Representing Spatial Structure Through Maps and Language: Lord of the Rings Encodes the Spatial Structure of Middle Earth

Abstract

Spatial mental representations can be derived from linguistic and non-linguistic sources of information. This study tested whether these representations could be formed from statistical linguistic frequencies of city names, and to what extent participants differed in their performance when they estimated spatial locations from language or maps. In a computational linguistic study, we demonstrated that co-occurrences of cities in Tolkien's Lord of the Rings trilogy and The Hobbit predicted the authentic longitude and latitude of those cities in Middle Earth. In a human study, we showed that human spatial estimates of the location of cities were very similar regardless of whether participants read Tolkien's texts or memorized a map of Middle Earth. However, text-based location estimates obtained from statistical linguistic frequencies better predicted the human text-based estimates than the human map-based estimates. These findings suggest that language encodes spatial structure of cities, and that human cognitive map representations can come from implicit statistical linguistic patterns, from explicit non-linguistic perceptual information, or from both.

Archaeology Through Computational Linguistics: Inscription Statistics Predict Excavation Sites of Indus Valley Artifacts

“…The idea is simple: Places that are spatially close to each other co-occur together in language, even when the texts in which they co-occur are not spatial in nature per se. And these findings were not limited to existing cities: Locations in fictional Middle Earth were reliably estimated from the novel The Lord of the Rings, with experimental findings indicating that readers of the novel rely on language statistics in their geographical estimates (Louwerse & Benesh, 2012). Similarly, using Arabic and Chinese texts, we were able to estimate the location of the largest cities in the Middle East and China, respectively (Louwerse, Hutchinson, & Cai, 2012).…”

“…As in our other studies (Louwerse, 2011;Louwerse & Benesh, 2012;Louwerse & Zwaan, 2009), we used latent semantic analysis (LSA), a commonly used algorithm that estimates similarities in meaning (Landauer & Dumais, 1997;Landauer, Laham, & Derr, 2004). LSA is a statistical technique that generates vectors representing terms (e.g., words) and documents (e.g., paragraphs).…”

“…Following our methodology in earlier investigations (Louwerse & Benesh, 2012;Louwerse & Zwaan, 2009), cosines between site vectors were entered into a two-dimensional approximations of LSA similarity matrices via multidimensional scaling analysis, and the resulting coordinates were compared with the actual longitudes and latitudes of each excavation site. For the purposes of predicting the relative locations of archeological sites in the Indus Valley, document vectors for each of the five sites were summed to create five site vectors.…”

Word-embeddings Italian semantic spaces: A semantic model for psycholinguistic research

“…Moreover, rocca (stronghold) is associated with names of famous Italian castles, like Calascio, Sanvitale, and Torrechiara; however, since Rocca is also the surname of the Alpine skier Giorgio Rocca, it also produces as neighbors names of other athletes, like Raich (Benjamin Raich, also Alpine skier), Schoenfelder (Olivier Schoenfelder, ice dancer), and Rivellino (Roberto Rivellino, football player). To a certain extent, also geographical information is captured in the model (in line with the results by Louwerse & Benesh, 2012). The neighbors of Belgio (Belgium) are either surrounding countries (Francia -France, Olanda -Netherlands, Germania -Germany) or Belgian cities (Namur, Charleroi).…”

A Systematic Review into Factors Influencing Sketch Map Quality

“…The majority of studies (6) indicate that sketch maps based on a real map are better than those based on a text [114][115][116][117]. Only one study failed to confirm this conclusion, finding no difference between the use of a map and a text [118]. One study confirmed that both a map and a text had a positive impact [119].…”