Introduction and Overview
It is becoming well known, as we will document later, that the proportion of the population older than age 65 in the United States is increasing at a rapid rate, that increased life expectancy means that a larger segment of this population is over age 85, and that the great majority of older adults choose to continue to live at home rather than be in assisted living facilities or in nursing homes. In American culture, it is both socially valuable and cost-effective to support the independence of this aging population in as many of the aspects of their lives as possible, whether by supporting their continued employment, independent living and healthcare arrangements, access to transportation, or educational enrichment. Many experts predict that technology can and will play an important role in supporting the independence of older adults, but to date there has been only modest evidence that this potential is being realized. The Workshop on Technology for Adaptive Aging was held to further understand the reasons why more progress has not been made and to stimulate further advances.
Each year the Administration on Aging of the U.S. Department of Health and Human Services publishes a report entitled A Profile of Older Americans. The facts in Box 1-1 are taken from the 2002 edition of that report.
B. Living Arrangements
It is clear that the population of aging (age 65 and older) adults in the United States is entering a period of rapid growth, as baby boomers age and life expectancy increases. The large growth expected in the “oldest old” will pose special challenges.
Given the importance and high costs of healthcare for older adults, applications of technology that could improve health maintenance and healthcare, or reduce the associated costs, would be especially valuable.
Many needs of older adults would seem to be amenable to technological solutions. What has been lacking to date is a coherent effort simultaneously to understand these needs and bring together that understanding with the expertise both to design solutions and to successfully bring them to market. The workshop was an effort to begin to do just that.
D. Health, Healthcare, and Disability
NOTE: Principal sources of data for the Profile of Older Americans are the U.S. Bureau of the Census, the National Center on Health Statistics, and the Bureau of Labor Statistics. The profile incorporates the latest data available but not all items are updated on an annual basis.
SOURCE: Administration on Aging, Profile of Older Americans (2002).
Selection of Topics
At our first meeting, the steering committee discussed several ways of organizing the workshop content and agreed to assign papers based on domains, or areas of function and activity, that we identified as important in the daily life of older adults. Many candidate domains were proposed, and we eventually selected six: communication, employment, health, learning, living environments, and transportation.
The changes people undergo as they age, specifically cognitive and sensory changes and changes in motor performance, were seen as applicable to all of the domains, so we decided to assign these as “overview” papers, eliminating the need for each domain-specific paper to address
them. In addition, we decided to assign a paper on methodological issues in research on aging and technology, another topic applicable to all specific research areas.
There is a great deal of excellent research on technology applications for older adults being performed in Europe, Asia, and elsewhere. However, the limited resources for the workshop led the steering committee to select mostly American researchers as authors, and thus the work of researchers in the United States is emphasized in this report. Research conducted in other countries is frequently cited in the workshop papers, and some authors have worked with foreign scientists, both here and abroad.
KEY ISSUES FOR THE ACCEPTANCE OF TECHNOLOGIES
As the steering committee continued to explore the topics and issues to be addressed in a workshop, we identified a series of issues that, if not addressed effectively, might pose barriers to the acceptance and therefore the success of a technology in the marketplace. First we considered a series of potential barriers to acceptance of the technology from the perspective of the aging population themselves. These are issues that could affect potential users′ motivation or willingness to try or use a technology, their ability to use it effectively and be satisfied with the results, or other aspects of their interaction with technologies. All workshop authors were asked to address these issues in their papers to the extent that they were relevant. This discussion of potential barriers is not intended to discourage investments in technology, but rather to highlight for innovators not familiar with the literature on the aging population the interesting challenges that are there to be conquered.
Control, Autonomy, Agency, Dignity
One of this country′s founding principles is the primacy of individual liberty, the inalienable right of the individual to be free to choose. Embedded in this notion is the belief that individuals can, in fact, make independent choices, that we know our own minds, and that we can initiate effective action to bring about a desired outcome without directly harming others. In the arena of aging, these issues are complicated by the onset of certain diseases, such as dementias, and by the life-and-death context in which certain types of decisions are made. Technology will not be used or accepted if it does not respect these individual rights and allow a person to age with dignity.
Universal design, that is, the design of products and environments to be usable by all people to the greatest extent possible without the need for adaptation or specialized design, is now a national goal. Universal design incorporates the concept of customization in the sense that it specifies design requirements that allow adjustment of information presentation and response modes to accommodate those with differing abilities and preferences. However, there are also individual differences among the aging population that transcend universal design criteria. In the field of aging especially, universal design, while important, is not sufficient because there are many special requirements. Large print does not help the individual with impaired cognitive function who has difficulty remembering the first part of a sentence coherently with the concluding part. Aids for this limitation would be disruptive for more cognitively normal individuals. We need to understand how to take account of the range of abilities, needs, and desires of older users to ensure that the technology can adapt to their individual differences.
Culture and Language
Applications of new technologies are designed to appeal to potential users. They will not be accepted if they confuse or alienate individuals on the basis of their cultural or language norms. Designs, descriptions, written instructions, and advertisements, for example, present the technology in a particular way. Depending on the cultural diversity of the potential user group, an approach that may appeal to one subgroup may discourage or turn off another. Thus it may be imperative to consider different designs and approaches for different cultural contexts.
Expectations and Stereotyping
As people age, their expectations of themselves and others may change. These expectations can present major barriers to the acceptance of new technologies. Recent literature has suggested that older adults both stereotype themselves and are stereotyped by others (e.g., with respect to speed of movement, memory, sensory function) and that this stereotyping can often be detrimental to their motivation or performance. There are great individual differences in how people adjust their expectations and in the affect associated with these changes. Also, each individual brings to a technology unique expectations for its performance based on knowledge, beliefs, and experience with that and similar technologies. Expectations need to inform the design and application
of technological solutions and the way in which they are marketed and introduced to potential users.
Privacy issues have been a focus of growing concern as new information technologies gain more widespread use. Whenever use of a technology involves communicating information of a personal nature, it directly affects the acceptance of a technology. Apprehension that private information (e.g., health, financial, or personal data) may be captured by persons or organizations for uses other than those intended by the user, without the user′s knowledge or consent, is one of the unintended consequences of improving the flow of information necessary for sound decisions. For example, a technological application, such as a medical monitoring system, may require that users provide information they regard as sensitive and would rather not share. Systems need to be developed, applied, and marketed with the assurance that an appropriate level (and perception) of privacy will be maintained. The factors that influence individual privacy concerns also need to be better understood.
It seems obvious that no technology should be made available to seniors if it presents an unreasonable risk of danger or harm to its user, whether in normal operation or as a result of the failure or malfunction of a technology in use. From the perspective of the user, even the perception of risk presents a barrier to acceptance. Nevertheless, some technologies that are now being made available to consumers could pose safety problems, especially for older users. One example is the attention distraction implicit in the proliferation of information displays (navigation aids, restaurant listings, Internet access) now being introduced in automobiles. Issues of safety in design must be addressed as well as methods of conveying hazards to older users through instruction and warning systems.
