Kortum P. (ed.) HCI Beyond the GUI. Design for Haptic, Speech, Olfactory, and Other Nontraditional Interfaces

Подождите немного. Документ загружается.

9.5

TECHNIQUES FOR TESTING THE INTERFACE

The ability of the Food Simulator to represent virtual foods was evaluated.

The major objective of this experiment was to discover whether the device could

represent differences in food texture.

Experiment 1

Task:

Participants were asked to bite the device and answer whether the

virtual food was a cracker or cheese.

Participants:

The participants (22 to 24 years of age) were six university

students (male) who voluntarily participated in the experiment.

Procedure:

Each participant tried 10 virtual foods. Crackers and cheeses were

randomly displayed, with the subjects being requested to distinguish between

the two foods.

Results:

The answers were 98.3 percent correct.

Discussion:

All of the participants were perfectly able to distinguish between

the two virtual foods. The device thus succeeded in representing hardness of

the food, perceived by the first bite.

Experiment 2

Task:

The following three foods were displayed:

Food A:

Cracker (same as in the previous experiment)

Food B:

Biscuit (“Calorie Mate”)

Food C: Japanese crispy snack (“Ebisen”)

These foods possess a similar texture. Figures 9.15 and 9.16 show the

force profiles of Foods B and C, respectively. The same force control method as

shown in Figure 9.11 was used to simulate Foods B and C. The participants

were asked to bite the device and select one of the three foods.

Participants:

The experiment was performed using the same participants as in

Experiment 1.

Procedure:

Each food was displayed 60 times; thus, each participant sampled

180 virtual foods. The subjects were requested to distinguish among three

randomly displayed foods.

Results:

The percentages of correct answers followed:

Food A:

69.1 percent

Food B:

59.1 percent

Food C:

28.2 percent

Discussion:

Force profiles of the cracker and the Ebisen were very similar.

The peak force of the Ebisen was greater than that of the cracker, but most

9 Taste Interfaces

302

[N]

[s]

0.5 1 1.5

FIGURE

9.15

Measured force of a biscuit.

Only one peak is observed.

[N]

[s]

0.5 1 1.5

FIGURE

9.16

Measured force of a Japanese snack.

Two peaks are observed that are very similar to those of the cracker.

9.5 Techniques for Testing the Interfa ce

303

participants were unable to distinguish the Ebisen from the cracker. The peak

force of the Calorie Mate was the greatest of the three foods. The internal

structure of the Calorie Mate is soft so that the force profile did not exhibit

a second peak. This difference contributed to the 60 percent of correct

answers.

9.6

DESIGN GUIDELINES

Design of the Food Simulator is in a very preliminary state and it has much room

for improvement. However, the following items should be considered for design-

ing the device.

F Linkages of the force feedback device should be fit to the structure of the

user’s jaw.

F The mouthpiece of the force feedback device should be thin, but it has to sup-

port large biting force. The material of the linkage should be carefully chosen.

F Due to sanitary issues, the mouthpiece should be covered by disposable mate-

rial. The material should not infringe taste perceived by the tongue.

F The injector tube for the chemical taste should be set at an adequate position

so that the liquid drops on the tongue.

9.7

CASE STUDY

A case study of a taste interface can be found at

www.beyondthegui.com

9.8

FUTURE TRENDS

A major advantage of the Food Simulator is that it can display biting force by the

use of a simple mechanism. Although the apparatus is inexpensive, it can effec-

tively display food texture. The method of force control requires improvement,

but most of the participants in the Special Interest Group on Graphics and Interac-

tive Techniques (SIGGRAPH) greatly enjoyed the demonstration.

On the other hand, major limitations of the current prototype are concerned

with its thickness and the weight of the linkages. The linkages are 2 mm thick.

Because great force must be supported, the device must be fabricated in steel.

When users finish biting, they can feel the unnatural sensation of the linkage

thickness. The weight of the linkage causes unwanted vibration.

The user’s teeth contact the flat surface of the linkage. This flat surface

degrades food texture. A person feels the force on independent teeth while biting

real food. Moreover, the current device applies force only to the teeth. The texture

9 Taste Interfaces

304

of real food is perceived, in part, by the tongue. However, displaying food texture

to the tongue is very difficult.

