Using a fragment-based drug design strategy, two biomedical interesting fragments, resveratrol and coumarin were linked to design a series of novel human monoamine oxidase (hMAO) inhibitors with a scaffold of 3-((E)-3-(2-((E)-styryl)phenyl)acryloyl)-2H-chromen-2-one, which demonstrated a very interesting selectivity profile against hMAO-A and hMAO-B: some compounds with this scaffold are selective hMAO-A inhibitors, whereas some are selective hMAO-B inhibitors. The small changes in the substituents of the coumarin moiety led to this interesting selectivity profile. The most potent selective hMAO-B inhibitor D7 has a selectivity ratio of 20.93, with an IC₅₀ value of 2.78 μM, similar or better than selegiline (IC₅₀ = 2.89 μM), a selective hMAO-B inhibitor currently in the market for the treatment of Parkinson's disease. Our modeling study indicates that Tyr 326 of hMAO-B (or corresponded Ile 335 of hMAO-A) may be the determinant for the specificity of these compounds. The selectivity profile of compounds reported herein suggests that we can further develop both selective hMAO-A and hMAO-B inhibitors based on this novel scaffold.