Ruge’s Week 10 Summaries
The Uncanny Valley: Effect of Realism on the Impression of Artificial Human Faces
This paper did not intend to discover the valley, as that was documented on earlier works, it instead sought to quantify its existence and better document what may be the leading factors. The beginning also applied the research from robotics into that of virtual environments. To best understand where the valley fell and when specifically the function of the line changed participation were offered a image of a obviously virtual character that was slowly and methodologically morphed into a human appearing face. The participants rated the each image from pleasant to unpleasant on a numeric scale. Throughout multiple variations of the experiment where they used different characters, and morphed different features in different orders, the obtained very different results. This proved not necessarily that the valley doesn’t always exist, but rather people are more sensitive to different features. Specifically the eyes. The interesting discussions in the paper revolve around the definitions of ‘pleasant’. It is not clear what about certain images makes people uncomfortable. They were able to account that change the size of the eyes relative to other features makes a larger effect, but that doesn’t necessary quantify ‘uneasy’. It is interesting that generally people prefer something 50% human to something 95% human.
Variations in Physiological Responses of Participants During Different Stages of an Immersive Virtual Environment Experiment
The purpose of the paper was to evaluate the use of biological and physiological monitoring devices to gauge immersion, and quantify what variables effect it to what varying degree. The experiment used the monitoring devices to monitor the stress levels of the participants as they interacted in the environment. The results showed that participants became stressed most likely due to the introduction to a new environment and technology. became less stressed as they became accustomed, and then became stressed again as they became bored. The lessons learned in this paper could be applied to many other papers to better account for the effects that VR and AR technology can have on stress levels as a whole. Through better designed testing methodology, perhaps with longer training periods, more analysis of the environments, and some psychological work, the data achieved could be more useful. The other papers discussed in this post due a better job of accounting for such error. An interesting topic to lead from this would be to connect studies done by many game companies in level design and apply them to the design of such tests. Valve did some work when creating the Half Life 2 games, that dealt specifically with controlling player focus and attention.
Physiological Measures of Presence in Stressful Virtual Environments
This paper used biological sensors to measure the reaction to virtual stimuli and use those reactions to gauge the presence and how closely the virtual world simulates the real one. The main environment consisted of a training room that allowed the users to become acquainted with the virtual environment while they used the equipment. This helped filter the possible error that would be caused by sudden excitement about the experiment itself. The training room was connected to a room with a virtual hole in the floor. The users were asked to go to the edge and drop balls into the hole. The experiment varied some parameters including the frame rate of the video feed, the depth of the hole, and the time spent in the simulation. The key component that made a the largest difference in providing a greater amount of presence to the participants was a thin plywood ‘cliff’ that matched the edge in the simulation. This allowed the participants to feel the edge of the pit with there feet. It proved how key a tactile response can be.
Comments are closed.