#8 - option 2

In Kraut et. al's article, the main focus was to study "the ways in which visual information is used as a conversational resource in the accomplishment of collaborative physical tasks." By varying the amount of visual information avaliable in their experiment (which consisted of a worker and a expert helper completing a collaborative physical task--repairing a bicycle), Kraut et. al were able to study the part that visual information played in maintaining situational awareness as well as achieving conversational grounding (i.e. mutual understanding). In order to achieve situational awareness, both participants must be conscious of where they are at in completing the task (i.e. how much more is there to do before the bicycle is repaired), and what one another is doing at the moment. This will enable them to coordinate their communication to the other's needs. To achieve conversational grounding requires 1) the helper to phrase their utterances so that the worker can understand them as well as the intended meaning and 2) the worker to acknowledge that they have understood the worker.

The first prediction was that: 1) the worker and helper who were co-present would achieve the highest performance on completing the task, followed by the pair who used the video system devised by the experimenters, followed by the pair who used audio cues alone. The second prediction was that: the less the amount of available cues as to provide situational awareness, the more explicit the worker's requests to helpers.

Contrary to their predictions, it was actually found that the use of the video system (which consisted of head-mounting the worker with a camera so that the helper can see, from a first person perspective, what the worker sees) did not improve performance significantly compared to the lack of the video system. The second hypothesis, however, was proven.

A concern I have is that the experimenters never really clarified what the shared visual space consisted of. Only by introducing and telling the participants what each can see will mutual knowledge and common ground be achieved. Perhaps this affected the outcome of the experiment?

Another concern that I have is that if the experimenters used a different kind of camera (e.g. one of higher quality with little bandwidth limitation), or positioned the camera differently, would that affect the outcome of the experiment?