Biting exercises contribute to human health. Some of the many application

areas for the Food Simulator follow:

Training:

The Food Simulator can be programmed to generate various forces

other than that of real food. Elderly people can practice biting with reduced

resistance to the teeth. On the other hand, increased resistance enables younger

people to perceive the difficulty in biting experienced by elderly people.

Entertainment:

The Food Simulator can change the properties of food while chew-

ing. A cracker can be suddenly changed to a gel. The user can enjoy a novel expe-

rience while chewing. This kind of entertainment contributes to the chewing

capability of children.

Food design:

Preferred resistance by the teeth can be found using the Food Simu-

lator. Such findings could contribute to the design of new foods.

This chapter has demonstrated a haptic device that simulates the sense of bit-

ing. The device presents a physical property of food. The Food Simulator has been

integrated with auditory and chemical taste sensations. The chemical sensation

was successfully displayed by a tube and an injection pump.

The device can be used in experiments of human taste perception. Taste is

a multimodal sensation, making it very difficult to control modality by using real

food. The device can display food texture without chemical taste, unlike real food.

This characteristic contributes to experiments in modality integration. Future

work will include the psychological study of sensory integration regarding taste.

REFERENCES

Chandrashekar, J., Hoon, M. A., Ryba, N. J., & Zuker, C. S. (2006). The receptors and cells

for mammalian taste.

Nature

444:288–94.

Davide, F., Holmberg, M., & Lundstro

m, I. (2001). Virtual olfactory interfaces: Electronic

noses and olfactory. In Riva, G., & Davide, F., eds.,

Communications through Virtual

Technologies: Identity, Community and Technology in the Communication Age

Amsterdam: IOS Press, 160–72.

Kawamura,Y., & Kare, M. R. (1987).

Umami: A Basic Taste

. New York: Marcel Dekker.

Kohyama, K., Ioche, L., & Martin, J. F. (2002). Chewing patterns of various textured foods

studied by electromyography in young and elderly populations.

Journal of Texture Stud-

ies

33(4):269–83.

Kohyama, K., Sakai, T., & Azuma, T. (2001). Patterns observed in the first chew of foods

with various textures.

Food Science and Technology Research

7(4):290–96.

Nakamoto, T., Ustumi, S., Yamashita, N., Moriizumi, T., & Sonoda, Y. (1994). Active gas

sensing system using automatically controlled gas blender and numerical optimization

technique.

Sensors and Actuators

B 20:131–37.

References

305

Szczesniak, A. S. (1963). Classification of textural characteristics.

Journal of Food Science

28:385–89.

Szczesniak, A. S. (2002). Texture is a sensory property.

Food Quality and Preference

13:215–25.

Takanobu, H., Takanishi, A., Ozawa, D., Ohtsuki, K., Ohnishi, M., & Okino, A. (2002).

Integrated dental robot system for mouth opening and closing training.

Proceedings of

ICRA

, 1428–33.

Toko, K., Matsuno, T., Yamafuji, K., Hayashi, K., Ikezaki, H., Sato, K., & Kawarai, S. (1994).

Multichannel taste sensor using electric potential changes in lipid membranes.

Biosen-

sors and Bioelectronics

9:359–64.

Toko, K. (1998). Electronic tongue.

Biosensors & Bioelectronics

13:701–709.

Tsuruoka, S., Wakaumi, M., Nishiki, K., Araki, N., Harada, K., Sugimoto, K., & Fujimura, A.

(2003). Subclinical alteration of taste sensitivity induced by candesartan in healthy

subjects.

British Journal of Clinical Pharmacology

305:807–12.