How much training is required to make use of a technology? If it is excessive, the technology may be improperly used or not used at all. Thoughtful design of training is especially important for older adults. Good training motivates and empowers people to use technology; poor training may discourage them. Training must be appropriate to its audience′s knowledge, skills, and abilities, as well as to what the audience wants and needs to learn. With a DVD player a failure to do this may be
only an annoyance, but with a glucose monitoring system it can have life-threatening consequences.
There are special considerations in designing appropriate instructional materials and programs for this audience that go beyond universal design to take account of perceptual and cognitive aging. It is particularly important to minimize the need for dividing attention across instructional materials, make appropriate use of memory aids so that working memory is not overburdened, and provide fundamental knowledge that older users may lack.
The degree of trust the older person places in another person, a device, a service, a procedure, or a system may be measured by expressed or observed satisfaction, acceptance, or willingness to use. Typically, trust increases as individuals accumulate exposure, familiarity, and a history of successful use of a technology. However, trust can be weakened or destroyed by unreliability, unpredictability, or excessive complexity. Distrust may lead to lack of acceptance or reliance on the technology. There is also the potential for “over-trust,” or an inappropriate dependence on technology, which could have serious consequences if the technology fails.
There are countless examples in which a lack of product usability has led to a failure of a product in the marketplace. The barriers to successful usability among older adults are even higher than for the general population. Many factors (e.g., preference, ease and speed of learning, success probability when experienced) are important in determining the ease with which users can learn and use a product. Training, usability, and customization are mutually interactive. Improved usability results in reduced need for training, while additional customization may make effective usability more challenging and require more extensive training. Keeping technological supports simple and easy to use should be integral to meeting the needs of the aging population.
ISSUES FOR THE SUCCESSFUL INTRODUCTION OF NEW TECHNOLOGIES
In addition to potential barriers from the user′s perspective, many other considerations can limit the likelihood that new technologies will successfully make the transition from research into marketable products. Here we consider the practical and economic issues that are potential barriers to the introduction of new technology in this domain.
Technology Transfer from Research to Manufacturing
Researchers who have ideas for useful products for older adults and who develop a product through the prototype stage may underestimate the time and cost to bring a new idea to market. Having a good idea that “works” at the laboratory prototype stage is only the very beginning of a progression of efforts necessary to get a product into the hands of a user at an affordable price. These costs include the need to engineer for reliability and safety as well as for manufacturability. If one starts with a laboratory prototype, the development path will be somewhat different depending on the extent to which novel hardware is involved.
Consider first a product that is primarily software. After user testing and refinement of the prototype, there will be a point at which the decision to actually produce a product will be made. This will be based on input and analysis from the marketing and sales organizations, from the development team who must estimate the development cost and schedule, and from a variety of other sources. Factors to be considered are reliability, safety, production cost, potential for market penetration, and so on. The outcome of this stage is not only the go-no-go decision, but also definitions of the target market, a concept of operation, a financial pro forma (estimating the magnitude of the return on investment as a function of time), hardware and operating systems to be supported, production software language, target performance specifications, and initial requirements documents that specify what features will and will not be included.
The next stage is detailed design of the software during which the detailed requirements are fleshed out, an architecture defined, and software specifications written. When this stage is complete, the actual coding can begin and, ideally in parallel with software development, technical documentation should be developed. However, the process is far from complete when the coding is done.
A lengthy process of testing ensues, including repeatedly exploring each aspect of the functionality to ensure that it is operating as intended and then testing the product and the documentation to ensure that they meet usability, performance, and safety goals. Usability testing is especially important when the market is the aging population. Product testing concludes with a phase of “beta testing,” in which the product is released selectively to potential market segments for prerelease evaluation. Mean-while, marketing and sales are operating in parallel with the developers to formulate the distribution, marketing, and advertising plans.
When the product involves significant manufactured hardware, many of the same steps are accomplished, but there are added decisions about produceability and ease of manufacturing as well as the specification and design of the entire manufacturing process, including the design and
construction of production and assembly lines. There is an entire design and development process associated with manufacture that is accomplished in concert with the actual product development process.
Reflection on this complex of processes reveals just what a significant decision it is to begin product development, because it sets in motion a very large commitment to follow through to successful deployment. It illustrates why there have to be large segments of the aging population for whom the product is attractive before a new product decision is undertaken. At the same time, we should not lose sight of the fact that future generations will be much more computer savvy than today′s older adults, and they may present a considerably larger and more stable market for technological support as they reach senior status.
If there are liability issues associated with the use of a given technology, they will have serious consequences for the willingness of manufacturers to offer the technology, as well as for user acceptance. These issues may affect the manner in which a technological solution should be designed, used, marketed, and supported. For example, what are the implications of an “easy-to-use” glucose monitor that is used improperly and gives erroneous readings, leading to adverse health effects?
The Market Is a Moving Target
Gerontologists use the term cohort to refer to a birth cohort, those people born at approximately the same historical time. Older people in a particular cohort may have a different reaction to technological innovation because they have had different opportunities to learn about and use various technologies. For example, those currently over age 70 may have completed their working years without ever having learned to use computers or a cellular phone. Those now about to retire have had very different experiences, and one can expect that future cohorts will have still different experiences and perspectives to consider. A product developer needs to take account of the fact that the target audience is not static, but changing as new cohorts move into senior status.
Defining Products That Can Be Customized
In addition to changes with cohort, as noted above customization may be needed to overcome barriers to individual differences in the acceptance of technology. This need for customization may become a barrier to the development and marketing of products for an older market, because highly adapt-
able products may be more expensive to design and manufacture than one-size-fits-all products. Some products and systems are likely to require skilled personnel to install or prepare them for use by the individual consumer and to periodically readjust them to the customer′s changing needs.
Economics of the Marketplace
Manufacturers are loath to begin development of a new product unless there is a clear and sufficiently large market because development costs must be underwritten by and amortized over projected sales. All the planning and other costs of the marketing itself, including advertising, packaging, instructional literature, distribution, warranties, and product liability, must be recovered. There is a role for the government in supporting some of these development costs in the interest of seeing desirable products reach the marketplace.
Sources of Funding
The discussion of funding at the workshop focused mainly on sources of funding associated with medical and health-related technologies. Many products will find a ready market among seniors. However, there are also technology products that meet unique needs, but for which the relatively small market either will not justify the required commercial investment or will result in unrealistic cost to the consumer. There are basically four alternative ways to underwrite these investment costs: (1) The government may directly contribute to the development costs through grants or contracts. (2) The government may subsidize the cost to the consumer through programs such as Medicare and Medicaid. (3) Private health insurance companies may justify reimbursing the cost by demonstrating reduced costs for hospitalization or reduced requirements for medication. (4) In a limited number of cases, the individual consumers may be able to afford the technology.
