Semantic representation in the white matter pathway
Fig 2
The construction and result of behavioral RDMs.
(A) The multi-arrangement method. Twenty college students were asked to arrange object pictures according to their semantic (or modality-specific attribute) relatedness by dragging the items on a screen with a mouse. The distances between items on the screen would transform into an RDM. If two items, e.g., scissors and axe, showed a close distance, then they were assigned a low value in the RDM. (B) The results of the behavioral RDMs. Three broad types of distances were measured: semantic similarity, modality-specific attributes (shape, manipulation, color, and motion), and control models that are also potentially relevant to object naming (early visual, phonological, and object category matrix). The values of dissimilarity were transformed to percentile for display. Red indicates low dissimilarity (high similarity) and blue high dissimilarity (low similarity). (C) Visualization of the semantic RDM using multidimensional scaling. (D) The correlations among various behavioral RDMs. The object line drawings were done by the first author Y.F. The underlying data for this figure can be found at https://osf.io/h7upk/?view_only=52b8f86cffa14ed4844e4a1b9cd429cb. F&V, fruit and vegetable; MDS, multidimensional scaling; RDM, representational dissimilarity matrix.