Computers and Social Change Main Contents | Chapter Contents

Computers and Social Change
Part Two. Ergonomics: The Human/Technology Interface


Is using a computer sometimes, literally, a pain in the neck? Do your eyes have trouble focusing on the screen? Does your upper arm ache from repeated strain on your mouse finger's tendon? I know mine do as I write this sentence. Physical stress is one of the physiological problems with the video display terminal operation. Ignoring my own advice to you in this chapter, I've been sitting here for five hours.

Are computers fun? Are they convenient? Does using one make you feel competent and in control? These are some of the psychological satisfactions of a "user friendly" tool. At the moment, my answers are no, yes, and yes.

But how frienly can a computer be, when "friend" is a social relationship? Are you talking more to computers and less to people? Or are you using the computer as a way to communicate, perhaps the way I'm talking to you via the World Wide Web. These are some of the social issues of ergonomics, the study of the human/technology interface. Most computer ergonomic research is micro-level. It studies the physiology and psychology of the immediate connection between computers and people.


The effects of the computer on the individual is often considered a technical question involving user-friendly interfaces between the person and the machine. Human factors research (as ergonomics is also called) is a relatively new field that looks at the match between the machine and the human being as a biological organism (Helander, et. al., 1984 ref). Cognitive factors of perception and learning are also taken into account in order to create interfaces that accomodate the user. A "good" interface is user- friendly; it meets human needs. But to understand the human/computer interaction we must consider what it is that human beings need.

4.1.1 A Functionalist Model of Human Needs

The model of human needs developed by A. H. Maslow is one of several psychological approaches to the growth of human personalities. As shown in Figure 4-1, it represents an overlapping set of needs whose relative strengths vary according to individual situations and personalities. Basic biological needs are strongest determinents of our actions when we are children. As our personalities develop, the "higher" cognitive and psychological needs become more important. For example, we may ignore our hunger in order to finish a computer program before dinner. Maslow's thoery assumes a moral progression from personalities organized to satisfy immediate wants to those oriented towards acting responsibly in the social world. Using Maslow's model, we could ask how computer use satisfies or frustrates individual need. The physiological requirements of human beings -- food, shelter, and air are some of these -- are not usually directly met by computers. However, computers indirectly contribute to the production

not scanned in yet
Source: David Kretch, Richard S. Crutchfield, and Egerton L. Ballachey. Individual in Society. New York: McGraw-Hill, 1962, page 77.
and distribution of clothing and food. Computers may also be used to regulate furnaces and thermostats to provide for a tolerable temperature. Sensory input can also be considered a physiological need, since the human organism cannot interact with its environment without information. Research on cognitive factors in computing addresses the issues of how compter-generated information can be matched to human perceptal characteristics.

The next two levels of need refer to peoples' social needs. Society provides for the safety and security of its members and gives them a sense of group membership. Social institutions like the family and social relationships like friendship provide a sense of love and belonging. Although a person can survive physically without others, that person's self-identification as a member of a social group is a basic part of his or her social nature. Computers act on this level of need through their effects on institutions and relationships.

The higher needs defined by Maslow are the individual psychological ones. Self-esteem is social insofar as the approval of others is a contributing factor, but self-esteem can also be personal. Self- actualization is the need to act on the basis of one's sense of self, and to feel that one's activities are successful in whatever terms one values. In a study of self-actualizing persons, Maslow found them

more efficient in perceiving reality, more accepting of themselves and others, more spontaneous in their relationships, with a tendency to centre on problems and their solution; to have a quality of privacy and detachment, and autonomy from cultural influences, a freshness of appreciation, a capacity for transcendence and oceanic feelings, a deep identification with humanity, more profound human attachments, a humorous and democratic character structure, and a rare capactiy to resolve moral dichotomies and dilemmas (Hampden- Turner, 1983 ref).

While all social theorists would agree with Maslow that individuals have biological, social, and psychological needs, not all would agree with his selection, or his priorities. There is great disagreement over what peoples' biological requirements are and where biology gives way to culture in defining them. Some theorists would add competition or territoriality to the list; others would argue about whether self-esteem is based on social status or on private, internal feelings of satisfaction. Others would question whether the needs for personal identity are "higher" than the social needs to fulfill the obligations and gain the benefits of group membership. Environmentalists would argue that there is a higher need to preserve and protect the earth for future generations; philosophers and theologians would make a case for an even higher level of spiritual need.

4.1.2 Capitalism and Human Needs

Critical social psychology argues that "as society becomes increasingly commodified, social interaction and the conditions which make it possible disappear" (Wexler, 1983: 97 ref). Advertizing creates artificial social needs and develops personalities whose self-satisfaction is derived from consumption (Ewen, 1976 ref). As the market for commodity production expands, it replaces social relationships in supplying material security, recreational, and even emotional needs (Braverman, 1974: 276-277 ref). From this perspective, computer- mediated social interactions represent an additional intrusion of the commodity market into social life.

According to this perspective, the goal of ergonomics research is not to meet any individual needs, but to match humans with computers in ways that raise productivity. Improved comfort or user satisfaction are most important in cases where imporved satisfaction leads to productivity increases. Where there is a conflict between productivity and comfort, design choices will be made for productivity.


Use of a computer is a physical activity involving hands, eyes, ears, and body. The goals of improving the health and safety of computer users is usually compatible with the goal of higher productivity. This is because the factors causing physical stress or damage also tend to make people less productive.

Northeastern's Office of Environmental Health and Safety provides links to information on office ergonomics containing current information on preventing physical problems.