This issue was discussed at the workshop by Jonathan Skinner of Dartmouth College. If the government reimbursed for such devices, there would be great industry interest. The government can use the argument that, even if a product or service is not profitable, it should be supported because it saves lives. However, government reimbursement for medical and assistive devices is very limited, and many of the innovations being discussed for future development will cost much more than those available today. If use of new devices in home healthcare reduced the incidence or cost of hospital visits, both the government and private health insurance providers could justify covering the cost, but, according to Skinner, so far there has been no study completed that indicates that home
healthcare actually does reduce hospital visits. In summary, there is a need to evaluate technology in comprehensive studies that examine the dollar costs, the impact on other types of healthcare costs, and the value to patients.
STEERING COMMITTEE COMMENTARY ON THE WORKSHOP
The body of the workshop proceedings consists of nine papers that address the role of technology in supporting individuals as they adapt to the changing needs and abilities associated with aging and in ameliorating the barriers to acceptance of the technology from as broad a perspective as possible. The steering committee′s commentary on each paper reflects the discussion at the workshop. For references supporting this material, the reader is referred to the papers themselves. First we consider the three invited papers that provide a background for the discussion of the role of technology.
Changes with Aging
The authors of the first two papers provide a summary of what is known about the life-span effects of aging on sensory, perceptual, and motor performance and on cognitive functioning and the ability to operate in a world involving social interaction, in which individual intentions, the demands of multiple external tasks, and feedback come together to produce behavior.
The paper on “Cognitive Aging” by K. Warner Schaie and the paper on “Movement Control in the Older Adult” by Caroline J. Ketcham and George E. Stelmach provide the theoretical and empirical context for the current knowledge and future projections that appear in the other papers. As in any research arena, there are gaps in the knowledge base concerning changes across the adult life span in perception, cognition, and motor control. Several of these “knowledge gaps” merit brief comment.
First, although knowledge of age-related changes in circumscribed cognitive processes (e.g., attention, working memory, perceptual speed) is sufficient, in many cases, for application to the design and evaluation of technology aids, there is a need for research that examines the manner in which different processes interact across the adult life span (e.g., the manner in which attention influences memory and vice versa). Clearly, additional research is required to explore potential changes in these important interactions and the manner in which they impact the design of technologies and technology aids for older adults.
Second, the great majority of current knowledge about cognitive aging has been obtained in well-controlled but relatively impoverished laboratory tasks and environments. However, technological solutions to age-related changes in cognition will be realized in complex environments, making it necessary to determine whether the laboratory-based research findings “scale-up” to the kinds of environments in which people live their lives. This scaling-up process is clearly beginning to occur to a limited extent in the human factors community and must increase in the near future.
Third, as is evident from the Schaie and the Ketcham and Stelmach papers, researchers already know a great deal about cognitive and motor changes across the life span. This information has, in turn, been used to design a number of different interventions that have shown promising improvements in a variety of perceptual, cognitive, and motor processes of older adults. Interventions have, thus far, included intellectual training, physical fitness training, and nutritional interventions. Clearly this theoretically based trend in interventions needs to continue and expand with the use of new technologies that enable the measurement of behavior and inferences about cognition in complex tasks and environments. Research is also needed on the mechanisms that underlie the beneficial effects of such interventions, the impact of multiple coordinated interventions on the cognitive vitality of older adults, and the ways in which technology can support the beneficial effects of these interventions.
Fourth, in recent years the study of changes in human brain physiology and structure has blossomed with the development of new neuro-imaging technologies, such as positron emission tomography, functional magnetic resonance imaging, and near-infrared optical imaging. These technological developments have enabled scientists to understand, in a way not previously possible, the changes in brain function and structure that underlie changes in cognition across the adult life span. We believe that the information garnered from the application of this technology in theoretically driven studies is crucial to the development and extension of the understanding of the impact of current and future technologies on the cognitive vitality of older adults.
Methodological and Measurement Considerations
A number of questions that focus on ways to develop scientific evidence to support the efficacy of new technologies are addressed in the methodology paper by Christopher Hertzog and Leah Light. People today expect technology to compensate for their frailties and offer them improvement in their circumstances or, at the very least, a slowing in the
rate of decline. But which of these outcomes is the most appropriate, and how is that choice determined? It is not difficult to think of ways in which improvement should be self-evident. People can do things they could not do before. They can do things better, more quickly, more easily, with less pain. They can do them by themselves. But to determine whether the rate of functional decline is slowing requires choosing a point of comparison other than one′s own prior behavior or experience. The question “How would I be doing if I did not have the use of this technology?” is answerable through a controlled comparison to someone else with like limitations but without the technology. Comparable to clinical trials for new medical treatments, a technological placebo is required.
But have scientists developed and tested the necessary measures not only to assess the impact of a specific technology, but also to compare across technologies? One of the measures most often used to assess technologies is speed—can the person do things faster than before? Whereas speed has been emphasized as an outcome in evaluating technologies relevant to the workplace, it is less clear how speed should be weighed as one outcome among many. For example, if use of a technology does not increase the speed of production but does reduce the wear and tear on workers, produce less stress, or reduce certain types of injuries, how do researchers assess the overall impact of its use? If they rely on a cost-benefit analysis, how do they assess all costs and benefits within a single metric, especially when either the costs or the benefits may occur over a protracted time frame?
To the extent that measures of accuracy are used, researchers too often rely on the summary measure—was the response correct or incorrect?—rather than analyzing the process that led to the outcome and determining where in that process the sequence of events may have departed from what it should have been. When looking at measures of overall satisfaction or quality of life, researchers often rely on measures that are likely to be confounded with more generally attributable sources of contentment or discontent, or on measures that are relatively crude, making it difficult to determine whether any meaningful change did, in fact, occur. Finally, when evaluating technology relative to the goal of enhanced autonomy, are researchers also assessing how the replacement of human contact by technological capability may impact other areas of a person′s life?
Clearly, adaptive technologies—those that accommodate a variety of user characteristics and often compensate for users′ disabilities—offer some advantages. By extending the range of human performance, they allow older adults to reintegrate into their daily regimens routine tasks that have become too difficult for them. But the promise of independence involves a caution, because dependence is not being eliminated but rather
redistributed among people and technology. Users may still need to depend on the reliability of the technology and on the people who service, repair, and update that technology. As this kind of technology is introduced, it is important to avoid severing the personal relationships that are important to the older population.
