The recognition of multiblock copolymers on nanopatterned surfaces has been investigated by molecular simulations. All the copolymers (A(n)B(12-n))(5) are composed of 60 square-well segments, but with various architectures by changing n. Segment density profiles, radii of gyration, pattern transfer parameters, and three adsorption conformations (tail, loop, and train) are examined quantitatively. It is found that the copolymer can recognize the adsorbing stripes on surface and the surface vicinity. The recognition affinity becomes stronger with increasing the stripe width, the adsorption strength, and the number of adsorbing segments in copolymer chain. From surface to bulk phase, the shape of copolymer changes from elongated to elliptical, and finally to globular. Among the three adsorption conformations, tail has the greatest average size while train has the smallest. With the increased number of nonadsorbing segments, the average size shows an increase in tail but a decrease in train.