A list of other Ergonomics servers can be found at UCSF/UCB Ergonomics Program

4.2.1 Physical Stress

Muscular and skeletal stress are frequently reported by computer users. Like any other job requiring long periods of sitting at attention, using a terminal can be uncomfortable. With properly designed furniture, much of this stress can be alleviated. Figure 4-2 (not yet scanned) shows the design parameters for an ergonomic workstation. The chair height prevents leg strain; the keyboard is at a height to prevent tension buildup in the back of the neck caused by holding the arms at an awkward position. Wrist rests, advocated by some designers, have been found by researchers to be unnecessary or even a hindrance to someone constantly using the keyboard.


Unless a workstation is to be used by only one person, furniture must be adjustable. Designs for the average man or woman are usually unsuitable for just about everybody except those few whose body proportions happen to be exactly average. For example, I am of average height for a woman when sitting down and of average height for a man when standing up. Before adjustable workstations were common my legs were "too long" and my body was "too short" for on-the-job comfort. From the perspective of ergonomics, variations in human proportions are a "given"; the designer's task is to make a machine that accomodates the variety in human form.

No amount of design, however, can change the fact that the human body did not evolve the ability to remain motionless for long periods of time. Frequent changes in position, and physical exercise, are both essential for maintaining muscle tone and avoiding stress and fatigue. Improvements in voice-operated computers may be the best way to match active humans to immobile machines. In the meantime, computer users and workstation designers are developing more awareness of the physical consequences of their work. "User friendly" hardware shouldn't give the user a pain in the neck. When you have the choice, you should plan your own work to avoid long sessions seated at a computer terminal. I find short walks and long backrubs ideal.

4.2.2 Vision

The human eye is a rather delicate optical system. The most well- documented health problems associated with computer use have been eyestrain, with sometimes permanent damage. Eyestrain can result from both poor lighting (Sliney, 1985 ref) and from poor resolution of displayed characters. Glare can be reduced by keeping a screen away from windows and ensuring that no light fixtures shine directly on the screen. Overhead lighting for workstations works less well than indirect lights on the material being typed in an otherwise dim room. Flourescent lighting creates a special visual problem, because the cathode ray tube in the terminal and the flourescent bulb both fluctuate rapidly. This may cause a flicker effect that is very stressful to the eyes. Incandescent light (the ordinary light bulb) is recommended. Eye specialist Lowell Glatt (1985 ref) recommended in the mid 1980's that people with any visual problems should not be hired to work at video display terminals. Today, following the passage of the Americans with Disabilities Act, such a policy would be considered discriminatory in the United States.

The productivity factor of most concern to designers of screen displays is the eye's ability to distinguish shapes. In addition to causing eyestrain, failure to make visual discriminations can result in higher error rates for whatever type of work is being done. Character fonts (the shape of letters and numbers) can be designed to improve people's recognition of individual characters. When characters are displayed as a pattern of small dots (pixels) on a screen, the greater the number of dots used, the greater the character's readability. In character fonts made in a 7x9 matrix (the smallest number of pixels for a readable European language character set) the following characters are commonly confused:

Proper contrast between character and background is an important lighting factor. Sharp edges between a character and its background improve our ability to see it. Many personal computer screens are quite fuzzy. Although light characters on dark backgrounds are most common, glare and user error rates are reportedly lower for dark characters on light backgrounds.

Improvements in display resolution include the use of color combinations (like the popular amber on black) that are easier on the eyes. Research on the physiology of color perception resulted in the recommendations in Table 3. Although the use of color in screen displays can reduce fatigue and be psychologically more interesting, about 8% of all males (and about one half of one percent of females) are colorblind to some extent, most commonly to red, green, or both (Birren, 1978).

Ergonomic factors in screen design include information about how people read. In English and other left to right languages, screens that present material from left to right are more successful than top to bottom ones. This applies as well to the number pad found on many terminals and calculators. When the numbers 7, 8, and 9 are on the top row, users make more errors than when the top row is 1, 2, and 3. Although many of our designs are "traditional", even experienced users adjust easily to a new layout and quickly improve their performance.

A final example of a visual discrimination problem built into a traditional design is the data entry form with little boxes for each letter. Filling in the boxes by hand takes 16% longer than it would to simply write the material on a straight line. Reading the boxes takes 28% longer than reading material written on a line. The boxes also result in a higher error rate when someone types from them onto a computer keyboard (Wright, 1984). The people who first designed forms with boxes probably thought they were matching human data recording techniques with the requirements of efficient keypunch machine operation. But microergonomic research showed us that an interface that doesn't match human perceptual requirements is less efficient for data processing.


4.2.3 VDT Radiation

Radiation is an emotionally charged concept, more likely to be associated with nuclear explosions than with familiar objects like microwave ovens or color television sets. Video display terminals (VDT's) emit electromagnetic radiation, as shown in Figure 4-2. So do other electrical and electronic appliances. Most of this is low frequency, non- ionizing radiation, as shown in Table Figure 4-3. The dangerous ionizing radiation associated with nuclear reactions is in a much higher frequency range. Medical X-rays, not computers, account for over 95% of our exposure to it (Grundy, 1978).