In the attempt to assess the impact of technology on people′s lives, scientists must move beyond an individualistic research design that looks at the ramifications of technology use in isolation. They must be sensitive to the contextual factors of both environment and personal history. Whereas experimental studies allow the most leverage in assessing causal relationships, the limitations placed on generalization beyond the experimental design are troubling. People will be relying on these technologies in circumstances subject to far less control than a laboratory, and although it is possible to gain important insight from randomized experimental designs, experimental frameworks do not adequately substitute for more naturalistic assessments that allow the variability in environmental conditions and the influence of other persons to be considered. Furthermore, researchers must be sure that they are defining the unit of analysis appropriately. Are they assessing the impact on the individual? On the individual and the spouse? On the individual and the caregiver? People live in a complex social and environmental milieu. Therefore, part of the assessment should involve the impact of technology on these relationships and the possibility that technology can reshape and restructure these relationships in both positive and negative ways.
Researchers should also build dynamic assessment into their framework, allowing them to study the process of adaptation to a new technology. The impact of the technology is likely to develop with time, and these developmental processes may proceed at different rates and in many different ways, depending on the purpose of the technology, the individual using it, and the social and environmental context into which it has been placed. Dynamic analyses are more demanding in terms of measurement, data collection, and the skill level of the analyst; they can be more intrusive for the user; and they introduce additional complications into the sifting of information into categories of “error” and “real change.”
Some adaptive technologies may allow the incorporation of monitoring capability directly into the device itself. Combining technological tools and measurement technology in the same device can produce a wealth of timed data that will track all sorts of indicators in real time. This wealth of information, however, requires analytic tools that are not currently well developed in the sciences. Other sciences are developing new analytic methods to work with new kinds of information, and behavioral scientists may benefit from advances made in these apparently unrelated fields.
The information flow in research on aging and technology is information about users as they live their lives. Quantum particles and genetic materials have no claim to privacy; people do. American society has greatly expanded its observations of individuals in public places in the interest of security. Will researchers also monitor people′s activities in their own homes or in care facilities? How do researchers minimize the risks that can result from the failure of the technology without unaccept-able intrusion into people′s privacy? How do they understand and measure the construct of privacy itself?
Finally, the development of technology through private-sector research and design operations is inevitably linked to market dynamics. The size of the potential market and whether the device is affordable within that market are key issues that must be addressed in an encouraging way. The potential for technology to provide significant benefits to those with access to it raises ethical questions regarding the role of government in both the development and the distribution of these technologies. Should research into particular adaptive technologies be discouraged because those who need the innovation are a relatively small group, or because a larger group who could benefit from the advance cannot afford to purchase it? To what extent should market dynamics be allowed to shape development and access? Clearly, devices aimed at lucrative markets will be developed in any case. But the funding of research into technologies with potential applications to disadvantaged individuals also increases the upward cost pressure on government programs to make these technologies available on the basis of need. What is society willing to pay to provide wider access to these technologies? These are all questions that need to be discussed as scientists move forward with projects designed to build the knowledge base required to make these advancements a reality.
We now discuss the papers that address the potential opportunities and impacts of a wide range of technologies on a spectrum of practical domains that the steering committee deemed important to the aging population. The topics are discussed in the order they appear in this volume.
Susan Kemper and Jose Lacal address the communication needs of older adults and emerging technical solutions for those needs. Communication may be between people, from a person to a system, or from a system to a person. Person-to-person communication is one of the most common methods of transmitting information, whether by face-to-face
conversation, by a telephone or other telecommunication method; through writing on paper, electronic mail, or text messaging; or through symbolic means with icons, symbols, or pictograms.
Communication is also a key component of technology use—the user has to tell a system what to do (e.g., make the house warmer) or the system has to tell the user something (e.g., blood glucose level is in the normal range). Person-system communication may consist of spoken messages, button presses, or text entry, and the system may communicate back by spoken, text, or symbolic messages.
In addition to basic activities of daily living (e.g., bathing and toileting), successful independent living requires the capability to carry out instrumental activities of daily living such as managing a medication regimen, maintaining the household, and preparing adequately nutritious meals. Such instrumental activities of daily living are dependent on successful communication between people and/or systems. In addition, many older adults engage in what have been referred to as enhanced activities of daily living, involving the ability and willingness to adapt to changing environments, to accept new challenges, and to engage in lifelong learning. These enhanced activities of daily living may be viewed more as luxuries than necessities, but they may be key to staying fully functional and maintaining a high quality of life. Such activities also involve communication at various levels.
Older adults may have communication difficulties because of age-related changes in perceptual, cognitive, and motor function: Text is too small. Illumination is insufficient. Hearing loss limits speech understanding. Cognitive deficits, combined with degradation in perceptual-motor skill, limit access to people and events both in the immediate environment and in the wider world. At the same, time new technology, such as miniaturized and inexpensive sensors, computers, and high-bandwidth technology, has great potential to enhance the communication of older adults. This can take the form of two-way video and audio communication with loved ones or transportation providers; on-line shopping, education, or entertainment through the Internet; or automatic monitoring of location by security agencies and of health parameters by hospitals or physicians. In this paper the authors discuss the most critical problems in communication for older adults and the technologies currently available or on the horizon, both specific products and infrastructure, that show the most promise to cope with these problems. The authors consider the actions required to allow older adults to access, understand, accept, and trust these remedies so as to improve their quality of life.
For communication technologies to truly be successful in meeting the needs of older adults they must be developed with consideration for the cultural norms of the potential user group and an understanding of the
fundamental age-related barriers to communication. For example, speech comprehension may be improved somewhat by making the message louder and slower. However, as the authors of the communication paper argue, such changes will be insufficient to ensure the success of communication technologies. Researchers on cognitive aging have shown that there are other important variables that are relevant to the design of speech communication systems, such as prosody, working memory, and the pros and cons of “elderspeak.” The practical relevance of these findings must be conveyed to technology designers.
This domain exemplifies the importance of translational research—designers have to know what the researchers on cognitive aging know, and the researchers should know more about the trends in technology design and be able to provide guidance about capabilities and limitations of older adults. Prototype technologies must be tested with the target user population, who may be suffering from hearing deficits, working memory declines, attention deficits, and arthritis. In addition, the specific communication needs of older adults can serve as the impetus for development, provided that designers are informed of these needs.
The two papers on learning and employment, out of necessity, examine a younger elder population in their 50s, 60s, and 70s. In particular, Sara Czaja and Phyllis Moen, in their paper on employment, examine the “younger old,” a vast majority of whom are fully able to carry out the responsibilities associated with an active lifestyle, but who traditionally have exited from the work force. Retirement in America, although no longer mandatory, begins at an age as early as the late 50s, is prevalent at age 62, and is the norm by age 65. In this paper the authors focus on this younger old group.
Many of the other papers in this volume examine the realities of aging today based on research involving persons born in the 1910s, 1920s, and, to a lesser extent, the 1930s. As researchers look at the issues associated with work and the influence of technology on people′s working lives, they will be looking at very different populations, those born in the 1940s and 1950s, with different life experiences and expectations than those born in the earlier part of the twentieth century. America′s baby boom generation—those born between 1946 and 1964—are today′s older workers approaching the traditional retirement age.
