For most of her life, Rose Franke was a voracious reader. A retired high school teacher, she had, over a lifetime, acquired an extensive library stocked with the classics, from Chaucer to the Brothers Grimm. Then, like thousands of Canadians in their 60s, she developed glaucoma, which caused her vision to fade. Now in her late 70s, she is learning braille and has tried using podcasts and audio books. She even managed to track down a braille copy of Virgil in Latin. But not all books are available in these formats, and they can be difficult to use – especially if you’re someone who likes to study the text closely.
A volunteer recorded himself on CDs reading books, and Franke was able to use them because her CD player had large, raised buttons. However, her grandson Xavier Snelgrove wondered if there might be a simpler way to recreate the warmth of one-to-one reading. At the time, Snelgrove, then a fourth-year student in engineering science, was working at the Technologies for Aging Gracefully lab (TAGlab) – a group within U of T’s department of computer science that aims to help the old, the infirm and the impaired lead better, more social, more autonomous lives. The lab was founded in 2009 by Ronald Baecker, a computer science professor with extensive experience in computer animation, computer graphics and, most important, human-computer interaction.
At around the same time Snelgrove was thinking about how to assist his grandmother with reading, Baecker was having his own social epiphany. Attending a family party where an iPad was the gift of honour, he watched in amazement as a group of people whom he hadn’t thought were particularly interested in technology took up the tablet with glee. So great was their enthusiasm that Baecker wondered if his lab could find a way to use the iPad to help people read.
Back at the lab – actually a suite of offices on the seventh floor of the Bahen Centre for Information Technology – Baecker and Snelgrove discussed options. What they envisaged eventually turned into an iPad application that allows readers to record their own audio versions of downloaded books. The application, which they call the Accessible, Large-print, Listening and Talking ebook, recreates the experience of reading aloud with (or to) a friend or loved one. While a regular iPad can “read” books aloud using a standard computerized voice, the TAGlab’s ebook allows users to substitute the computer voice with the voice of someone they know. Friends or family members record their version of the text at the same time as they’re reading aloud. The computer picks up the thread if and where they leave off, though not, as yet, in the original human voice – the technology is not quite there yet. Text and audio are synchronized so you get the best of both regular and audio books, and previously read text is marked so it’s easy to find your place. Many of the app’s features aren’t yet finalized, but it’s hoped that even people with visual or motor impediments will be able to control the playback using voice, touch or a highly simplified keyboard.
While still in a prototype stage, the new ebook is a good example of what TAGlab does. It uses technology as a support, not a replacement, says Mike Massimi, a PhD student and assistant director of the lab. “We’re taking something that people already do, which is read out loud to each other, and we’re just inserting the technology there as a support to help the social experience of reading. . . . You could very well have a text-to-speech machine read every single book to you and some people would say it’s the same outcome, but it’s not. There’s a layer of humanity that we try to address.”
In the future, computers may be so good at mimicking the tone and rhythm of speech that their synthesized sentences will be indistinguishable from a human’s. But at the moment, says Liam Kaufman (MSc Medical Science 2008, BSc Computer Science 2011), a research associate who is working on the ebook, human speech has the edge. “Computer-synthesized text [or text-to-speech] . . . rarely conveys emotions or personality. Recordings capture both the story and the reader’s emotions and personality.”
As time goes on, TAGlab’s research is almost certainly going to become more important. The oldest baby boomers are now turning 65 and seniors are one of the fastest-growing age groups in Canada. According to the 2006 census, more than one million Canadians are now aged 80 and up. By 2050, a third of the population in high-income countries is expected to be 60 or older and, for the first time in the history of the world, the old (defined as 60 and up) will outnumber the young (those under 15). With an aging population comes a change in values. Things that were once taken for granted – mobility, autonomy, the ability to cook, clean and do small things for oneself – become both fraught and perplexing.
TAGlab is determined to make aging a little easier. Staffed by everyone from undergrads to post-docs, the 30-person lab looks for ways to help older people, or those with physical or mental challenges, to live better lives through technology. To date, the lab has achieved its biggest success with MyVoice, a smartphone app that helps people with aphasia or other speech impairments to communicate. The app is location-based, meaning it uses GPS technology to supply the user with a context-sensitive list of words and phrases – coffee-related words when the user is at Tim Hortons, for example, or “What’s playing?” for a trip to the movie theatre. (See “Say the Right Thing,” Autumn 2011.)
Baecker says TAGlab is interested in almost anything that empowers people in gentle, non-intrusive ways. While other researchers are developing technologies that monitor every aspect of an older person’s existence – sensors that watch for falls, for example – TAGlab’s approach is to support people in everyday tasks, with the goal of preserving their dignity and autonomy. Take the issue of faltering memory. Because so much of our sense of self depends on remembering who we are and what we’ve done, much of the lab’s work has investigated ways to bolster memory. In one experiment, older adults with dementia or mild cognitive impairment were sent on outings equipped with lightweight cameras that automatically took pictures every minute or so. They carried the cameras around their necks and afterwards they were shown pictures of their experience. “One of the things we found,” says Kaufman, who worked on the project, “is that it would increase their memory of a given event.” Simply wearing the camera to a museum, park or a family gathering helped people remember the specific event better. And, of course, remembering what’s happened around you makes it easier to connect with other people.
