-- KarlinBark - 23 July 2004
More testing...

Questions to ask the test subject:

Number correct on arm
1. 16/18
2. 11/18
3. 16/18
4. 17/18
5. 18/18
6. 14/18

Number correct on back of hand
1. 15/18
2. 16/18
3. 16/18
4. 17/18
5. 18/18
6. 16/18

Did the high force hurt?
1. No
2. No
3. No
4. No
5. Yes
6. No

Was the low force detectable?
1. Yes, no problems detecting the low force
2. Yes, but sometimes it was harder to detect. On one occasion, I applied a low force, but the subject couldn?t tell it was there
3. Yes, but it took awhile to become accustomed to the low force?sometimes wasn?t sure if a low force was being applied or if the tester was just moving the scale around.
4. There were some problems detecting the low force
5. When a low force was applied immediately after a high force, it was a bit difficult to notice it?the only reason the subject knew it was a low force was because the tester indicated that he was applying a force
6. Yes, but some problems

Was it easy to distinguish between medium and high?
1. Sometimes it was difficult, especially the medium?would think that the medium force was a high force, until an actual high force was applied, then the subject realized that the previous force was a medium force
2. Same as above.
3. Yes, but only after some time passed and the subject was able to become accustomed to the different feelings.
4. sometimes it was difficult, subject felt as if he was guessing
5. it wasn?t always easy, but once the subject got used to the forces, it was easier to distinguish them
6. Sometimes it was difficult

Where was it easier to distinguish the forces, on the hand or on the arm?
1. The back of the hand was easier
2. The back of the hand was easier
3. The forearm was easier.
4. The back of the hand was easier
5. The back of the hand was easier
6. The back of the hand was easier

Misc. Observations
1. none
2. The subject did not remember the calibration at first. Most of the incorrect responses for the forearm occurred because of mis-calibration (he was off by one each time, i.e. each time I applied a low force, he said it was medium, and I applied a medium force and he thought it was a high force?until I applied a high force, then the subject realized the calibration was off, and mentioned that his responses before must have been incorrect).

Also, this subject had more hair on his arm than previous subjects. Instead of testing on the ?outside? of the forearm, we had to test on the ?inside? of his forearm, where there was less hair. However, we had problems with the tape sticking to his skin. He also stated that because the tape began to slip, it created problems for him in distinguishing the forces. He said the sensation of the tape un-sticking from his arm created a distinct feeling and it interfered with the skin stretch forces he felt. However, he wasn?t sure if he thought the extra sensation was helping him distinguish high forces, or if it hurt.

3. It was difficult to have the tape stick onto this subject?s arm?.unusually difficult. For some reason the tape that was used on other subjects would not stay on this subject?s arm. To compromise, a larger piece of tape was used to stretch the skin. Having a larger piece of tape, would, in my opinion, make things more difficult
4. none
5. none
6. none

Li and I discussed the status of the project. And we both discussed the fact that we weren?t entirely convinced that skin stretch was much better than vibration in terms of giving a person more information. The advantages of vibration are how easily it can be applied, and the low cost of vibration devices. Skin stretch provides challenges in reducing slip, the different skin properties of each individual. We hypothesize that the ability to discern different magnitudes of skin stretch is much like the ability of a person the judge the weight of an object by holding in their hand. Some people can estimate the weight of an object fairly easily and relatively accurately, but this occurs only because of past work experience and knowledge, whereas others will have no perception of weight not because they do not have the ability to, but because they have no prior experience with weights. Skin stretch, seems to be somewhat similar. It seems like there is a memory curve that is affecting the ability of people to determine the amount of skin stretch. A more reliable use of skin stretch would be to limit skin stretch to binary commands, either skin is being stretched or it isn?t. However, the binary aspect of the skin stretch would give it no more sensitivity or ability to translate more data than a pager motor that vibrates upon impact. Li and I started to discuss what we thought was a more compelling idea, to compare vibration to skin stretch, mostly because we weren?t convinced that skin stretch was a method worth pursing if it did not provide any more information than vibration. So Li noted that one of the disadvantages of vibration was that it was not a realistic feeling. It is very rare that in the real world, that vibration will be the main source of touch/contact. Objects don?t vibrate, humans don?t vibrate. But, we thought that skin stretch could provide a more realistic contact. As Mark mentioned before, studies have shown that not all haptic/touch feedback has to be completely realistic, just sort of (i.e. arcade games where a pod/game seat moves or tilts?.the machine only tilts 5-10 degrees, but to the user, it feels like they are being thrown about, 45 degrees or more). So we?ve decided to move on to test the ?realness? of skin stretch. By proving that skin stretch provides a more realistic response, we think it would help to solidify the direction we have been pursuing.

To compare which is more realistic, skin stretch or vibration, we?ve decided to do another simple test. Most of the vibration applications in virtual environments seem to be impact or collision detection. So we would like to explore this application. We plan to apply some sort of collision to a user?s hand in three different ways. First, we would apply a real object. Then, we would apply the skin stretch, then the vibration and ask the subject which of the two felt more like the first object hitting their hand. We could do this test with several different objects, varying in hardness. We could simply strap on a pager motor to simulate the vibration, and we could use a similar method we have been using now to simulate skin stretch. More thought must be put into how we will strap on devices, and consider using a visual cue as well to simulate the act of an object hitting the subject?s hand. I guess we?ll work on that next week.