Many scholars and journalists have pointed out that the events and circumstances surrounding the establishment of retirement as an institution have changed or are about to change—its timing, its finality, and its all-or-none nature. Sociologist Matilda White Riley noted that often there
is a structural lag that occurs between society′s institutions and its evolving culture, forcing the institutions to play catch-up to the social behaviors. Retirement—formalized in the late nineteenth century and institutionalized with the passage of the Social Security Act in 1935—may be one of those institutions lagging behind social behavior. The paper on employment and technology acknowledges this impending change.
One of the many factors in the mix of influences on tomorrow′s older working and retirement population is the pervasiveness of technology. Technology has already altered the use of time at work and the nature of labor; it has influenced the ability to respond to both acute and chronic diseases and has promoted lifestyle adjustments to accommodate changing bodies. And it will continue to do so. For the baby boom population weaned on technology, the capacity to ride the crest of change seems particularly evident. Although aspects of this change that is swiftly approaching have been discussed here and elsewhere, the combined effect of multiple factors is significant and has been referred to collectively by steering committee member Scott Bass as the “perfect storm.” Like the perfect storm in nature, the entrance of the baby boom generation into later adulthood presents a confluence of independent variables playing off one another in perfect but dramatic harmony. This fact is highlighted early in the volume to point out that both aging and developments in technology are moving targets whose developments are rapidly evolving and changing. Figure 1-1 identifies 12 factors contributing to the upcoming dramatic change, and discussed below:
Cohort differences and attitudes toward work and leisure: In national surveys, baby boomers have indicated that they want to work after traditional retirement ages. About half have indicated that they need to work for the income and the other half indicated they wanted to work for the social contact and stimulation. Most said they are interested in part-time or seasonal work.
Removal of some of the disincentives to work after age 65: Historically, public policy encouraged those age 65 and older not to work, thereby making room for younger workers. Pensions and Social Security benefits were weighted to provide incentives for early retirement. For the most part, these incentives have been removed to make selection of the age of retirement neutral. In addition, Social Security earnings penalties for older workers have been reduced or eliminated.
Demand by employers for skilled workers: As productivity has become associated with working smarter, more efficiently, and drawing on new technologies, experienced and skilled workers are in demand. In selected industries, there are shortages of highly skilled workers. A scarcity of well-educated, skilled workers of any age sought by employers who are losing them to retirement would be unfortunate.
Potential labor shortages: Labor shortages are cyclical and are influenced by the overall economy. In times of economic expansion, even nontraditional labor markets are recruited, including retired workers. However, even in projections of most growth in the gross domestic product (around 3 percent per year), the demographics, which include smaller cohorts of skilled younger workers, point to selective labor shortages. Even in times of recession, there are sectors with labor shortages, such as nursing and teaching.
Concerns about pensions and Social Security: With the increase in longevity, older people will need resources to support themselves. Many are concerned that Social Security will be too little and too precarious to support their needs. In addition, pensions, particularly those invested in stocks, are subject to the ups and downs of the larger economy, possibly forcing older workers to consider working longer and later in life.
Changes in perceptions of older workers by employers: At one time, human resource managers voiced negative stereotypes of older workers. Recent surveys have shown that many of those negative images have faded and are being replaced by positive images of reliable and conscientious employees.
Health of the older population: Successive cohorts of older populations are healthier and less impaired than previous cohorts. Most older individuals of around age 65 are active and capable of full-time work.
Educational background: Younger cohorts of aging adults in the United States reflect higher educational attainment. Lifelong learning, once an
unrealized concept, has become the expectation in the changing workplace. Strong basic skills are needed as a foundation for learning; new and novel skills are a consequence of new technologies and changing work patterns. Today′s older workers are better prepared for lifelong learning and on-the-job retraining.
Consumer activity: The baby boomers have been robust consumers, setting trends and developing habits driven by consumption of quality goods and services. To support this habit as they age, baby boomers will need sufficient income to support the lifestyle to which they have become accustomed. This factor alone is a salient one in considering the need for income maximization among aging boomers.
Changes in the nature of work itself: Historically, labor relied on manual effort, was standardized, routine, and hierarchical, involved sequential decision making, necessitated high volume, and relied on basic skills. The nature of work itself has changed. The economy of the twenty-first century is much more dependent on services. With customization and attention to detail that characterize transactions, collaborative planning and teamwork are needed, quality and high productivity are rewarded, and the workplace draws on higher order skills, such as problem solving, planning, innovation, and process efficiencies. It is these skills of the modern economy in which experience and maturity are assets.
Shift in labor from industrial to service sectors: America was an industrial giant in the first half of the twentieth century. Associated with industrial might were environmental hazards, occupational risks, and costs of unionized labor. Today, industrial production has migrated to nations with lower labor costs and the capacity to make goods cheaply and efficiently. American strength now lies in the service sector, providing for the management of goods and services to customers. In labor involving brawn, older workers were disadvantaged; in labor involving services, older workers are competitive with younger counterparts.
New technologies: New technologies have changed the way information is organized, cataloged, and disseminated. They have influenced the way organizations do business and the way workers interact. Large numbers of baby boomers are computer literate and are adaptable to ubiquitous computing environments. In addition, new technologies have provided greater access to mainstream occupations for people with disabilities and have given those with debilitating diseases or infirmities that might once have triggered a premature retirement decision the ability to continue or return to work.
Czaja and Moen′s chapter examines the dynamic and changing environment surrounding work, technology, and cultural expectations. Their recommendations point out that much of the literature about aging and human performance has focused on cohorts who were born in the 1920s
and 1930s and whose performance has been measured in laboratory studies at the upper boundaries of retirement age. The experiences of and attitudes toward work and retirement for those growing up in the Great Depression or during World War II are quite different from those of the emerging Vietnam-immersed baby boom cohort. Studies examining human performance in a laboratory setting of those in their 70s and 80s may be different from observing people in their 50s and 60s in their natural work setting, in which they have made adjustments to their work and the workplace has made adjustments for them. As a result, Czaja and Moen argue that there is “limited empirical data on the practical implications of aging for work activities.” It is to this end that they make a recommendation for a research program to better understand the performance of current cohorts of older workers in actual working environments, rather than performance on laboratory-based tests.
A second area of research that Czaja and Moen call to our attention is a more detailed understanding of gradual age-related change in abilities associated with aging and the specific skill requirements of jobs in different sectors. Longitudinal research in this area would provide detailed insight into the concomitant effects of health, ability (at different ages), work force expectations and actual performance, influences of technology, uses of technology, changes in the workplace and jobs, and a variety of different employment sectors and their effects on different age, ethnicity, and gender subgroups over time.
