Magnolol and honokiol can be widely expected to efficiently extract from the magnolia bark residues for the full utilization of traditional Chinese medicine resources. However, an accurate and rapid extraction has been still a great challenge on the magnolia bark residues, due mainly to the relatively similar chemical structures and physicochemical properties of natural active substances. Deep Eutectic Solvents (DES) as excellent reaction media exhibit a promising prospect to extract the natural active substances. At the same time, microwave heating can also greatly contribute to determining the pharmaceutically active components in natural plants. In this study, ten environmentally friendly and economic DESs were synthesized for the selective extraction of lignanoids (including magnolol and honokiol) from the magnolia bark residues. A microwave-assisted DES approach was also conducted under the conditions of solid loading of 1:15 g/mL, the extraction temperature of 50°C, the extraction time of 30 min, and the microwave power of 160W. It was found that the choline chloride/levulinic acid (ChCl-La) was the most efficient solvent for the lignanoids extraction with a product yield of 20.31 mg/g. It infers that the ChCl-La was beneficial for the extraction mass-transferring, due to a comparative lower viscosity, but higher molecular diffusivity. Besides, the hydrophilic DES presented the higher solubility of lignanoids, according to the principle of like dissolving like. Some cofactors of solvent extraction were evaluated under the same microwave power, including the solid-liquid ratio (1:10, 1:15, 1:25, and 1:40 g/mL), the molar ratio of DES components (1:2, 1:3, and 1:4), and solvent water content (30%, 40%, and 60%). The lignanoids yield was also predicted during the DES extraction under various conditions of extraction temperature (30, 50, 70, and 90℃) and extraction time (30, 60, 90, and 120 min). The optimal extraction yields were achieved for the honokiol and magnolol (9.93 and 22.95 mg/g) under the extraction conditions of the solid loading of 1:25 g/mL, DES components molar ratio of 1:2, the water content of 30%, extraction temperature of 90°C, and extraction time of 90 min. Among them, the water content most significantly influenced the product's yields. Therefore, a suitable amount of water was added to decrease the viscosity of the extraction system, particularly for the sufficient hydrogen bonding to interact with products. The yield of total lignanoids (32.87 mg/g) was 395.03% higher than that of the methanol in Chinese Pharmacopoeia (6.64 mg/g) under the same conditions, indicating that the ChCl-La was an efficient solvent for the lignanoids extraction from magnolia bark residue. The molecular mechanism of extraction was investigated by Density Functional Theory with the Gaussian 16 program package. A Kohn-Shame wave function was selected to characterize the interaction between lignanoids and the solvent in the optimized composite system. Subsequently, the interaction between molecules in the composite system was visualized using a reduced density gradient. It was found that the interaction between solvent and magnolol was the hydrogen bond, followed by Van der Waals force, whereas, the bonding between solvent and honokiol was only Van der Waals force, indicating that the extraction rate was dependent on the strength of the hydrogen bond. More importantly, the binding energy of DES solvents with lignanoids demonstrated that higher energy was required for the binding of honokiol and ChCl-La than that of magnolol and ChCl-La, indicating a higher yield of the latter.