I knew we had been successful when 80-year-old Grace Curran told us that her daily life had gotten easier and her self-confidence improved. "My memory is lots better," she says, "and I feel a little bit prouder of myself in different things." What made the difference was the Brain Fitness Program, a neuroscience-based training program she completed on a computer, even though in her own words she "was definitely computer illiterate." "I thought the program was easy," she says. "I didn't have any difficulty with it at all."
While the computer platform offered invaluable benefits for the Brain Fitness Program, it also posed challenges for our user interface (UI) designers because of our target audience: people aged 60-plus. What those benefits are, and how we resolved the UI issues, are the subjects of this article.
As a scientist, my expertise is in neuroplasticity, a subfield of neuroscience. The brain's natural plasticity is what allows us all to learn new skills, from sitting up in babyhood to performing a masterful violin concerto in maturity. In skill acquisition, the brain progressively "specializes" by modifying the effective wiring that supports more-successful practice attempts. That progressive connectional specialization is the fundamental basis of learning.
Although scientists once thought that plasticity was limited to early childhood, by the mid-1990s scientific research had shown that the brain retains its plasticity throughout life. Not only does the physical brain continue to specialize its information-processing machinery in mature life (old dogs can learn new tricks), but it is also possible to steer brain plasticity to progressively train the brain out of earlier maladaptive learning that can have serioussometimes devastatingconsequences for the person in the form of dyslexia, schizophrenia, depression, and so on.
Research has shown that harnessing plasticity to retrain the brain requires heavy, repeated exposure to specific, progressively changing, corrective stimuli. Dyslexia can serve as an example. Dyslexic children develop an abnormal neurological representation of the sounds in aural language that bear meaning in words. To correct the problem requires a radical revision in the wiring that represents these critical sounds. These large-scale changes can be achieved only through elaborate, controlled, fine-grained learning progressions. To etch the new, corrected pathways for each crucial sound requires thousands of repetitions in which the child with dyslexia has to make decisions about which sound he or she heard. In due course, the brain gradually improveseven renormalizesits representation of those critical word sounds.
The notion that we can drive corrections in the brain's physical connections and its information processing by intensive, computer-guided training means the beginning of a therapeutic revolution. Patients with mild cognitive impairment, early-stage Alzheimer's disease, stroke, and traumatic brain injury are all identified targets of this class of therapeutics. These treatments have the great advantage of being noninvasive, relatively simple in conception and production, and completely natural ("organic") in their neurological impacts.
In 2002 my colleagues and I founded Posit Science to provide effective programs for better cognitive function in older adults. Posit Science's first offering, the Brain Fitness Program, improves auditory processing, which in turn improves memory, focus, and thinking.
The computer's value in these training regimens cannot be overstated. An effective brain-training program must meet many requirements, most of which are far better delivered via computer than through other media or by direct human-delivered therapies, as you can see in the list below.
1. Brain-training programs must deliver incredibly precise, scientifically exact stimuli.
The stimuli in the Brain Fitness Program include "frequency sweeps," the most elemental building blocks of language; idealized, synthetic versions of phonemes; and syllables, words, and sentences that have been specially processed to emphasize the pieces that are important for your brain to capture. The purpose of these highly stylized stimuli is to gradually etch clear representations of each sound in the brain. Until 32-bit sound cards became commonplace, these stimuli were difficult to deliver to a broad audience.
2. Training programs must dynamically adapt to individual performance.
Effective brain training must train the brain at "threshold," the uppermost edge of ability, so that the brain can make gradual improvements. Keeping the training at threshold requires constant adaptation to the individual's performance. A software program built with algorithms that efficiently govern this adaptability is much more effectively dynamic than a human trainer or another delivery mechanism.
3. Changing the brain requires that the individual makes thousands of controlled decisions.
Any new, renewed, or altered skill in the brain requires thousands of controlled decisions. Through quick, constant delivery of precise stimuli, computers can force the brain to make thousands of controlled decisions in a single hour. Compare this with the relatively few number of decisions your brain might have to make in a classroom setting.
4. Brain-training programs must give immediate, recognizable, and consistent rewards.
A brain that feels rewarded (or that anticipates a reward) increases its flow of dopamine, a neurotransmitter that is a gating factor in learning and memory. Computers can deliver consistent rewards directly after a correct answer, increasing the dopamine flow and therefore the likelihood of memory creation and retention.
5. For a brain-training program to have real meaning, it has to be widely available to those who need it.
One of the greatest advantages of the computer format is accessibility. It means that ifwhenwe figure out how to develop a training program that prevents Alzheimer's disease or a similarly devastating indication, we can get it out at a relatively low cost and relatively quickly to the many millions of people who need it.
Although the computer platform was a given for the reasons described above, several UI questions arose when we turned to developing the Brain Fitness Program. Our target audience was adults aged 60-plus, many of whom were into their 80s and beyond. This audience is diverse: Some have severe cognitive decline, others are extremely sharp; some are afraid of computers, others are computer engineers; some have vision, hearing and/or motor control issues that pose a challenge for computer usage; others are very adept in these areas. Based on our combined expertise in the science of aging and computer program design, we made the following decisions to accommodate this diverse audience.
Mouse use. To reduce nervousness, the Brain Fitness Program requires only mouse usage. For the first several days of the program, users just have to point and click; limited dragging is required later. 80-year-old Grace reported that this was one of the things that encouraged her to do the program. "We didn't have to use the keyboard or anything like that. I guess people that are experienced with computers could use a keyboard...I was at ease with just using the mouse."