According to recent research, VDT terminals meet the guidelines established by the American National Standards Institute for safe levels of exposure to low frequency radiation (Murray, 1984). However, because it is difficult to measure extremely low frequency electromagnetic fields (below 10 kHz), and because human neurological and cell processes occur in the 8-25 Hz range, scientists do not have enough evidence on the biological effects of VDT radiation in this range to conclude that they are harmless (Harvey, 1984). Static Electricity
If your computer terminal gives you small electric shocks, especially after you walk across a wool rug, it's not radiating -- it's building up an electrostatic charge. Just as sox cling together when they come out of the drier, this can attract dust to the screen, interfering with your vision. Some research suggests that particles can
also be attracted to the VDT operator, causing in rare cases a skin rash (Tjonn, 1984). Companies make anti-static features (including floor coverings and de-ionizing devices) to protect floppy diskettes and other magnetically encoded data from being accidentally erased by static electric discharge.

Static buildup may also affect your mood. An imbalance of positively- charged ions in the air (as in the Santa Anna wind which sometimes blows across Southern California) can cause irritability and fatigue. A balance of negatively charged ions (as often found in mountaintop areas) are reportedly invigorating. Some research has found that negative ion generators reduce the frequency of headaches and make employees feel more comfortable and alert (Hawkins, 1984). Microwave Radiation
The Risks of Excessive Microwave Exposure. A long-term study of career officers in the Polish military has found a positive correlation between cancer and long, job-related exposures to high levels of microwave and radiofrequency radiation (Dolnick, 1985). Microwaves can cause cataracts, according to studies by opthamologists. In the U. S. a lawsuit for cataract damage believed to be caused by VDT radiation was settled out of court, awarding $10,000 to a woman who was unable to read for more than 10 minutes, or to tolerate bright light. The New York Workman's Compensation Board ruled that there was a "recognizable link" between VDT work and a claimant's eye damage (VDT News. 1985, Jan/Feb:8). However, microwave ovens and color televisions are more common sources of microwave exposure than are VDT terminals. When appliances and terminals are functioning properly, they do not "leak" dangerous levels of microwave radiation.

Terminals can be shielded to prevent emissions. Although the Computer and Business Equipment Manufacturers Association called the announcement of a VDT radiation shield an "irresponsible" appeal to the fears of people who know little about radiation ("Cbema...", 1985), an IBM report recommended shielding older terminals. It concluding that the newer ones were better designed and represent no hazard (VDT News March/April, 1985:4-7). People who are worried about the possible effects of display terminal radiation should also be concerned about their TV set. Doubling the distance between yourself and the screen will reduce your exposure to any possible emissions by 75%.

4 .2.3.3 Possible Reproductive Hazards
The possibility of a connection between VDT's and reproductive hazards is an issue characterized by few facts and much controversy. One study reported that continuous exposure to an extremely low frequency (50 Hz, 5 kV/m) field over several months reduced the sperm viability of male rats; another reported an increased rate of birth defects in the children of fathers working at high-voltage electric power plants where they had a long-term exposure to a 50 Hz, 400kv field (Nordstrom, et. al., 1983). This possible damage to male reproductive cells is believed to be because some sperm cells become heated, the way a microwave oven heats food (Jensh, et. al., 1982).

Women, whose reproductive organs are better protected by layers of body tissue, have also been studied for possible reproductive damage. Although clusters of problem pregnancies have been reported among VDT operators, the evidence has been insufficient to establish any radiation hazard. Physical stress, for example, might be a contributing factor in miscarriages. Several major studies are underway to evaluate any potential problems (VDT News, July/August, 1985:4-6). A large recent Finnish study of birth defects found no evidence that they were related to VDT use by the mothers (VDT News, Jan/Feb, 1985:12). Defining VDT Health Risks
Defining health risks is more than a scientific question of what radiation is produced by computers and what the effects electromagnetic fields have on biological organisms. It is also a social issue, defined by the concerns of employees and their unions, the interests of manufacturers and businesses, and the policy-making institutions of government and law. My own research on reproductive protection policy for industry makes me appreciate how difficult it is to make decisions in the absence of good information (Perrolle, 1993).

The possible health effect of VDT radiation remains an unresolved issue in ergonomic research, in public policy, and in computer design. Should the hazards finally be assessed as serious, shielding users from extremely low frequency emisions would become a social priority in computer technology. In 1985 the U.S. Congress concluded that there is insufficient evidence of risk to establish VDT regulations, but several states have passed laws doing so. Based on its best available information, the Human Factors Society proposed an American National Standard for Human Factors Engineering of Visual Display Terminal Workstations. For information on the current status of ergonomic recommendations, you should consult the current version of the standard.

4.2.4 Computers for the Disabled

Although most microergonomics research focuses on the industrial applications of computer technology, another important area which studies the match between humans and machines is bio-engineering. There the focus is on medical procedures, artificial limbs, and adaptive technologies for the blind, deaf, or physically impaired. In the functionalist model of the human/computer interface, computer aids for the disabled are a clear example of meeting the most basic sort of human need. From a critical perspective, these would be viewed as contributing new products to the market for medical commodities and as making disabled people better able to do productive work. Another issue is that of the status of disabled persons in society. The medical model of a disabled person as having a physical problem that should be corrected by changing that person ignores the solution of changing the society (and its technologies) to accomodate a physically diverse population. Communications for the Deaf
One of the earliest telecommunications networks, the TELEX system, supported a class of specially-designed terminals (TDD's) for the deaf. Intended to replace the telephone, they allowed subscribers to transmit written messages to one another. By the 1980's, any personal computer with communications capability could be used to transmit written text translated between the 5-bit code for the TDD's and standard ASCII. Current technologies for the deaf can be found through the DEAFTEK site.

