Assignment 8
Kraut et al's paper explores the function of visual information in communication that is focused on the completion of a collaborative physical task. First, visual information provides situational awareness - knowledge of the actions of the other participant and the progress of the task. This ability to draw conclusions about the state of the task allows people to plan out relevant utterances geared toward the completion of the task. Second, when people share the same visual information, they increase their common ground, allowing for more efficient communication. When no visual cues are present, one expects to observe many utterances which contain long, explicit descriptions and clarifications.
The researchers conducted two experiments to analyze the effect of visual technology on task-based communication, in which a worker was aided by an experienced helper with different technologies. Surprisingly, it was found that communication that utilized both audio and video did not result in substantially higher performance in the completion of the task than communication which just used audio. However, the types of utterances varied considerably, as predicted. The lack of visual information prompted workers to give more explicit descriptions and helpers to provide more statements of acknowledgement. Also, workers' descriptions were followed by help in the visual condition more often than in the audio condition because in the former, descriptions were viewed as implicit requests for help, while in the latter, they were interpreted as attempts to ground the helper.
One of my concerns (also mentioned by Kraut et al) is that the results may be too dependent on the specific type of video equipment used. How significantly might the data change if the camera were in a fixed position (on a tabletop, say) rather than attached to the worker's head? A constant view of the entire scene may allow for faster completion of the task. Could this be the reason why there was no notable difference between setups that used video and those that did not? Why did the experimenters choose to use a head-mounted camera rather than a fixed one?
I would also like to know whether gender played a role in the outcome of the experiment. For example, if the two experts in Experiment 1 were comprised of one male and one female, was there a difference was there a difference in how effectively they helped the workers? Similarly, were male or female workers more successful in completely the task efficiently?
posted by Evan @ 10:58 PM
3 Comments:
This comment has been removed by a blog administrator.
Good breadth and depth in your summary of the main points of the research. In regard to your concerns/questions, I agree with your sentiment in most of them. I think that attaching a camera to a worker's head would change the data significantly. Having a camera on someone's head just seems unnecessary in that it doesn't help with communication process. People's heads are always moving around, and it just makes communicating over the video system that much harder and more awkward. Fixing the camera on a table top would likely generate significant differences because it would be easier to make conversation by looking into the screen and not having to worry about the view moving. I feel like this definitely could have been the reason why there were no significant differences found between the tasks. The logic of why a head-mounted camera was chosen puzzles me a bit as well.
With regard to your question about gender, it seems like everyone in the class has shared that sentiment. It seems like it could have played a major role in the data for a wide variety of reasons, and definitely would have been worth looking at. I'm surprised it wasn't though of, especially given that this study is only a couple of years old. (In fact ... the word "gender" isn't even mentioned once in the paper!)
I agree with Christina that having a camera that can be controlled by the user would be a good way of fixing the head-mounted camera problem. I feel that this would be better than the solution of having the users share the same visual field, as having someone glance in multiple directions could be confusing and disorienting. While I agree that gender could have played a role in this process, I think that the researchers were correct in leaving that out at this stage. An analysis of gender in this context would probably not be possible until much later.
Post a Comment
<< Home