Current discussions on the mechanism of the omission of locative words are limited by the nature of the noun head, failing to explain the omission of spatial words governed by the same type of noun head in different sentence patterns and address the underlying motivation for spatial words causing nouns to become locative. This paper used Distinctive Collexeme Analysis to examine the semantic distributions of the synonymous motion constructions “N1 + Vjìn + N2” and “N1 + Vjìn + N2 + lǐ”. Based on their spatial roles, verbs were divided into two categories: Vspatial relation and Vspatial region. The study showed that Vspatial relation tended to co-occur with “N1 + Vjìn + N2”; Vspatial region tended to co-occur with “N1 + Vjìn + N2 + lǐ”. Thus, this paper constructed the interaction continuum of verb spatial roles and noun semantic roles, pointing out that the omission of lǐ is the interaction between verbs’ spatial role and the location, with the cognitive motivation being the schema transformation mechanism of lǐ.