Visual aids for lip readers included a computerized pair of eyeglasses which display a visual signal generated by a small voice analyzer. Designed as an aid to lip readers, the display distinguishes sounds which look deceptively alike. An anatomically correct talking head that can be used as a video screen reader is currently under development. Data Access for the Blind
Specially-equipped terminals have been developed for the blind, who may also expect to benefit from improvements in voice- activated systems. The Kurzweil reading machine was an early practical result of voice analysis research. It uses an optical scanner and sound synthesizer to "read" printed text aloud. Although it was expensive (about $30,000) and limited in capability, it allowed the blind to read text printed on paper. Technologies for braille writing and printing were developed, but serve only the minority of visually disabled people who read braille. There have also been experiments in transmitting computer-generated visual signals that bypass the eyes. While this is not expected to restore sight, it was hoped that it would enable some blind people to distinguish shapes. Like ear implants for the deaf, these technologies are controversial among the community of people with disabilities.

The most useful adaptive technology for blind computer users is the screen reader, which reads computer files and directories aloud. Not all files can be accessed by screen readers. For example, many cdrom directories are inaccessible. The widespread introduction of the GUIs (graphical User Interfaces) such as Microsoft Windows has created a challenge for adaptive technology designers and users. Windows are difficult for screen readers, despite some promising developments such as Georgia Tech's "Accessible X".

One design idea proposed in the 1980's for viewing low resolution graphs (though I haven't seen a resulting product) was a computerized tactile grid which allows the blind to "see" graphics output. A colleague reports that there is now a technology that allows special paper to be used in any laser printer. When heated, the graphical output is raised on the paper,

Screen readers can be connected to a text only web browser (like lynx) to allow blind web surfers to hear web pages. However, what lots of web pages sound like is:

image, image, click here for a great image, image, click our image map
As more information moves onto the worldwide web, the design of accessible web pages has become an important social and ethical issue. In putting this book on the web, I have been trying different ways to provide alternatives to scanning in images of figures and tables. [but since I haven't had time to put in the tables and figures yet, you can't view or hear the results...maybe audio files would work as alternatives to graphics. %-) any ideas, anybody?] Robotics for the Physically Impaired
Robots are tools for manipulating objects by remote control. Although most current developments in the field are aimed at military and industrial interests, the same technology can be applied to preserve the health of people working with hazardous chemicals, and to provide hands and arms for people whose mobility is limited. At the University of Utah, researchers use advanced robotics and computer graphics to design artificial arms that can be controlled by the nervous system of the amputee. Similar graphical techniques are coming into use to analyze body motion in sport and dance (Voy, 1984 ref; Deken, 1983: 137-143 ref). Voice recognition and other non- keyboard data entry systems, although still in the experimental stages, should benefit individuals who have physical trouble handling books, telephones, and other information equipment. Also in the experimental stages is a computer-controlled device to transmit nerve impulses to the legs of people paralyzed by spinal cord injuries. Meeting Human Needs?
Like the Kurzweil reading machine, many of the new computer applications for the disabled are prohibitively expensive. A voice synthesizer for a mute child cost $100,000 in 1985. Although many electronics products have dropped in price, they are still expensive for the disabled, who as a group have lower than average incomes Other products are announced with great enthusiasm while they are still experimental. Bio-engineer Howard Chizeck (1985 ref) criticizes commercial promises of electronic devices to let the paralyzed walk again, arguing that it is holding out false hopes for a product that is really years away. There is a gap between what a computer professional knows the technology can do, if resources are available to develop an application, and what is practically feasible for the majority of handicapped people. Chizeck asks the question, "Who will Pay?", and concludes that computer applications that save health care costs and make people employable will be paid for by insurers. Those which only improve the quality of life, like being able to walk when you can already get to work in a wheelchair, may not be. Although the technology is available, it may not be affordable.


Psychology is the study of human beings as they relate to the world, to one another, and to themselves. While the sociologist takes the relationships among people in larger groups as a starting point for study, the psychologist focuses on the individual. In a review of the literature on self-concept, Louis Zurcher (19 ) identified four rather distinct ways in which individuals think of themselves. If you would like to try Zurcher's model on some actual data, write the phrase "I am ... " twenty times. Fill in the blanks before reading the next paragraph.

The physical self is our experience of ourselves as a biological organism. If you wrote "I am six feet tall" or "I am hungry", you were describing yourself in this way. The computer/human interface influences this dimension of self-concept by changing our physical activities and physiological experiences. If your statements included a description of your roles ("a student", "a parent", "a vice-president"), your particular social relationships ("Mary's best friend), or your group identifications ("an American"), you were thinking of your social self. This dimension of the self is of most interest to sociologists. The feeling of self-esteem is part of the social self-concept.

In religious or mystical experiences people report an oceanic self that is part of the whole universe. If your sentences described your relationship to the sacred, you were describing this dimension of self concept. Many cultures encourage people to think of themselves in this way; ours tends to view mysticism outside of conventional religious institutions as a form of deviance. An oceanic sense of oneself as having special insight into or power over the universe can be considered medically a symptom of mental illness. It can also considered sacrilegeous by people who believe in God. Because the computer interface can heighten our feelings of connectedness and our feelings of power, some observers think the technology will encourage the oceanic self concept.

There is also a dimension of the self that thinks about its own existence. If you wrote a sentence like: "I am someone who thinks about who I am.", you were experiencing your reflexive self. Self-actualizing people are reflexive. They not only act in satisfying ways; they comprehend the meanings and consequences of their actions. Some psychologists think that, by reflecting on our experience of the computer/human interface, we will change our understanding of what it means to be human.

