Shimojima, A., & Katagiri, Y. (2008). An eye-tracking study of exploitations of spatial constraints in diagrammatic reasoning. In G. Stapleton, J. Howse & J. Lee (Eds.), Diagrammatic representation and inference (pp. 74-88). Berlin: Springer.

Shimojima, A.; Katagiri, Y.

2008

Shimojima, A., & Katagiri, Y. (2008). An eye-tracking study of exploitations of spatial constraints in diagrammatic reasoning. In G. Stapleton, J. Howse & J. Lee (Eds.), Diagrammatic representation and inference (pp. 74-88). Berlin: Springer.

Spatial representations, such as maps, charts, and graphs, convey different levels of information, depending on how their elements are grouped into different units of objects. Therefore, how people set boundaries to graphical objects to be interpreted and how they maintain the object boundaries during the given task are two important problems in understanding the way people utilize spatial representations for problem-solving. Table comprehension process was experimentally investigated in terms of eye gaze control behaviors when people were required to read off􀀀 information distributed over large-scale objects, e.g., a row or a column, of the given table. Evidence was found that a large-scale object can be bounded by a single attentional shift to it, and that they can be retained as coherent objects for subsequent reference. These findings suggest the existence of a higher-order information processing in the comprehension of a spatial representation, based on rather intricate processes of attention management.

The process of extracting higher level information from a spatial representation depends on how easily the relevant object can be perceptually integrated, and made available to higher-level cognitive process: to extract higher-level information demanded by the given task, some spatial layouts (e.g., columns in column-wise tables) require only a single shift of attention to an object, while other layouts (e.g., rows in column-wise tables) require multiple shifts of attention. Second, the process of manipulating higher-level information thus obtained can be facilitated by perceptual-motor activities involving eye movements: to recheck higher-level information, some spatial layouts (e.g., column-wise grouping under a column-wise task) allow us to exploit retained large-scale objects by directing eyes back to them, while other layouts (e.g., row-wise grouping under a column-wise task) do not.

32

108