Abnormal protein aggregation in the nervous tissue leads to several neurodegenerative disorders producing intracellular inclusions or extracellular aggregates in specific brain areas. The accumulation of the Amyloid Beta peptide in the brain is proposed to be an early important event in the pathogenesis of Alzheimer disease. Recently, four prenylated compounds the active constituents of Psoralea Fructus (PF), have demonstrated some anti-Alzheimer effects both in vitro and in vivo. These compounds exhibited strong inhibitory activities on Aβ42 aggregation. Regardless of the significant research efforts in this field, the molecular mechanisms of protein ligand binding remain somewhat unrevealed. Within this framework, computer simulations represent a powerful tool able to connect experimental findings to nanoscale molecular events. In this work, a combined molecular modeling study was performed on four active constituents of Psoralea Fructus (PF) as Aβ42 aggregation inhibitors. Starting from docked conformation of the compounds, a 100-ns molecular dynamics (MD) simulation and binding free energy calculations were carried out to determine the binding modes of the ligands and to identify main interacting residues. The binding free energies calculated by the MM/PBSA method showed the importance of the van der Waals interaction. A good correlation between the MD results and the experimental AB42 aggregation rate was observed. The results from this study can identify the key amino acid residues in AB42 binding site and can provide some insights into the development of potent novel AB-42 aggregation inhibitors.