4.3.1 The Psychology of Stress

The stresses experienced by computer users are psychological as well as physical. One theory proposes that stress has two components: one the physical effects of the human/machine interface on the body; the other the degree of control people have over their activities (Karasek, et. al., 1981). For example, secretaries report more signs of stess (headaches, eyestrain, and back problems) than do their bosses (Balshem, 1984). High stress jobs are those with some physical strain and very little autonomy.

Jobs like air traffic controller are reportedly high stress occupations, but studies have shown air traffic controllers to suffer few physiological symptoms. They seem to enjoy the pace, responsbilities, and status of their jobs. However, they have little voice in decision-making or advancement into management, which is quite authoritarian (Landsbergis, 1985). A study of Silicon Valley found electronics assembly work one of five most stressful occupations in the region, along with air traffic controller, intensive care nurse, police officer, and teacher (Global Electronics Information Newsletter 6, January, 1981:4). Most of these are not physically stressing jobs so much as they are positions of great responsibility with very little discretion.

Managers and individuals can use research findings about stress when they plan work activities. The way that work itself is organized can affect physical strain. Workers with more control over their activities report less physical stress. Stress can be reduced by an organization of tasks that do not require continuous keyboard operation. Long sessions at the keyboard can be broken up. Activities that require standing or walking can be interspersed with sitting. In addition to improving productivity, such arrangements can help prevent chronic and permanent muscular and skeletal damage.

Stress also seems to have a temporal dimension. Research on assembly- line workers has shown that people do not like to work with perfectly regular, clock-tick motions. Instead, people like to be able to control the tempo of their own work. When they are able, assembly-line workers prefer to alternate speeding up and slowing down the line. It gives them a feeling of control over their work, and breaks the monotony.

4.3.2 Response Time

Although human reactions to computer response times are partly psychological, there is some evidence that human beings have a limited range of rhythms to which they can physically adjust. According to Edward Hall, entrainment occurs when people interact with one another. During this process all participants begin to synchronize their speech and motions. Sea chantys and other work songs are some of the cultural aids to entrainment; music and dance are expressions of this social process. If you have ever found yourself getting "into the swing" of some activity, you are probably experiencing entrainment.

Programs which pace the user at a fixed speed have the same effects as assembly lines; they have been most successful in software designed to help people increase their typing speeds by "driving" them to type faster. Because people have chosen the typing software to improve their own skills, the response time demands are a challenge -- not an imposition. When response time was increased in an on-line computer system at work, employees initially reported higher levels of stress (Turner, 1984).

Although much of the research on computer response time is concerned with speeding up data processing, it seems possible that some patterns of human/machine interaction do not suit human beings. In a study at Ogden, Utah, medical researcher Charles Seizler found rotating shift work associated with cardiovascular, sleep, and digestive disorders. As with jet lag, the people in his study had trouble moving their internal clocks forward. In other words, you have more trouble changing your schedule to get up several hours earlier than usual than you do staying up several hours later.

Since interaction with the computer is often interpreted by the user as "conversation", a slow computer can have a psychological effect similar to waiting for a particularly slow speaker to finish a sentence. As an experiment you can try talking to someone with long pauses between all your words and watch their reaction.

Table 5 lists response times found acceptable to computer users. Notice that acceptable response times for actions that are analogous to manipulating objects are shorter than those for conversation-like interactions (for example, waiting for the computer to answer your question). Recent research by IBM has found user productivity increased by as much as 100% when "conversational" response times are decreased from two seconds to a few tenths of a second. Such findings have been criticized by analysts who warn that we need more and better ergonomic research before we rush out to buy expensive new equipment for the sake of a fractional increase in response time (Lyman, et. al., 1985).

We also need to understand what faster response time means to users. We could, for example, look at the study as showing that higher productivity can be obtained by pushing people to work faster. But faster response time may instead represent a change in the way the user thinks about the human/machine interface. Higher productivity may be obtained by thinking of the computer as a tool that does not require the

human conversational tempo. If we think of a computer as a device for turning pages or moving data rather than as a conversational partner from which we expect appropriate pauses, we may handle information much more quickly. Since a few tenths of a second is a reasonable time to wait before the "thing" we push moves, perhaps the shorter interaction times represent an end to thinking of computer interaction as "conversation" and the beginning of thinking about it as a tool. 4.3.3

4.3.3 Conversations with a Computer

Can you really have a conversation with a computer? Or is it a reification of human social relationships to imagine ourselves communicating with a machine instead of with its programmer? People sometimes treat computers, automobiles, teddy bears, and other objects as if they were living beings with wills of their own. Software for children is often encourages this with error messages like: "I DON'T KNOW HOW TO " Even adults can relate to software as if it were a person. 4.3.3 Computer Programs as Counselors
The ELIZA program simulates a Rogerian psychotherapist, chosen because Rogerians try to have the patient take all the initiative in conversation. Using a keyword driven sentence genertor, ELIZA produces sentences like: "TELL ME MORE ABOUT YOUR MOTHER." in response to a user's sentence containing the word "mother". If you typed: "My mother country was Czechloslovakia.", ELIZA would "ask" for more information about your mother. If you typed: "I want to commit matricide.", the version of ELIZA I have wouldn't "know" you wanted to kill your mother, because "matricide" isn't in its keyword dictionary. Joseph Weizenbaum, who created ELIZA as a simple demonstration of an artificial intelligence technique, was appalled by the degree to which people dervied "inter-personal" satisfaction by "conversing" with his program. To social psychologist Sherry Turkle, these converstions provide "the illusion of companionship without the demands of friendship" (1984b). The Future of the Conversational Interface
Artificial intelligence researcher, Richard Bolt, believes that computer interfaces of the future should become even more conversational. He thinks that computers should "know" where users are looking and accept speech and gestures as input, responding to the presence and normal behavior of the human. But Bolt's experimental interfaces that respond when you point or look at them are not conversations in the social sense. Instead, they are projections of one of our deepest fantasies -- to have our wishes become deeds without effort on our part. The idea that a glance or a gesture could have a physical impact on the world satisfies a powerful desire for mastery and control that some people bring to the human/computer interface. Commercial software companies have begun to take advantage of our willingness to talk to computers by selling counseling software on topics ranging from sexual impotence to how to make a better sales pitch (Rice, 1986; "Psycho-computing," 1985). In advertizing, the promise that a new computer system "lets you be the master of your universe" appeals to oceanic fantasies of power.