9 Taste Interfaces

306

CHAPTER

Small-Screen Interfaces

Daniel W. Mauney, Christopher Masterton

10.1

NATURE OF THE INTERFACE

The use of small screens as a tool for displaying dynamic information is becom-

ing ubiquitous. Displays range from very simple screens as seen on clocks,

microwaves, alarm systems, and so on, to highly capable graphical displays as

seen on mobile phones, medical devices, handheld gaming devices, and per-

sonal digital assistants (PDAs). The number of products sold with small screens

is staggering. Mobile phone sales, just one category of small-screen products,

were estimated at 825 million units in 2005 (Gohring, 2006) and have over 2 .6

billion subscribers worldwide (N ystedt, 2006). In comparison, personal com-

puter (PC) sales were estimated at 208.6 mill ion units in 2005 (Williams &

Cowley, 2006).

Small-screen design primarily makes use of the visual system. Questions in

small-screen design often center on how small data elements can be displayed

so that they are properly seen or recognized. To gain a deeper understanding of

the answers to this question, a brief background of light, eye anatomy, and the

sensitivity and acuity of the eye are discussed.

10.1.1 Light

Light is generally described by its wavelength and intensity. The subjective or

psychological correlate to wavelength is color (more accurately known as hue).

Thus, the color that people see is their perception of the light’s wavelength. Wave-

length is measured in nanometers (nm). The subjective or psychological correlate

to intensity is brightness. Brightness is the impression produced by the intensity

of light striking the eye and visual system.

People do not normally look directly at a light source. Most of the light seen is

reflected from surrounding surfaces. The measurement of the intensity of light

focuses on two aspects: the amount of light falling on an object, called

illumi-

nance

, and the amount of light emitted from or reflected from a surface, called

luminance

. The common English unit of illuminance is the foot-candle (ft-c),

while the metric unit of illuminance is the meter-candle (m-c). The common

English unit of luminance is the foot-Lambert (ft-L), while the metric unit of lumi-

nance is the millilambert (mL). The amount of light emanating from a source is

called

radiance

. To put all of this together, consider the act of reading a book.

The amount of light generated by the light bulb is called radiance, the amount

of light falling on the book’s pages is called illuminance, the amount of light

reflected from those pages is the luminance, and the amount of light perceived

by the visual system is called

brightness

Contrast

is a measurement of the luminance difference between a target and

its background. It is often expressed as a ratio, such as 10:1.

10.1.2 The Eye

As light rays enter the eye, they first pass through the cornea, the pupil, and the

lens to ultimately fall on the retina. The cornea and the lens refract (or bend)

the light so that images come into focus on the retina. The iris controls the

amount of light entering the eye by dilating or constricting the size of the pupil

(the round black opening surrounded by the iris).

The retina is composed of nerve cells and photoreceptors that are sensitive

to light. There are two types of photoreceptors: rods and cones. Rods are heavily

concentrated in the peripheral region of the retina, while cones are concentrated

primarily in a small pit called the fovea. Many rods share a common optic nerve

fiber, which pools their stimulation and aids sensitivity to lower levels of light.

In contrast, cones have a more or less one-to-one relationship with optic nerve

fibers, which aids their image resolution, or acuity. Vision accomplished primarily

with cones is called photopic vision, and vision accomplished primarily with rods

is called scotopic vision. Only in photopic vision do people actually perceive

colors. In scotopic vision, the weak lights are visible, but not as colors. Instead,

all wavelengths are seen as a series of greys.

10.1.3 Sensitivity

In poorly lit environments, scotopic vision dominates because rods are far more

sensitive to light. Since rods are located primarily in the periphery of the retina,

sources of low light are best seen in the visual periphery. This explains why it is

easier to see a faint star at night when not fixating directly on it.

The sensitivity of the eye to light is heavily dependent on the wavelength of the

light. Figure 10.1 shows the spectral threshold curves for photopic and scotopic

10 Small-Screen Interfaces

308

vision. These curves show that the rods are maximally sensitive to wavelengths of

around 500 nm (yellow–green), while the cones are maximally sensitive to wave-

lengths of around 550 nm (green). It also shows, as noted before, that scotopic

vision is generally more sensitive to light than photopic vision.

10.1.4 Visual Acuity

The smallest detail the eye is capable of resolving at a particular distance is known

as the minimum visual acuity (Grether & Baker, 1972). Visual acuity of an object

depends on both its size and its distance from the viewer. A small object that is

Logarithm of relative radiant flux required

for threshold vision

(arbitrary units)

400 500

Scotopic

(rod)

vision

Photopic

(cone)

vision

600

Wavelength in nanometers

700

FIGURE

10.1

Spectral threshold curves for photopic and scotopic vision.

