The present study is concerned with how the Chinese learners of English grammaticalize different English syntactic rules. The ERPs (event related potentials) data were collected when participants performed English grammatical judgment. The experimental sentences varied in the degree of the similarity between the first language Chinese (L1) and the second language English (L2): (a) different in the L1 and the L2; (b) similar in the L1 and the L2; (c) unique to the L2. The P600 effect was found in L2 for structures that are similar in the L1 and the L2 and that are unique in L2, but there was no P600 effect of sentence type for the mismatch structures. The results indicate L1-L2 similarity and L2 proficiency interact in a complex way.