Semantic Content of Tactical Planning Analysed Through Multimodal Dialogues

摘要

This study addresses the semantic content of tactical planning conveyed throughhuman multimodal dialogues over a map. Specifically, the aim of this study is to determine the significants (e.g., spatial objects, tactical concepts) of a tactical planning task that commanding officers convey through the concurrent use of verbal and graphical (hand) actions. Seven military officers provided verbal and graphical estimates of a tactical situation portrayed on a topographic map. These sessions were videotaped. The graphical components consisted of graphical actions involving a physical reference to a topographic map. These graphical actions were defined as visual-graphical signifiers in a previous report. A set of significants was determined from the analysis of the commanders multimodal dialogues in their verbal and visual-graphical components. These multimodal dialogues were analysed using lexical, syntactical, pragmatic (planning phases) and semantic levels of linguistic analysis. The significants (N=382) were categorised, in a representational scheme, along two dimensions: spatial classes and type. The first dimension grouped the significants in four classes: geographical features of the area of operation, tactical features of the area of operation, combat power, and courses of action. The second dimension categorised the significants according to their type, i.e., whether they consisted in objects, concepts, attributes of objects (or concepts), or relations pertaining to objects (or concepts). The significants identified in this study represent an intrinsic part of the knowledge used during tactical planning. These significants were deduced from all but two of the planning phases. Since commanders used both verbal and visual-graphical signifiers to identify the same set of significants, then both types of signifiers should be paired for the interpretation of other.