The paper by Eric Dishman, Judith Matthews, and Jacqueline DunbarJacob discusses research in applications of technology, especially information technology, to health and healthcare. Older adults represent the largest consumers of healthcare resources in the United States. They are also the largest users of assistive devices, with more than half of all community-residing individuals age 85 and over using one or more devices. Inasmuch as health and functioning are key components of quality of life, technologies that address these concerns have the potential of significantly improving the quality of life of older individuals. The authors discuss current and emergent technologies designed to maintain and enhance the health of older adults. Examples of health-related technologies include monitoring devices, decision support systems, emergency response systems, sources of health-related information for consumers, as well as technologies designed to enhance patient and institutional compliance with treatment and lifestyle regimens. Several challenges face those who are developing new technological solutions for health and wellness.
Information overload. One of the hallmarks of current and emerging
technology is its ability to provide vast amounts of information. For example, sensors and monitoring devices can inform caregivers or others about a person′s location, what he or she is doing, and even their biological and physical functioning. Such information can be very useful in helping older patients evaluate their own status or in providing a relative or healthcare professional with assurances that all is well. However, information also has the potential for creating anxiety, distress, guilt, or, alternatively, a false sense of security. There may be too much information for an individual to process, leading to anxiety and distress. The information may be ambiguous or inconsistent, making it difficult to interpret, and it may even be unreliable. From a social perspective, the mere availability of monitoring technology may put pressure on older persons to be monitored when they wish not to be, and it can put pressure on family members to monitor their relatives in ways they may not wish to be monitored.
Need for data reduction and integration. The technical capacity now exists to generate vast amounts of data about older persons and their environment, but having data is not the same as having useful information. Success in developing hardware to generate data is not matched by a corresponding ability to convert data to useful information. In other words, the capacity of current hardware far exceeds the capacity of the available software. Thus, much more effort is needed in developing the means to reduce and integrate data to make it useful in decision making. Moreover, the data reduction and integration will vary for different users. An older person with some functional impairment is likely to need more data reduction and decision support than a family member or a healthcare professional.
Ethical, liability, and acceptability issues. For technology to be useful, it should facilitate decision making and ultimately impact behavior. This raises questions about who is responsible for wrong decisions or actions, particularly if they are the result of a technological failure.
The ubiquitous presence of sensing and monitoring devices has the potential of undermining privacy. Individuals may be monitored without their consent or awareness. Although this already happens in public spaces, what are the implications of being monitored in private spaces such as someone′s home or office? A challenge is to develop technologies that also provide feedback to the person about what is being monitored and when.
New technology is often expensive. There are already large socioeconomic disparities in the health and well-being of older persons. Will the availability of useful but costly technology further widen the gap between people who do and do not have access to this resource?
Technology has the potential of undermining functioning in older persons. Most older individuals are very resilient and are able to develop compensatory adaptive strategies for dealing with life challenges. Technology that provides too much support has the potential of eroding these
adaptive capacities. Successful technological supports will maximize people′s independent functioning and perhaps sense and adapt to changes over time in their needs and abilities.
Learning will continue to play a vital role in the lives of tomorrow′s elders. As described in the paper by Sherry Willis, there are three general arenas in which this learning will be carried out. First, having considerable leisure time as well as a desire for knowledge and information, older adults will be learning informally about matters, such as health and travel, that are important to their everyday lives. Second, older learners are expected to need training and retraining that will enhance their employability and work efficiency. Finally, and overlapping both of these domains, along with the rest of the population, older adults will be learning to use technologies to support a variety of activities, including learning about other matters.
The diverse learning needs of older adults make it likely that the nature of learning itself will be quite variable. Traditional, highly structured learning is exemplified by the numerous formal courses and training programs available on the World Wide Web. A very different model is the spontaneously organized, nonhierarchical community of learners, who meet virtually or in person to pursue a common interest or problem, often without a designated teacher. This latter approach to learning is often more motivating and effective. However, a suggestion that traditional learning environments should be eliminated ignores relations between task demands, learner characteristics, and the learning environment. That is, a student-centered approach to learning may be ideal for highly motivated and cognitively robust individuals who have general learning goals that emphasize problem solving over the performance of specific and concrete tasks. This same approach may not work at all for teaching entry-level computer skills to those with cognitive deficits and no background knowledge. The match among task demands, environmental affordances, and individual abilities is especially important when discussion centers on technology and the older learner.
There have been some remarkable recent efforts to clearly delineate the physical, perceptual, and cognitive requirements of technology-mediated tasks, such as banking and blood glucose monitoring. Similar research needs to be carried out on the use of the Internet, synchronous and asynchronous learning applications (e.g., NetMeeting, Vclass, and CentraOne), and immersive environments such as driving simulators. Some of the learning tasks enabled through these technologies are highly structured and linear, whereas others are ill defined. Only by clearly
understanding the demands of these tasks can technologies be efficiently built to enable them for users of all ages.
Task demands can rarely be separated completely from the environments in which they are carried out, and this is certainly true in the area of technology for learning among older adults. Many of the technologies discussed in the paper by Willis have substantial divided attention, working memory, and task-switching requirements that can be reduced through intelligent design. Acknowledging the need for matching technology to the older learner, there have been several sets of guidelines for web-site development, in-vehicle telematics systems, and automated phone response systems. This research needs to be continued so that the older person is not left muddling through a learning tool that is filled with bells and whistles or heavy with content but cannot be used.
Learning environments can be optimized to task demands only when the understanding of user abilities is sufficient. Although there have been some attempts to find the predictors of individual differences in searching web sites and databases, these are small in number and do not extend to more complex learning tasks and applications. Researchers do not know if age-related changes in speed of processing, verbal and visual spatial memory, attention, and executive control account for individual differences in the performance of these tasks. Developing that knowledge base will help both designers and users.
There are two particularly interesting ways in which research could focus on individual differences that affect the older learner. The first concerns preferences for structure in learning. It may be that many older adults prefer highly structured tasks and environments, so that learner-centered and loosely organized materials are less effective. The second research area deals with mental models of technological entities. Engineers and software designers, as well as people who have grown up with technology, have highly detailed representations of such constructs as links and cookies, file management, logical structures, and so on. It is possible that imparting these models to older users will confuse them early in learning, but it may in the long run give them a better understanding of the technology that is being used to support their learning. If this makes learning less frustrating and more enjoyable, then everyone will benefit.
There is the persistent issue of cohort-based delay in exposure and access to technology. Certainly tomorrow′s elders will comfortably and frequently use today′s technologies, such as the web, automatic teller machines, and personal data assistants. They may not face the same barriers to access and use that challenge many older people today, although they will still be susceptible to age-related changes in movement control, perception, and cognition that impact interactions with technology. Moreover, technology will continue to develop, elders will continue to move
out of educational and work settings, and thus future generations of older learners will still have to cope with technology lag. How learning experts and technology developers can reduce the effects of this lag is a central challenge, the solution to which will benefit not only older adults, but poor, disabled, and geographically isolated people.