4.3.4 Children's Ideas and Adult Attitudes

Computer users report a range of attitudes, from excitement and satisfaction, to fear or boredom. "Computer phobia", the fear of using computers, is a rational response to some individual experiences with computer use. Attitude differences with regard to age and sex have been found, and some observers suggest that an obsessive fascination with the computer is a new form of neurotic behavior. From studies of children relating to computers, some researchers have predicted that the computer will radically change our views of ourselves and our world. Mastering the Machine
In Mindstorms: Children and Powerful Ideas, Seymour Papert argues that the computer is an ideal tool for teaching children abstract mathematical and spatial relationships. Papert and his MIT colleagues developed the LOGO language, and with it taught children geometry and programming, as a form of play. With a terminal, they directed the activities of a robot "turtle". Children easily learned to draw by writing turtle programs. The children eagerly taught one another techniques they had discovered, and proceeded on their own to learn more about what the turtle could do; in addition, they invented programs for themselves. The lesson: that computers are controlled by humans (rather than vice versa) was clear.

Even for adults, LOGO is a rewarding introduction to computer programming. It is as simple as the elementary BASIC language yet contains the recursive procedures and list-processing features of higher-level languages like LISP. The new user gets the immediate satisfaction of seeing the turtle (usually a small triangle on the screen) do just what he or she told it to do. More importantly, after playing with LOGO the beginner has a fairly broad sense of what a computer can do.

For the children, LOGO was an introduction to computers that bestowed a sense of mastery over the machine. For them, the computer is a fascinating toy that does all kinds of things after you give it orders. They learned that they can use a computer to act out an imagined design. The experience also taught them the value of making mistakes. Many of their favorite discoveries were made after giving the turtle a "wrong" command. The Second Self
To better understand how children think about themselves in relation to computers, Sherry Turkle observed a children's play group for five years. During that time she recorded their conversations, their interactions with electronic toys, and their interactions with one another. She compared her findings with Piaget's earlier work on how children develop reasoning and moral judgment. In looking at how children understand what it means to be alive and human, Turkle found some interesting differences from Piaget's findings of the 1940's. Children who played with computer toys put more emphasis on feelings and emotions as defining characteristics of human beings and less stress on reason and calculation. Her findings, reported in The Second Self: Computers and the Human Spirit, suggest that computer use does not, as many observers had feared, make people less sensitive to the emotional and interpersonal aspects of being human. If anything, it appears that instrumental rationality (or goal-oriented logical reasoning) may become less important in our culture's definition of "being human". Computer Phobia and Self-Esteem
"Computer phobia" is a popular term for an irrational fear of using the computer. To a psychologist, phobia is more than fear. For instance, my reluctance to go to the dentist is rational if I am afraid of the pain of having my teeth drilled; it is a phobia if the very thought of seeing a dentist makes me tremble, and I can't make myself drive down the street his office is on. From a sociological viewpoint, computer phobia is often a case of "blaming the victim" (Ryan, 1971). This occurs when any possible problems with hardware reliability, software design, user interface, documentation, and real threats to the user's ecomonic status and self-esteem are ignored. Instead, the reluctant user is blamed for being irrational. In the popular "Dilbert" cartoon series, the computer is often depicted as a hostile object threatening human control over situations and interactions. But the solution to "computer phobia" is not to focus on the irrationality of the reluctant user, but to analyze the basis of his or her fears.

A person's experience of interaction with a computer can affect self- esteem in positive ways, for example, a person can be proud of being a good programmer. Computers can also contribute to self-actualization, for example through the satisfaction a programmer gets from making a program LINK COMPUTERS AND SELF-ESTEEM IN CARTOONS work. Or computer experience can have a negative effect on self-worth if the experience makes a person feel incompetent or unfulfilled. For example, people who feel competent and skillful because they are expert typists may feel less satisfied when first learning to use a word processor. Initially, they will be less able to apply their skills to turning out high-quality finished text.

If a person is highly skilled and performs well within an existing organization, new equipment may render his or skills obsolete. This is a threat to job status, and to individual pride and satisfaction. It can also raise real barriers to the successful implementation of computer technology in organizations (Warner, 1984). If the computer is introduced in ways that allow people to preserve their positions and enhance their skills, less "phobia" will result. User-Friendly for Whom?
In some cases, the basis of reluctance to use computers is that they are difficult to learn and people are afraid of looking foolish. The situation calls for good documentation, non- threatening training, and the opportunity for users to see for themselves that the new way is "better". This assumes, of course, that the new way is better. In some cases, in our enthusiasm to computerize our activities, we embark on the process with equipment and procedures that are in some ways less effective. Word processors, for example, are ideal tools for text that must be revised, corrected, and reprinted. They are often not as quick as a typewriter or a pen for a short memo or note. Sometimes the new quipment breaks frequently and cannot be easily fixed, causing very real frustrations. Figure 11 shows the printout from a computerized taxi meter. Although management might dismiss the driver's complaints as computer phobia, the driver maintained that no matter how often the computers were repaired, they did not stand up to potholes.