One of the challenges of this kind of work, says Massimi, “is that you try not to think like a computer person, even though you are. If you ask most computer people to solve a problem, they’ll solve it in a particular way that’s very efficient, very quick. . . . So if you ask them to create a system that reads books to blind people, you’re going to get basically a tape recorder that plays back a file.” But TAGlab researchers come from a variety of backgrounds and have more than just coding on their resumés – they’ve taken courses on everything from psychology and medical science to graphic design. (“We look for people who haven’t just studied math and computer science,” says Baecker.) And they think differently.
In the case of the ebook, says Massimi, they realized that it was “not just about transmitting information from a file to your ear as fast as possible. It’s about the experience of reading. It’s about being able to connect to other people around reading. It’s about . . . having the freedom to choose what to read, how you read it and when.”
As exciting as all these developments are, no amount of technology is going to help an elderly person who doesn’t have an adequate support system – or who is resistant to technology. But that, too, is part of the lab’s raison d’être. The team works diligently to get feedback about their projects. In a process called iterative design, they go back and forth between concept and reality, testing their ideas against the actual needs of their target users. They tested an early prototype of the ebook with Snelgrove’s grandmother and they’re now trying it out with a person who has multiple sclerosis. “We don’t just create a design that we want to use,” says Kaufman. “We will create a design and see how the user likes it and then make changes based on that.”
The best example of the lab’s user-friendly philosophy may be the communicating picture frame – a computer so simple it doesn’t look like a computer. Discreetly disguised as a bedside picture frame, it’s actually a touch screen (with a wireless connection) designed to help people in chronic pain communicate with a minimum of effort. One tap on its picture and the computer sends a simple message to a friend or loved one – either “I’m missing you” or “I’m not feeling well today.” With a good support network, the user will get a cheery video message back. Friends and family can either create a new video message or use a previously recorded one. A notification pops up when the new message arrives, the user taps the screen and the video starts playing. In the future, the user may also be able to replay old video messages. The technology itself is more complicated than it sounds, but to the person sending the message, it’s seamless. “One touch by the person with chronic pain and they get a great deal of information back from the family,” says Massimi.
When their own expertise in a field runs dry, the TAGlab team often consults with experts in other areas, especially the social sciences. Over the years they’ve worked with people in nursing, neuropsychology, social work and neurology. In the case of the communicating picture frame – and other strategies for alleviating loneliness in chronic-care settings – the team has collaborated with Véronique Boscart, a nurse-researcher at Toronto Rehab who taught in the Lawrence S. Bloomberg Faculty of Nursing for six years. Boscart works with people in chronic care and she’s acutely aware that the best solution does not always require new technology. Some of the people she works with would be happy with a laptop and an Internet connection. But Boscart likes TAGlab’s bottom-up approach and is happy to provide feedback on their ideas based on her own extensive clinical experience. “There are very few people who focus on solving clinical situations,” she says. “Quite often research happens in an ivory tower and by the time that knowledge is transmitted to the clinical setting a lot of it is lost. I think what TAGlab does is create real solutions for real people and real problems. That’s the beauty. Ron Baecker really thinks outside the box.”
It’s not easy working with senior citizens, says Baecker. Members of the TAGlab team often become close to the participants, and a couple of the participants have died during the course of the team’s research.
Baecker himself doesn’t seem too daunted by age. He’s 69 and will become a professor emeritus in 2013 but he hopes to keep working until age 99. Asking for any longer – expecting to live into the three digits – is “presumptuous,” he says. He traces back the inspiration for his current research to an academic paper on electronic prostheses he read about a decade ago. Nearing 60 at the time, and dealing with an ill sister, Baecker recalls thinking that prosthesis technology hadn’t come very far since the paper was originally published – in 1990. He and some U of T colleagues decided to try conceptualizing a few possibilities of their own. They came up with ideas for electronic memory aids and presented them at the first international conference on technology and aging, held in Toronto in September 2001. They went on to collaborate with health professionals at Baycrest, building technologies that could improve the lives of people with Alzheimer’s and amnesia. “I started thinking this is really what I should do for the rest of my research life,” says Baecker.
With about a dozen projects on the go, the TAGlab seems well on its way to realizing its goal of empowering older adults. By keeping its technologies deceptively simple and easy-to-use, it aids the old and infirm without ever impinging on their autonomy. Snelgrove’s grandmother will have to wait a little longer for her own personal copy of the Accessible, Largeprint, Listening and Talking ebook, though. The lab likes to make sure its products are glitch-free before they’re released, and the app is still sprouting features. But it will come to market, sooner or later. Baecker has a strong track record for developing viable products (prior to TAGlab he was involved with four startups, one of which, the webcasting firm Captual Technologies, was recently sold to Desire2Learn) – and it’s part of the lab’s ethos.
The goal, says Massimi, is to discover people’s needs, develop the technology to address these needs and then actually get the technology into people’s hands. “That’s what we like to do. We like to take stuff out of the lab and get it working.”
By bringing artificial intelligence into chemistry, Prof. Aspuru-Guzik aims to vastly shrink the time it takes to develop new drugs – and almost everything else