In this paper, Ann Horgas and Gregory Abowd describe technology for use in an older adult′s living environment. It includes a discussion of technology for such settings as assisted living and nursing homes, but, consistent with the fact that the vast majority of older adults live in their own homes, the focus is on technology for use in the home. A major goal of such technology is extending the period during which people can age in place, that is, remain safely in their homes. Toward this end, the technology helps compensate not only for motor and sensory deficits but also, importantly, for cognitive decline: there is a growing body of research that aims to develop cognitive aids to help ensure that older adults perform necessary routine daily activities. An additional key goal of the technology is to support the caregivers of older adults. In many cases, caregiving is provided by an aging spouse, while in many others, it is provided by adult children who do not reside with their parents. Cognitive aids can support this latter group by providing them with frequent, detailed information about their parents′ activities and status. This in turn supports aging in place, because adult children are more satisfied with such living arrangements when they can monitor their parents′ well-being.
The maturity of the technology described varies, from devices and systems that are nearly ready to be marketed to concepts and prototypes that are still in the design and development stages in the laboratory. Before becoming widely used, however, all the technology discussed must overcome four key challenges:
Cost. In most cases, the major expenses of the technologies described are not hardware-related, and when they are, these costs can be mitigated by economies of scale. Potentially more intractable expenses arise from the need to customize the technologies to individual users, a requirement that is closely related to the second challenge. Universal design principles do not apply in a straightforward manner to cognitive aids, since by definition these systems must represent the details of their users′ daily activities.
Ease of use. It is essential that the technologies be made completely transparent to use, and, if the user is to install them, they need to be
transparent to install as well. One way to facilitate this is by customizing a system to a user′s particular situation, but customization can be expensive. Advanced machine-learning techniques may help mitigate this problem by automating the customization process.
Reliability. Reliability is essential for technologies that are put into the home, where technical support staff are not immediately available. Although we are all familiar with the frustrating experience of dealing with unreliable software systems, we also deal daily with many instances of highly reliable technology. Cable television is one example. In addition, the design of “self-healing systems” is a very active current area of research, and there is similarly a large body of knowledge on the design of fault-tolerant systems. These and related engineering techniques will play a large role in the development of future technologies for the living environment.
Privacy. Many of the technologies described in this paper involve extensive monitoring of the routine daily activities of an older adult. Clearly, such monitoring raises important privacy concerns that will need to be addressed by technology designers to ensure that the rights and dignity of the users of the technology are respected.
Although these challenges are significant, it should not be forgotten that the technology that gives rise to them is being created to handle equally significant challenges that result from the dramatic demographic shifts currently under way. As the proportion of older adults skyrockets, there will not be enough younger adults to serve as full-time caregivers in the home, and technology will have to take up some of the slack. The technology discussed in this paper—if it is able to meet the four classes of challenges mentioned above—offers the hope of a better and safer quality of day-to-day life, in the home, for many older adults.
The paper by Joachim Meyer starts from the assumption that transportation is more than getting from point A to point B. Mobility is an integral part of everyday independent living. In the United States, transportation is defined as “automobility,” wherein more than 92 percent of the public choose the car as their primary mode of transport. The next wave of retirees, the baby boomers, have been defined by the automobile since their youth, and today as adults most live in the suburbs or rural areas where alternatives to the car are few or nonexistent. However, the natural aging process and age-related conditions may affect the capacity of older drivers to operate a vehicle safely and comfortably. Meyer discusses both the promise and the challenges associated with the introduction of new in-vehicle technologies to assist the older driver. Although the
major focus is on the automobile because of its importance in American life, other modes of transport are also important to many seniors, and some discussion is presented concerning public transportation.
WHAT CAN BE LEARNED FROM THE WORKSHOP
The vision for the workshop was the identification of a list of exciting new technologies to support the aging population that could be recommended for immediate transition to application in the range of settings studied by the workshop participants. No such list was forthcoming from the invited papers or the workshop discussion. On one hand, we heard about the promising new directions in miniaturization, electronic connectivity, and software sophistication that would enable all kinds of communications aids and sensory instrumentation to enhance social support from relatives and peers, monitoring and control of health care and medication administration in the homes where seniors prefer to live independently, and potentially to enhance their instrumental activities of daily living and quality of life. On the other hand, in each domain we heard about the serious challenges that the development of truly useful technological support would encounter, from trust, privacy, and safety to acculturation, autonomy, and dignity. At this stage in the field, there is simply not enough mutual understanding between specialists in aging research and technology developers to enable sensible, prioritized development of such a list. Instead, in the paragraphs that follow we suggest the kinds of further developments that are required and some of the means to accomplish those developments that the steering committee considers to have the greatest potential to bring about the most useful and usable technological support of the aging population.
Although we of course already knew this at some level, the workshop strongly reinforced the fact that producing technology that is genuinely supportive of the aging population is a many-faceted problem that requires multidisciplinary effort across the developmental spectrum, from basic research to evaluation of operating products in use in the field. The nation′s current approach, as it is with many problems, is to “let a thousand flowers bloom.” However, in this domain, such an approach has the effect of letting available technology seek its own market, and as discussed above, there are too many reasons, involving both user acceptance and economic reality, why such an approach will facilitate only the most widely applicable technological supports.
There are many ways government agencies can play a role in promot-
ing this kind of multidisciplinary collaboration. First, they could foster mutual education among industrial technologists and specialists in aging research in government and university laboratories. University faculty could be supported to spend summers or sabbatical leaves on site in industrial settings where potentially relevant technology is being developed. Graduate student training grants could be focused specifically on multidisciplinary training in laboratories conducting aging research, together with internships in industrial and field settings where potential technology interventions are being developed and evaluated. Specific work groups, workshops such as this one, and conferences could be sponsored to promote education across disciplines among established experts.
Second, government agencies could sponsor unconventional types of translational research and development that explicitly call for collaboration between technologists with ideas for product innovations and experienced specialists in research on aging. There is great interest among some of the major U.S. corporations in tapping into the growing market created by the aging population, but they lack the expertise needed to develop products that are truly responsive to the needs of the aging population. Program projects are an existing funding mechanism that is suitable for this purpose, but they would require that sufficient resources be committed to support both halves of the venture, the corporate developers and the specialists in gerontology. Thoughtful negotiation of the allocation of intellectual property rights is another prerequisite. Without this kind of support, it is unlikely that these sorts of collaborations will happen serendipitously.
Collaborative Ethnographic Studies
There is a need for research studies employing sophisticated, structured observation, taking the user′s perspective in settings in which the aging population is found, in order to understand the real needs, the impact of the physical and social environment, and the range of individual differences that must be accommodated. Recently a new breed of ethnographers has become involved in collecting this kind of information for the design of computer-supported applications across a wide variety of settings. But without an understanding of the potential of technology, such studies will lack a focus. What is needed is collaborative research in realistic settings with observation and input from technologists who can educate the observers and the users about potential technical innovation and at the same time learn what the needs are and begin to shape the available technology to be responsive. It is difficult for users to articulate what will be useful to them without some examples of what is practical and feasible. With multiple prototypes to examine, they can interpolate and extrapolate from examples to recommend valuable innovations. Fur-
thermore, this kind of human factors approach to a “requirements analysis” from the perspective of older users should become an integral part of the technology development process.