User-friendly software can be as much of an affront to human self- esteem as systems that are frustratingly difficult. If "the highest expression of the engineering art is to design a machine that can be operated by an idiot", (Noble, 1980) then how are the operators of that machine to feel a sense of status and accomplishment? One solution is to design "natural" user interfaces in which the machine matches the habits of the people. One example of this as an experimental electronic mail system. Its command language was contructed based upon the words chosen by naive users to accomplish mail searching, reading and deleting (Good, et. al., 1984). Another solution is to provide "training wheels" to make software extremely friendly for new users yet allow experienced users to avoid annoying prompts and menus (Carroll and Carrithers, 1984). New multimedia interfaces are being designed for users at different levels of skill and for those who prefer visual explanations (Thomas, et. al., 1985; "Multimedia...", 1985; Mozeico, 1982). In the experimental APEX system, documentation is by demonstration--the user is shown techniques through computer generated motion pictures (Feiner, 1985). Bolt's conversational interfaces are extremely user-friendly.


4.3.5 Sex and the Compulsive Programmer

The sense of mastery, identified by psychologists as an important part of positive attitudes towards computers, has become for some people an obsession. Among many of the adults interviewed by Turkle, a feeling of empowerment contributed positively towards self-esteem. Computer users report feeling satisfaction and a sense of accomplishment from getting the machines carry out their commands. But Joseph Weizenbaum finds that among computer hackers, elaborate fantasies of computer mastery are unrelated to real-world events or issues. The Compulsive Programmer
The compulsive programmer described by Weizenbaum is caught up in the process of programming, and views it as an unending contest between hacker and computer. Working programs are not the goal--compulsive programmers rarely finish programs. Nor is the goal to realize a mental design--hackers' code is marked by patches, poor documentation and bugs, and is rarely planned in advance. The main goal for the compulsive programmer seems to be control over the computer. Not only does programming interfere with social relationships, but seems for compulsive programmers a substitute for interpersonal relationships.

The successful programmer, on the other hand, usually does like to finish projects and see a program actually working. Although the creative professional may put in long hours and spend the last weeks of a project ignoring everything else, the professional lives in a real world with other humans and company deadlines. His or her goal is to accomplish something in the world rather than to live out a fantasy of the grandiose but never- finished computer program. Professional programmers are part of a division of labor. Norms for their roles include communication and coopertion with others (including producing good documentation). While the professional programmer often gets psychological satisfaction from programming, the process itself is not a substitute for living in the social world.

Although many adults view computer hackers as young geniuses, more experienced computer profesionals recognize that, as employees, they are disorganized, difficult to manage, and often incapable of designing or adequately documenting software. When the software industry was beginning, compulsive programmers were often tolerated by companies suffering a shortage of skilled labor. Today, compulsive programmers are less employable. Most companies want reliable men and women who do the work that the company want done. When compulsive-programmer tendencies do crop up in professionals, it is often handled by management as a problem. Some psychologists, like Dr. Steven Berglas, argue that the computer profession does tend to attract rigid, control-oriented personalities. Psychotherapist Craig Brod argues that such "techno-centered" people represent a new type of problem personality. However, psychological tests programmers doing routine work in COBOL found their personalities far from neurotic (Guster, 1985). For even the most creative work, companies have a tendency to insist on real-world relevance. The Hacker as Robin Hood
In a time when the English nobility were converting public forests to private property, Robin Hood became a folk hero for asserting the commoners' right to poach royal deer. His descendants, if he had any, most likely worked in the factories of the Industrial Revolution and struggled for better working conditions. Hackers poaching information in the private data bases of the Computer Revolution insist that information is free. Like the Luddites, some turn to sabotage, as if crashing individual computer systems could halt a social change they find offensive.

Some older programmers see the shrinking opportunities for hackers as one sign that programming as a craft is becoming programming as a set of tasks. Many programmers found in the early home computers a satisfying hobby that compenstated them for the progressive fragmentation of their work. Later, when home computers became more common, the attractions for buyers were this challenge of craftmanship, the sense of mastery over technology, and the hacker's escapism. Today, compulsive programming often takes the form of recreation. In leisure activities, escapism and fantasy do not meet with social disapproval, though complaints from families of compulsive home computer owners and from friends of compulsive computer game-players indicate that "computer addiction" is still considered anti-social by many people.

Occasionally, hacking becomes a more serious form of social protest. A U.S. government employee who was officially reprimanded for making what he condidered a legitimate complaint changed the password on a federal data base and then "forgot" it. It had, he said, something to do with the Declaration of Independence. If they would reread that document they might figure it out (New York Times, Feb. 14, 1986).

One difference between the old community of hackers and activities of the newer cracker groups is that the older group had a hacker ethic based on values from a time before much online information was considered to be for-profit property. Hackers shared with the international community of scientists a belief that knowledge and information belongs freely to anyone who can understand it. Gender Differences in Computer Use
Curiously, most compulsive programmers are male. One way to understand why some young men, and an occasional young woman, become so engrossed with the computer that they neglect their appearance, their social relationships, their school work, and even their physical needs for food and sleep is to consider the process of gender role socialization. Socialization is the process by which children are taught their culture and their place in society. Gender is the term sociologists use for the culturally-defined roles appropriate to each sex. Gender-role socialization is the process through which boys and girls (biological sexual categories) learn to be masculine or feminine (cultural categories).

