Week 10 summary
Physiological Measures of Presence in Stressful Virtual Environments
Presence is defined as “the feeling of being there”. The best way to calculate the level of presence managed by a VE experience is by comparing the responses of people to a similar “there” situation, I mean a real situation. As noted by the paper, the feeling of presence is very subjective and as with other research in HCI, a self-reported observation may not be very reliable. In this paper, the authors try to quantify the feeling of presence by looking at certain physiological measures like the heart rate, skin conductance and skin temperature.
The authors conducted three experiments in order to measure these three physiological measures and studied the effect in each scenario. The three experiments were Effects of Multiple Exposures on Presence (Multiple Exposures), Effects of Passive Haptics on Presence (Passive Haptics), and Effects of Frame Rate on Presence (Frame Rate). In the tests the researchers tried to measure the reliability, validity, sensitivity, and objectivity of the physiological measures.
They found that presence measures decrease over multiple exposures to the same VE, but not to zero. This did not undermine their findings since, it has been known that the measure of presence reduces over repeated runs of the same VE experience. The heart rate was the best indicator of the presence among the other factors. Moreover, the passive haptic inclusion increased the presence significantly.
The Uncanny Valley: Effect of Realism on the Impression of Artificial Human Faces
Humans are complex creatures. Researchers in robotics have always tried to build robots that look and behave exactly like humans. Although that is a very scary thought, yet that has been the pinnacle robotic scientists try to achieve. However studies have showed an interesting phenomenon highlighted in this paper called “The Uncanny Valley”. This phenomenon states that if a robot (or any similar object like a doll, or virtual character) is very realistic and similar looking to a human but not exactly alike, it elicits an unpleasant impression in humans. This is generally more true (according to the paper) when an otherwise good replica has some jarring or abnormal features, probably causing the look of a disfigured person.
The researchers ran four experiments by presenting web based pictures of different artificial faces that had been morphed on a scale of perfect to abnormal. The variations of experiment was based on uniform morphing, morphing of eyes (or head) first, manipulations of eye size and variations in morphing to avoid confounding. The results of experiments seemed promising. None of the four types of morphing sequences showed that an almost perfectly realistic human appearance was a sufficient condition for the uncanny valley to emerge. It was generally in the case of a disfigured or abnormal looking feature on an otherwise perfect looking face. The uncanny valley was prominent when the eyes and the head showed the largest mismatch in the degree of realism. Participants gave the lowest pleasantness score when the eyes were 100% real human and the head was 0% real human. Similar was the case, when the head was 100% real human and the eyes were 0% real human, which validated the idea that people did not appreciate the idea of a hybrid or mixed “reality”.
This research would support the notion that when designing robots or characters, try to make them appear exactly human-like or intentionally nothing like humans at all. “Somewhere-in-between” is where the Uncanny Valley lies.
Variations in Physiological Responses of Participants during Different Stages of an Immersive Virtual Environment Experiment
This study uses various physiological monitoring techniques like the Meehan paper to study the level of presence in users however, unlike the Meehan paper which immerses the users in a stressful situation, this paper attemps to look at users in more general situations. The study is performed during an immersive virtual environments experience with 40 participants. They attempt to measure user’s response to virtual stimuli when there is a successful substitution of real sensory data by computer generated sensory data.
The experiment is run by having the user walk in a representation of a street using a CAVE-like system. The researchers further study the heart rate as a realistic measure of presence in virtual environments. However, the paper also reveals some limitations of using heart rate as a measure of Virtual Environment experiments, since these experiences due to the head mounts and the entire setup implicitly impose certain stress on the users, which could possibly skew the results.
The results indicate that there are different mental and physical stress levels in the different time segments of the experiment. The mental stress is high at the beginning of the experiment since the newness of the system, but then decreases at the beginning of the actual experiment, and it increases again at the end of the experiment. In addition, they found that the more realistic the environment the greater is the influence on the mental stress of participants. However, unlike the Meehan experiment, multiple expose to the same experiment did not reduce the heart rate significantly in the participant.
Comments are closed.