Formative Premarket Evaluation Studies
Another kind of study need spans the range from research to prototype application. Once a product idea has reached the prototype stage, there is a need for the equivalent of what in the software industry is called beta testing—that is, testing by the intended user population in real-life situations. This beta testing needs to be carried out by an interdisciplinary team that works in the aging user′s environment, and it needs to include specialists who understand the aging population and specialists in the conduct of formative evaluation. Such evaluation is not directed at assessment of the final outcome but rather attempts to discover the modifications to the technology that will enhance its effectiveness. Such studies will also reveal potential behavioral changes required of the users themselves to make the technology successful. Speakers at the workshop identified the need for more advanced evaluative methodologies to support such studies. It was mentioned repeatedly that the goal is a change on some outcome that is important to the life of an individual—in the activities of daily living or their overall “quality of life”—but clear definitions or evaluative criteria are not available to assess these features. The problem is complicated further by the fact that the introduction of technology will represent a change agent that will cause the aging population to adapt and change. The measurement regimens needed to take account of these potential dynamics have yet to be developed, and should be a priority for the National Institute on Aging.
Systems Implementation Studies
There is a third kind of study that is needed to support technology transition into widespread use. It is the study of actual “pilot” systems implementation in the broader setting with the opportunity to observe the impacts across the organizations and agencies that will be impacted. It was suggested that technology might be a primary means for integrating healthcare delivery in hospitals and clinics with primary healthcare in the home, or eventually for permitting significantly more care to be provided in the home and less in traditional healthcare environments. It has even been suggested that a change is needed in the overall healthcare delivery model to make affordable healthcare available to older adults. If system changes such as these are to be effected—and there are opportunities to do so across the domains in which the aging population potentially will
meet new technologies—there is a need for collaborative multidisciplinary “model implementation” studies among the organizations and agencies that will be impacted.
Technology transition can have impacts where they are least suspected. In a hospital in which computer-supported drug ordering was to be introduced, an interdisciplinary study was undertaken while the technology was being introduced to understand the broad potential positive and negative impacts of its introduction. The evaluation extended not only to the pharmacy and the doctors who would be using the system, but also to potential impacts throughout the hospital. Transitioning technology to the systems affecting the aging population will require the same kind of care in evaluation to ensure that the positive impacts outweigh the negative ones.
Improvements in Infrastructure
There is a need for substantial improvements in physical and electronic infrastructure before much of the potential of technological contributions can be achieved. The Americans with Disabilities Act and related legislation are making giant strides in improving physical accessibility in public spaces, but few new homes are being built with wheelchair access in mind and few existing private homes can accommodate even the most fundamental physical access needs without serious renovation.
Advanced communications technology was frequently mentioned in the workshop, not just by the speakers on that topic but also in the discussions of living environments, healthcare, and transportation. However, the application of communications technology presumes the availability of the infrastructure to support it. It was argued that broadband access can be assumed to be available in the home in the next few years, but at what cost, with what user interfaces, and with what technical support to the population that may already be having difficulty with the activities of everyday living?
It would also be useful to investigate ways to site or design environments suitable for older persons that encourage them to stay engaged in educational activities, stimulating leisure pursuits, or fitness activities. Some housing has been strategically placed near colleges or universities, where access to educational or intellectually challenging resources is easily available, but other approaches could also be considered.
Need for Training and Instruction
Although a goal of design may be to develop products that do not require training, this is probably an inappropriate goal. The best-designed product, especially if it is complex, is going to require some training. A better goal
is to ensure that the documentation and instructional support that is provided with the product itself is well designed. Older adults are generally more willing to use technologies if training is provided, but we need to better understand their specific training needs—how, when, and if training for older adults should differ from training for younger or middle-aged adults.
A recurrent theme of the workshop reflected the importance of providing training and instruction for the use of technology by older adults. Manuals and on-line help systems are notoriously ineffective and difficult to use—this aspect of technology development often receives short shrift in the design process. Moreover, system error messages are often vague or confusing and sometimes misleading.
Users must be taught how to interact with the technology and to understand sufficiently well how the system works to be able to tell when it is not working properly. Training and instruction must be developed following instructional design principles, and such education must be tailored to usage patterns; that is, if a system is used infrequently, the instructions must be easily accessible and interpretable. If the consequences of misuse are extremely critical, users must be trained to the level of perfect performance. Issues of retention over time and the potential need for refresher training must also be addressed.
Future developments in this area should also focus on the potential for the technology itself to provide user training, sometimes referred to as “embedded training.” The device could serve as a “coach” for the user by providing specific feedback and guidance when errors are made or introducing new features of the system only after the user becomes proficient with the basics.
Human Factors Tools and Techniques
The discipline of human factors is a multidisciplinary approach to design that puts the user at the center of the design process with the goal of developing safe, effective, and efficient user-system interactions. Within this discipline are tools and techniques that can be used to increase the likelihood that technologies will be designed to meet the needs of the users and that users will be able to safely and effectively use the technologies because they are designed well and accompanied by help systems, manuals, and instructions that are efficacious. Such tools and techniques include needs analysis to fully understand what the system should do and how it should do it; person analysis to recognize the capabilities and limitations of the target user population; task analysis to detail the components of tasks and ensure that the system functionality is appropriate, the expectations of the user are considered, and that error messages are clear at each point in the process; and in-
structional analysis to determine the form, content, and medium that will provide effective instructions for initial learning, retention over time, troubleshooting, and system maintenance.
The techniques of iterative, user-centered design and user testing provide methods of ensuring that systems will be usable. There are documented components to usability that provide guidance for design: learnability, efficiency, memorability, error avoidance, and satisfaction. Techniques such as rapid prototyping to develop mock representations of a product can be used to identify critical flaws early in the design process. Similarly, “wizard-of-oz” methods may be used to mimic what the system will do to measure user behaviors and expectation, prior to the expense of building the final product. Practitioners with these skills should be involved throughout the technology development process, whether it is the development of a single web site or the development of the smart home of the future.
These are exciting times for researchers and practitioners who are interested in making a difference in the lives of aging citizens. As this workshop has attested, technological opportunities abound, and it is likely that, whether the multidisciplinary communities that have a stake in making them successful participate or not, many of these technologies will reach the marketplace and some of them will be imposed on the older population in the interests of efficiency and cost-effectiveness. It is up to the readers of this workshop report to make sure that these technologies are developed in ways that will ensure their success and to ensure that future cohorts of the population age with greater dignity, feelings of self-worth, accomplishment, and happiness.