VideoLab

The VideoLab’s main objective is to advance linguistic research on the connection between space and language in human face-to-face interaction. Therefore, the VideoLab seeks to develop innovative techniques and methods of video analysis that serve to examine the construction of spatiality during interaction. These methods provide new insights into the multimodal accomplishment of interactional spaces and shall ultimately lead to a theoretical modeling of the interfaces between interaction and space.

Two main aspects characterize the interplay between space and interaction, and both of which are addressed by the VideoLab. Firstly: how do interactants create spatial preconditions for their interactions? And, secondly: how do they activate elements in their spatial surroundings in order to pursue their interactional objectives, thereby anchoring their interactions in their spatial-situational surroundings? 

The use of video data provides new insights into the role of space during interaction. Video recordings allow examining how speech and physical activities in space are interwoven.

Unlike other methods, such as questionnaire studies, video recordings permit studying the behavior of the participants while they are carrying out their everyday interaction. And, unlike field notes, for example, video recordings make it possible to repeatedly watch the interaction and to scrutinize even the smallest details of the temporal and spatial organization of the event.

However, working with video recordings also bears new challenges for linguistic analysis of conversation that the VideoLab will be addressing. These challenges can be found in theory construction, in the development of methods for data collection and analysis, and in the development of appropriate presentation techniques for audiovisual data and their analysis.

Theory

As far as theory is concerned, the VideoLab intends to further develop categories to describe spatial configurations formed by the bodies of interaction partners (see Adam Kendon’s F-Formations, Reinhold Schmitt’s Interaction Ensembles, etc.), as well as categories that make it possible to describe features of built spaces relevant to interaction. Finally, a theory that allows an analytical access to the interplay of interaction and built space shall be developed and tested (based, e.g., on Goodwin's concept of Graphic Fields).

Data collection

Even though the inclusion of video data represents a quantum leap in the analysis of interaction, much of what is relevant for and perceivable to the participants remains invisible to the eye of the analyst. Here, the aim is to apply advanced technical methods which will allow an accurate determination of the visual field and the focus of each person involved in the interaction, and thus to meet the demand of Conversation Analysis to adopt the participants’ perspective.

Eye-Tracking-Brillen im Einsatz auf dem SechseläutenplatzEye trackers in use on Zurich’s Sechseläuten square

Concretely, the technical devices involved are mobile eye tracking systems that look similar to glasses. They are supplemented by mini-cameras that can be attached to the subjects’ heads and thus also permit the documentation of the visual behavior of larger groups (e.g. guided tours). Despite working with cutting-edge technologies in data collection, the VideoLab strives to maintain the highest possible level of authenticity. Hence, the VideoLab does not intend to create controlled, and thus highly artificial, experimental environments that have little resemblance with everyday life. Rather, the technology in use should be oriented towards a reduced, low-threshold usage directly “out in the wild” (hence the usage of mobile eye tracking and the preference for inconspicuous technology versus maximizing data quality).

Representation of data

The development of analytical methods is closely associated with the need to develop new forms of data visualization and data presentation: How can transcripts that have usually been focused on language be adapted to the complex requirements of multimodal analysis? This also includes questioning the traditional presentation of transcripts and replacing them with other forms of documentation and data-presentation, such as a 3-D modeling of the interactional space, based on the collected video data, or computer-based representation forms that present the video data in reference to space instead of time.