Langston, W., Kramer, D. C., & Glenberg, A. M. (1998). The representation of space in mental models derived from text. Memory and Cognition, 26, 247-262.

@Article{,
  author = 	 {Langston, W., Kramer, D. C., & Glenberg, A. M.},
  title = 	 {The representation
of space in mental models derived from text},
  journal = 	 {Memory and Cognition},
  year = 	 {1998},
  OPTvolume = 	 {26},
  OPTpages = 	 {247--262},
}

Author of the summary: Jim Davies, 2001, jim@jimdavies.org

Cite this paper for:

• Spatial representations can be formed when reading
• Implicit spatial relations are not noticed from mental pictures formed from text
• Spatial representations formed when reading are topological or functional, not Euclidean.
• "It appears that the sorts of spatial relations encoded from text are all explicitly stated in the text, are based on past experience, are depicted in an accompanying picture, or are based on repeated presentations of the text and task demands to form a detailed spatial representation." [p27]

Main & Johnson-Laird 1982, etc.: "readers construct particular spatial representations when the texts are specific," but "readers use a more verbatim representation when the spatial layout is indeterminate..." [2]

Denis, Gonclaves, & Memmi 1995: images generated from verbal descriptions have metric properties.>

The experiments worked like this: Participants read a description of a spatial layout, and they manipulated the distance (in the layout) of a target object from the last object mentioned in the text. If they are `noticing,' then they should be able to answer questions about the closer objects better. The experiments failed to support the hypothesis, leading to the conclusion that the generated images are not euclidean in nature.

Experiment 1 had 2 distances. They found nothing.

Experiment 2 had 5 distances. Nothing.

Experiment 3 used audio instead of text. No noticing effect.

The rest of the experiments were more complicated; I won't go into them.

"It appears that the sorts of spatial relations encoded from text are all explicitly stated in the text, are based on past experience, are depicted in an accompanying picture, or are based on repeated presentations of the text and task demands to form a detailed spatial representation." [p27]

Reading in the absense of multiple passes or pictures rarely results in a Euclidean rep. [28]

Summary author's notes:

• Page numbers are from the online version.