Rods are maximally sensitive to wavelengths of around 500 nm, compared to

550 nm for cones. Scotopic vision is generally more sensitive to light than

photopic vision.

Source:

From Schiffman (1990), as adapted from Chapanis (1949);

courtesy John Wiley and Sons, Ltd.

10.1 Nature of the Interface

309

very close may appear larger than a large object that is far away. The distin-

guishing feature of an object’s size, therefore, is the size the object projects on

the retina, also known as its visual angle (Figure 10.2). The probability of an object

being detected is directly related to the visual angle, expressed in seconds,

minutes, and degrees, of that object subtended at the eye. Figure 10.3 graphs

the probability of detecting an object at different visual angles.

Factors Affecting Acuity

A number of factors affect acuity, including illumination, contrast, time, and wave-

length. In general, as illumination, contrast, and time spent focusing on a target

Target 1

Size = 2

Distance = 25

Target 2

Size = 1

Distance = 12.5

Target 3

Size = .5

Distance = 6.25

Visual angle

FIGURE

10.2

Impact of visual angle.

All three targets appear to be the same size because their visual angle is the

same.

Source:

From Schiffman (1990); courtesy John Wiley and Sons, Ltd.

100

0 0.2 0.4 0.6 0.8

Visual an

le (min)

Probability of detection (%)

1.0 1.2 1.4 1.6

FIGURE

10.3

Probability of detection for objects of varying visual angles.

This probability is directly related to the visual angle of an object subtended

at the eye.

Source:

From Grether and Baker (1972), as adapted from

Blackwell (1946); courtesy John Wiley and Sons, Ltd.

10 Small-Screen Interfaces

310

increase, acuity improves. Wavelength has little impact on acuity if there is high

luminance contrast. However, if there is low contrast, color contrast will improve

acuity. A red signal is most easily seen in low-contrast conditions, followed by

green, yellow, and white in that order (Sanders & McCormick, 1993).

10.1.5 Color

Color perception is the ability to discriminate among different wavelengths of light.

It is useful to note that the light itself is not colored. Instead, different wavelengths of

light produce the sensation of different colors. Thus, color is a psychological experi-

ence that different wavelengths of light have on the nervous system. There are three

psychological dimensions of color: hue, brightness, and saturation. Hue varies with

changes in wavelength, and the term is often used interchangeably with the word

color

. Brightness varies with the physical intensity of the light. Saturation refers to

the spectral purity of the wavelength. The addition of other wavelengths, white

light, or grey to a single wavelength will desaturate the color (Schiffman, 1990).

Approximately 8 percent of men and 0.5 percent of women have some form

of color vision deficiency (Sanders & McCormick, 1993). The most common form

of color weakness involves the inability to differentiate between red and green.

As such, color should never be used as a primary cue (the primary distinguishing

feature between two items), and requiring users to distinguish between the colors

of red and green should be avoided.

10.2

TECHNOLOGY OF THE INTERFACE

Until 1970, the primary display technology available to electronics designers was

the cathode ray tube (CRT). Although the CRT is still in use today and still out-

performs competing technologies in some areas (especially manufacturing cost),

it has two major drawbacks: size and power draw. Cathode ray technology

requires a great deal of power to fire electrons at the screen and a relatively great

distance in order to display those electrons on a reasonably sized screen. Despite

continued improvements in the CRT engineering process, these two important

flaws have not been overcome.

In 1970, the Swiss team of Schadt and Helfrich built the world’s first liquid

crystal display (LCD). While original LCD designs had their own shortcomings

(namely cost, response time, and contrast), the technology overcame the limita-

tions of the CRT by requiring much less power and a vastly decreased depth.

The continued improvement of the LCD has led to the emergence of an

enormous range of small-screen products, from the simple digital watch to power-

ful laptop computers. In this section, we explore the various technologies used in

current small-screen design.

10.2 Technology of the Interface

311