Open victorbarres opened 8 years ago
Figure 17: An illustration of “hypothesized” TCG processes in the two cases of Kuchinsky (2009)’s study. (A) illustrates the case where the sentence structure is specified first, resulting in a process consistent with the structural view (i.e., no cueing effect due to the verbal guidance principle), whereas (B) illustrates the case where a lexical item (woman) is specified earlier than the sentential structure (SVO), resulting in a process consistent with the incremental view (i.e., cueing effect). Since low threshold is assumed for both of the cases, utterances are made intermittently. The yellow shade represents the place where speaker’s attention focuses, and the dashed red oval marks subscene.
Kuchinsky example used by Jin. Just as in the previous example two questions:
Jin mentions three types of attentional processes involved in scene perception: Zooming-in, Zooming-out, and shifting. Can I implement those? Can this make the implementation of subscenes based scene representation better defined?