This study makes comparison of spatial descriptions for navigation between Japan and America from cross-cultural and geographic perspectives, based on 24 tourist guidebooks of four cities in Japan (Kyoto, Tokyo) and the U.S (Boston, New York City). The contents of maps and linguistic information in the guides were quantified and then analyzed. The results indicate that Japanese guidebooks use predominately imagery information of maps , while American guides mainly use linguistic information. Therefore, we can insist that there is a complementary relationship between these modes of spatial information vehicle, language and imagery. The results also demonstrate that a relative frame of reference with landmarks is the most fundamental sentence construction for giving directions. In principle, linguistic information can be used to complement the lack of imagery information in describing a given geographic environment, so its use rates increase in relatively unfamiliar environments. However, the contents vary with the environmental characteristics such as the regularity of street pattern. Difference in address systems between two countries also affected the way of sorting the sites, style of maps, and the use frequency rates of linguistic information.