Great controversy exists over the extent to which our culturally- defined categories are based on real biological differences between men and women are innate, there is actually very little biological basis for many of these differences. In rare cases, children whose sex was misidentified at birth have been raised to have the traits of the opposite gender. At puberty, with the emergence of physical differences, their status was changed, though their personalities remained more appropriate for the gender in which they were raised. Whatever the real differences are, however, the combination of biology and culture produces divergent behavior patterns between boys and girls. Among these are two directly related to computer use: mathematical competence and desire to achieve mastery.

It is important to remember that the facts that few girls become hackers and most girls seem less interested in computers than boys do not indicate biologically-based differences in ability. It may be that most, or even all, of boys' higher scores on tests of mathematical reasoning and higher achievements in engineering, sports, and other fields emphasizing competition and domination are due to the differences in the way children are raised, and social expectations about their behaviors (Caplan, et. al., 1985; Tobias, 19 ; Fox, 19 ). Also, since many girls have exceptional talents along these lines, the findings do not apply to individuals -- they are differences in the average performance of the two groups. The percentages of women who enter mathematically related occupations (12% of U.S. scientists and engineers) or computer science (30% of 1983 bachelors degrees) are much smaller than can be accounted for by lower test scores. Sociologists point to a complicated combination of subtle social pressures and outright discrimination, and blanket definitions of personalities as cooperative rather than competitive, verbal rather than mathematical, and interested in relating to people rather than to objects.

Women are not encouraged to develop control over objects, nor to engage in fantasies of power. It has been found that female computer science majors are more likely than their male counterparts to drop out of school (Campbell and McCabe, 1984). Some researchers have suggested that this is because adolescent girls are taught to avoid demanding situations, while boys are encouraged to deal with them (Fox, 1977; Wolleat, et. al., 1980). Men tend to dominate conversations, while women are expected to respond supportively and tolerate interruptions (Kollock, et. al., 1985). Even in preschool children, psychologist Malcolm Watson finds differing fantasy patterns. Boys tend to imagine fanciful adventures; girls more realistic and domestic dramas. Within the computer industry women are found, in disproportionate numbers, in jobs like documentation (using verbal skills) or technical support (involving personal interactions with customers). Whether because of discrimination ("women's jobs" in the computer industry are lower-paid) or women's preferences for these types of work, the pattern of female computer use is different from that of males. In both the compulsive world of hackers and the more ordinary world of industry, computer power is more often sought and obtained by men.

As computerization expands, such differences may decrease. There has been a trend towards less sex-role stereotyping in U.S. children's books (Kinman and Henderson, 1985). If the computer comes to be viewed as a general-purpose tool applicable to a wide variety of tasks, instead of a mysterious piece of "high technology" over which to obtain mastery, the psychology that confines computers to the male sphere of activity may diminish. Already office automation is producing computer systems designed specifically for a female clerical staff. Micro-computer manufacturers are trying to sell the idea of the computer as a household appliance. Also, women with mathematical talent and an interest in technology are increasingly ignoring traditional social definitions of what they are "supposed to" want to do. A 1985 study of California high school stuents found no differences in male and female attitudes towards computers (Fisher and Pulos, 1985).

4.3.6 The Reflexive User

Reflexive, a scientific word meaning "action based on automatic reflex", is also a philosophical term meaning "reflective or meditative." The term reflexive user is intended to describe a computer user who dvelops automatic habits but keeps the ability to think about what it all means. Despite children's enthusiasm and adult fears about the effects of computers on the job and in the video arcade, most computer users settle into a routine, and cease to think about why they are using a computer, and what the interaction is doing to them. For most of us, the psychological effects of computer use are unconscious. Once the computer's workings are mastered, the machine becomes an ordinary part of life. As in the case of higher productivity through shorter response times, using a computer by automatic reflex can make us more efficient. Acquiring those habits, however, can interfere with our awareness of the social consequences of computers. The reflexive user acquires the reflexes to use computers as tools but does not lose the capacity to reflect on the social meaning of the computer/human interface.

When I first used an automatic teller machine I thought about the jobs of tellers. I wondered if anyone were put out of work by my actions, and in what ways the work would change. My brother-in-law eventually married the teller who cashed his paycheck every Friday--the opportunities for that sort of social interaction might be reduced. Yet I had observed bank customers arguing with tellers and being generally unpleasant, so maybe the loss of social interaction on the job would bring improvements as well as lost opportunities. Then there was the awkward business of following the simple instructions for entering my card and code numbers while people waited in line and I felt slow and incompetent. Was this really a better way to get money out of a bank? Today the card goes in the slot and the numbers are punched without conscious thought. I use the teller machine by reflex; jobs for bank tellers declined by a third. Reflecting on the small nagging guilt at having contributed to the loss of someone's job is part of why I wrote the paper version this book. The web version grew out my experiments with students and colleagues to try to do something about the computer revolution's impact on our own situation.

Individuals experience the computer revolution as a series of minor changes in the world. Each small change is confronted as a temporary disruption in our comfortable patterns. Unless we reflect on our negative experiences with technology, they are soon forgotten or rationalized as part of our inevitably changing society. We adapt to change, alter our habits, and carry on. Unless we are caught up in the popular enthusiasm for the computer (or unless our jobs are eliminated or altered), most of us will notice very little change from day to day. Only when we pause to think over the experiences of our lifetimes, or when we listen to the accounts of older acquaintances, do we see the intersection of biography and history, and realize that we and our culture are changing.

Next Chapter | Main Contents | Chapter Contents