Button size. The buttons in the Brain Fitness Program are very large, especially those that are part of the training tasks. There are two main reasons for this: (1) it helps people see/notice the buttons, even if they are unfamiliar with computer programs (what "looks" clickable) and/or have poor vision, and (2) it enables people with poor motor control or hand tremors to keep the cursor on the clickable area long enough to click it successfully. In user testing with about 20 subjects, we found that these buttons are easy to use and intuitive, though in response to feedback we ultimately increased the active areas on the items that require dragging beyond the borders of the item. (This is not unlike what many games and websites for young children do.)
Full screen. We selected an 800 x 600 format for the program for two reasons: (1) it makes everything larger, and therefore easier to see, and (2) it prevents distractions with background objects or competing programs, helping to focus attention on the program.
Contrast. As people grow older, the lenses in their eyes gradually yellow. The net result is that certain colors are more difficult to distinguish, such as blues and greens. As a result, the program uses high-contrast art, text, and buttons (though we tempered them somewhat to prevent eye fatigue from a constant barrage of brightness).
Volume and headphones. The Brain Fitness Program hones auditory processing, which relies to some degree on hearingyet many older adults have some trouble hearing and/or wear hearing aids. To accommodate people with hearing loss, the program includes headphones that provide the optimal listening environment and can be worn comfortably even with hearing aids. We also put multiple volume controls in the program, so that users could adjust the important sound stimuli to one level and the surrounding noises (such as correct-answer "dings") to a different level. (One issue we haven't yet resolved: Several women have complained that they do not like to use the Brain Fitness Program after they get their hair done, because the headphones flatten their hair. This is a problem we hadn't anticipated!)
Text. Onscreen text for older adults should be large; the NIA recommends at least 12 point. (The Brain Fitness Program uses 14 as a minimum.) Large text helps viewers with vision deficitsa substantial percentage of the 65-plus population. In addition, very limited use should be made of highly stylized fonts, italics, reversed type, all caps, and anything else that complicates or clutters the screen. Text should be left-justified . Text should be as short, clear, and, where possible, consistent (such as in button names). Many of these rules also apply to printed materials, which we always print in 14 pointmuch larger than in most publications of any kind. The major difference between onscreen and printed text for the older audience: Onscreen text should be sans serif, whereas printed text should be in a serif font.
Exercise initiations. The Brain Fitness Program takes users through a step-by-step exercise initiation. Voiceover instructions redirect them if they make a mistake. The initiation prevents the user from entering the main task of the exercise until he or she demonstrates a clear understanding of task performance. If the user does not demonstrate mastery of the task, he or she is directed to call Posit Science for help.
Optional information. While we kept all required information very direct and simple, the program includes a host of optional features for those who are interested. These include progress meters, points systems, ample scientific information, and so on. The key is to make every feature that isn't essential entirely optional (or at least ignorable), so that the most impaired user can still perform the exercises. (Our next program, to be released in late 2007, takes this strategy much further. It includes many more game-like elements that will make for a more engaging experience for those who are interested, but that can be easily ignored by those who find them distracting or overwhelming.)
Coach kit. We offer a coach kit for our most impaired users. The coach kit enables one person (often an adult child) to support another through the program. It includes a headphone splitter, an extra pair of headphones, a best-practices guide for the coach, and a single-button mouse. These tools are helpful when aiding a more impaired or nervous user.
Printed materials and customer support. Many older adults are more comfortable getting information through printed materials or a live person than a computer screen. Our experience suggests that any program designed for the 60-plus audience should have clear printed collateral and a strong customer-support component to accommodate those users.
We ran an ongoing User Testing Lab to verify that our UI decisions enabled our users to complete the program and made several adjustments based on their feedback. The design has proven effective: In a recent survey of almost 1,000 program users living in retirement communities, 85 percent were able to complete 30 hours of training. "I'm not a computer person," says Robert, age 81. "But it was easy to learn, nothing to it. Usually I only use the computer to write letters and play solitaire. Anything else and I break it." 102-year-old Doris has vision difficulties that make reading nearly impossible. "But I could do the Brain Fitness Program," she says. Sixty-year-old Diane reports: "I have never used or liked computers. I have no natural instinct for what to do. But the Brain Fitness Program wasn't hard to learn. Using the program actually got me comfortable enough with computers to finally learn how to use email."
In fact, studies at the University of California at San Francisco and Stanford University show that even people diagnosed with MCI and early-stage Alzheimer's disease can complete the program, despite their significant memory and learning deficits. This is important news: we believe that a version of the Brain Fitness Program may help to prevent or halt the progression of MCI and Alzheimer's. (Already, preliminary study results indicate that the program holds real promise for people with MCI.)
For me personally, solving the usability issues for the older adult audience matters because a usable interface enables the delivery of effective brain health programs that can make real change in people's lives. Fred Ford sums it up nicely. "Ever since I started using the Brain Fitness Program it has improved every aspect of my life. Not only has it maintained my brain functions, it has improved my physical and emotional health as well." As he tells us, even his bowling average has gone up!
Michael Merzenich, Ph.D.
About the Author
Michael Merzenich, Ph.D., is a professor of neurophysiology at the University of California at San Francisco and chief scientific officer at Posit Science Corporation.
Figure 3. To accommodate different types of users, the Brain Fitness Program makes nonessential information optional. On this screen, the bar at top gives the interested user information on where he or she is in the stimuli and how changes in stimuli are determined. The computer novice or more impaired user can easily ignore the bar.
©2007 ACM 1072-5220/07/0700 $5.00
Permission to make digital or hard copies of all or part of this work for personal or classroom use is granted without fee provided that copies are not made or distributed for profit or commercial advantage and that copies bear this notice and the full citation on the first page. To copy otherwise, to republish, to post on servers or to redistribute to lists, requires prior specific permission and/or a fee.
The Digital Library is published by the Association for Computing Machinery. Copyright © 2007 ACM, Inc.