves of the proppant packs having homogeneous particle sizes. To verify the prediction of the DEM-LB model, we collected three types of proppants with mesh-20/40, 30/50, and 40/70 sizes and then conducted the conductivity testing experiment under specific mixing ratios using a fracture conductivity measurement cell. The laboratory-measured fracture conductivity versus proppant areal concentration curves show an overall good agreement with the modeling prediction from the DEM-LB coupled numerical model. It is confirmed that the conductivity curve of a proppant pack having various particle sizes cannot be simply obtained based on linear interpolation of the conductivity curves of the proppant packs having homogeneous particle sizes. This study uses laboratory experimental methods to provide insight into the role of effective stress, proppant particle size, and particle size heterogeneity on the fracture conductivity versus proppant areal concentration curve. The research outcome will facilitate the optimization of proppant selection and placement in hydraulic fracturing to maximize proppant-supported fracture conductivity and to maximize the final recovery factor.